CN106921533B

CN106921533B - Method, device and system for measuring network performance

Info

Publication number: CN106921533B
Application number: CN201510994478.5A
Authority: CN
Inventors: 林程勇; 董雯霞; 肖少然; 李勇
Original assignee: Tsinghua University; Huawei Technologies Co Ltd
Current assignee: Tsinghua University; Huawei Technologies Co Ltd
Priority date: 2015-12-25
Filing date: 2015-12-25
Publication date: 2020-02-14
Anticipated expiration: 2035-12-25
Also published as: CN106921533A

Abstract

The embodiment of the invention discloses a method, a device and a system for measuring network performance. The system comprises: the control node is used for receiving the measurement request, sending a first measurement instruction and a second measurement instruction, receiving a first result and forwarding the first result; the source detection node is used for receiving a first measurement instruction, sending a detection request message to a target detection node based on the first measurement instruction, receiving a detection response message sent by a target detection node, generating a first result according to the detection request message and the detection response message and sending the first result to the control node; the target detection node is used for receiving the second measurement instruction, receiving a detection request message sent by the source detection node and sending a detection response message to the source detection node based on the second measurement instruction; the source detection node and the source virtual client in the source physical server are communicated with the outside through the source virtual switch, and the target detection node and the target virtual client in the target physical server are communicated with the outside through the source virtual switch.

Description

Method, device and system for measuring network performance

Technical Field

The present invention relates to the field of communications technologies, and in particular, to a method, an apparatus, and a system for measuring network performance.

Background

For network services, both the service provider and the service consumer typically need to know about the network performance. For service providers, network performance may be used to measure the quality of service of a network service. For the service user, the network performance can be used to know the network situation. Generally, a measurement person of network performance can know the network performance between two network terminals through a ping operation. Assuming that a measurer needs to know the network performance between the network terminal a and the network terminal B, the measurer may manually write a ping command for the network address of the network terminal B on the network terminal a and manually trigger the written ping command, and the network terminal a may obtain a measurement result of the network performance between the network terminal a and the network terminal B, so that the measurer may know the network performance between the network terminal a and the network terminal B through the measurement result obtained by the network terminal a. It can be seen that in order to measure network performance between network terminals, a measuring person needs to manually write and trigger a ping command on the measured network terminal. It can be seen that the operation of measuring the network performance is cumbersome and complicated for the measuring staff.

Disclosure of Invention

The technical problem to be solved by the embodiments of the present invention is to provide a method, an apparatus, and a system for measuring network performance, so as to simplify manual operations of measuring personnel in a network performance measurement process, thereby enabling the measuring personnel to more simply and conveniently measure the network performance.

In a first aspect, an embodiment of the present invention provides a system for measuring network performance. The system comprises:

the control node is used for receiving a measurement request, sending a first measurement instruction to a source detection node, sending a second measurement instruction to a destination detection node, receiving a first result sent by the source detection node, and forwarding the first result, wherein the measurement request is used for indicating the control node to feed back a result for identifying the performance of a link from a source virtual client to a destination virtual client;

the source probe node is configured to receive the first measurement instruction sent by the control node, send a probe request packet to the destination probe node based on the first measurement instruction, receive a probe response packet sent by the destination probe node, generate the first result according to the probe request packet and the probe response packet, and send the first result to the control node, where the first result is used to identify performance of a link from the source probe node to the destination probe node;

the destination probe node is configured to receive the second measurement instruction sent by the control node, receive the probe request packet sent by the source probe node, and send the probe response packet corresponding to the probe request packet to the source probe node based on the second measurement instruction;

wherein the source probe node, the source virtual client and the source virtual switch are located in a source physical server, the destination probe node, the destination virtual client and the destination virtual switch are located in a destination physical server, the source physical server and the destination physical server are two physical servers, the source probe node is configured to communicate with the destination probe node through the source virtual switch, the destination probe node is configured to communicate with the source probe node through the destination virtual switch, the source virtual client is configured to communicate with the destination virtual client through the source virtual switch, and the destination virtual client is configured to communicate with the source virtual client through the destination virtual switch.

Optionally, the measurement request, the first measurement instruction, and the second measurement instruction all carry an identifier of the source virtual client, an identifier of the destination virtual client, and a type of an indicator to which a result of identifying performance of a link from the source virtual client to the destination virtual client belongs.

Optionally, the sending a probe request packet to the destination probe node based on the first measurement instruction includes:

determining a sending mode of the detection request message according to the type in the first measurement instruction;

and sending the detection request message to the target detection node according to the sending mode.

Alternatively to this, the first and second parts may,

the source probe node is a process in the source virtual client and the destination probe node is a process in the destination virtual client.

Alternatively to this, the first and second parts may,

the source probe node is a process in a first virtual machine, the first virtual machine and the source virtual client being two virtual machines located at the source physical server;

the destination probe node is a process in a second virtual machine, and the second virtual machine and the destination virtual machine are two virtual machines located in the destination physical server.

Alternatively to this, the first and second parts may,

the source detection node is further configured to determine whether the first result belongs to a first range, and generate and send a first alarm to the control node when the first result does not belong to the first range;

the control node is further configured to receive the first alarm sent by the source detection node, and forward the first alarm.

Alternatively to this, the first and second parts may,

the destination probe node is further configured to generate a second result according to the probe request packet, and send the second result to the control node, where the second result is used to identify performance of a link from the source probe node to the destination probe node;

the control node is further configured to receive the second result sent by the destination probe node, and forward the second result.

Alternatively to this, the first and second parts may,

the destination detection node is further configured to determine whether the second result belongs to a second range, and generate and send a second alarm to the control node when the second result does not belong to the second range;

the control node is further configured to receive the second alarm sent by the destination probe node, and forward the second alarm.

Optionally, the system further includes:

a source proxy node, configured to receive the first measurement instruction sent by the control node and forward the first measurement instruction to the source probe node, and receive the first result sent by the source probe node and forward the first result to the control node;

and the destination proxy node is used for receiving the second measurement instruction sent by the control node and forwarding the second measurement instruction to the destination detection node.

Alternatively to this, the first and second parts may,

the destination proxy node is further configured to receive the second result sent by the destination probe node and forward the second result to the control node.

Alternatively to this, the first and second parts may,

the source proxy node is further configured to receive the first alarm sent by the source detection node and forward the first alarm to the control node;

the destination proxy node is further configured to receive the second alarm sent by the destination probe node and forward the second alarm to the control node.

Alternatively to this, the first and second parts may,

the source proxy node is further configured to sign the first measurement instruction before forwarding the first measurement instruction;

the source detection node is further configured to verify the signed first measurement instruction after receiving the first measurement instruction, and send the detection request packet to the destination detection node when the verification is successful;

the destination proxy node is further configured to sign the second measurement instruction before forwarding the second measurement instruction;

the destination probe node is further configured to verify the signed second measurement instruction after receiving the second measurement instruction, and send the probe response packet corresponding to the probe request packet to the source probe node based on the second measurement instruction if the verification is successful.

Optionally, the source proxy node and the destination proxy node are the same proxy node;

the control node is further configured to, before sending the first measurement instruction, obtain a first route from the source proxy node to the source probe node and a second route from the source probe node to the source proxy node, send the first route to the source proxy node, send the second route to the source probe node, and, before sending the second measurement instruction, obtain a third route from the destination proxy node to the destination probe node, and send the third route to the destination proxy node;

wherein the first route is used for the source proxy node to send the first measurement instruction to the source probe node, the second route is used for the source probe node to send the first result to the source proxy node, and the third route is used for the destination proxy node to send the second measurement instruction to the destination probe node.

Optionally, the source proxy node is a process in the source physical server, and the destination proxy node is a process in the destination physical server.

Alternatively to this, the first and second parts may,

the control node is further configured to send first control information to the source proxy node and second control information to the destination proxy node before sending the first measurement instruction and the second measurement instruction, designate a first port on the source virtual switch for the source probe node and a second port on the destination virtual switch for the destination probe node, and establish a virtual local area network including the first port and the second port; the source detection node can communicate with the destination detection node through the first port, the destination detection node can communicate with the source detection node through the second port, and the virtual local area network is used for transmitting the detection request message and the detection response message;

the source proxy node is further configured to receive the first control information, and create the first virtual machine and the source probe node in the source physical server based on the first control information;

the destination agent node is further configured to receive the second control information, and create the second virtual machine and the destination probe node in the destination physical server based on the second control information.

In a second aspect, an embodiment of the present invention provides a method for measuring network performance. The method comprises the following steps:

receiving a measurement request by a control node, wherein the measurement request is used for instructing the control node to feed back a result for identifying the performance of a link from a source virtual client to a destination virtual client;

the control node sends a first measurement instruction to a source detection node and sends a second measurement instruction to a destination detection node, wherein the first measurement instruction is used for instructing the source detection node to send a detection request message to the destination detection node, receiving a detection response message corresponding to the detection request message and generating a first result according to the detection request message and the detection response message, the first result is used for marking the performance of a link from the source detection node to the destination detection node, and the second measurement instruction is used for instructing the destination detection node to send the detection response message corresponding to the detection request message to the source detection node;

the control node receives the first result sent by the source detection node and forwards the first result;

Optionally, the measurement request, the first measurement instruction, and the second measurement instruction all carry an identifier of the source virtual client, an identifier of the destination virtual client, and a type of an indicator to which a result for identifying a link from the source virtual client to the destination virtual client belongs.

Optionally, the method further includes:

the control node receives the second result sent by the target detection node and forwards the second result;

the second result is generated by the destination probe node according to the probe request packet, and the second result is used for identifying the performance of the link from the source probe node to the destination probe node.

Optionally, the method further includes:

the control node receives a first alarm sent by the source detection node and forwards the first alarm;

the control node receives a second alarm sent by the target detection node and forwards the second alarm;

wherein the first alarm is generated and sent by the source probe node when the first result is determined not to belong to a first range, and the second alarm is generated and sent by the destination probe node when the second result is determined not to belong to a second range.

Optionally, the method further includes:

the control node determining whether the first result belongs to a third range;

when the first result does not belong to the third range, the control node generates a third alarm and forwards the third alarm;

the control node determining whether the second result falls within a fourth range;

and when the second result does not belong to the fourth range, the control node generates a fourth alarm and forwards the fourth alarm.

Alternatively to this, the first and second parts may,

the control node sending a first measurement instruction to the source probe node and a second measurement instruction to the destination probe node, including: the control node sends the first measurement instruction to a source proxy node to instruct the source proxy node to forward the first measurement instruction to the source probe node; the control node sends the second measurement instruction to a destination proxy node to instruct the destination proxy node to forward the second measurement instruction to the destination detection node;

the method for receiving the first result sent by the source detection node by the control node specifically includes: the control node receives the first result forwarded by the source proxy node, the first result being sent by the source probe node to the source proxy node.

Alternatively to this, the first and second parts may,

the receiving, by the control node, the second result sent by the destination probe node specifically includes: the control node receives the second result forwarded by the destination proxy node, the second result being sent to the destination proxy node by the destination probe node.

Alternatively to this, the first and second parts may,

the method for receiving the first alarm sent by the source detection node by the control node specifically includes: the control node receives the first alarm forwarded by the source proxy node, wherein the first alarm is sent to the source proxy node by the source detection node;

the receiving, by the control node, the second alarm sent by the target detection node specifically includes: the control node receives the second alarm forwarded by the destination agent node, wherein the second alarm is sent to the destination agent node by the destination detection node.

before the control node sends the first measurement instruction and the second measurement instruction, the method further comprises:

the control node acquires a first route from the source agent node to the source detection node and a second route from the source detection node to the source agent node, sends the first route to the source agent node, and sends the second route to the source detection node, wherein the first route is used for the source agent node to forward the first measurement instruction to the source detection node, and the second route is used for the source detection node to send the first result to the source agent node;

and the control node acquires a third route from the destination agent node to the destination detection node, and sends the third route to the destination agent node, wherein the third route is used for forwarding the second measurement instruction to the destination detection node by the destination agent node.

Optionally, the source proxy node is a process in a source physical server, and the destination proxy node is a process in a destination physical server;

the control node sends first control information to the source proxy node and second control information to the destination proxy node, the first control information is used for instructing the source proxy node to create a first virtual machine and the source probe node in the source physical server, the second control information is used for instructing the destination proxy node to create a second virtual machine and the destination probe node in the destination physical server, wherein the source probe node is one process in the first virtual machine, the first virtual machine and the source virtual client are two virtual machines located in the source physical server, the destination probe node is one process in the second virtual machine, and the second virtual machine and the destination virtual client are two virtual machines located in the destination physical server;

the control node appoints a first port on the source virtual switch for the source detection node and appoints a second port on the destination virtual switch for the destination detection node, the source detection node can communicate with the destination detection node through the first port, and the destination detection node can communicate with the source proxy node through the second port;

the control node establishes a virtual local area network comprising the first port and the second port; the virtual local area network is used for transmitting the detection request message and the detection response message.

In a third aspect, an embodiment of the present invention provides an apparatus for measuring network performance. The apparatus is deployed with a control node. The device comprises:

a first receiving unit, configured to receive a measurement request, where the measurement request is used to instruct the control node to feed back a result for identifying performance of a link from a source virtual client to a destination virtual client;

a first sending unit, configured to send a first measurement instruction to a source probe node and send a second measurement instruction to a destination probe node, where the first measurement instruction is used to instruct the source probe node to send a probe request packet to the destination probe node, receive a probe response packet corresponding to the probe request packet, and generate a first result according to the probe request packet and the probe response packet, the first result is used to identify performance of a link from the source probe node to the destination probe node, and the second measurement instruction is used to instruct the destination probe node to send the probe response packet corresponding to the probe request packet to the source probe node;

a second receiving unit, configured to receive the first result sent by the source probe node;

a first forwarding unit, configured to forward the first result;

Optionally, the apparatus further comprises:

a third receiving unit, configured to receive the second result sent by the destination probe node, where the second result is generated by the destination probe node according to the probe request packet, and the second result is used to identify performance of a link from the source probe node to the destination probe node;

and the second forwarding unit is used for forwarding the second result.

Optionally, the apparatus further comprises:

a fourth receiving unit, configured to receive a first alarm sent by the source probe node, where the first alarm is generated and sent by the source probe node when it is determined that the first result does not belong to a first range;

a third forwarding unit, configured to forward the first alarm;

a fifth receiving unit, configured to receive a second alarm sent by the destination probe node, where the second alarm is generated and sent by the destination probe node when it is determined that the second result does not belong to a second range;

and the fourth forwarding unit is used for forwarding the second alarm.

Optionally, the apparatus further comprises:

a first determination unit configured to determine whether the first result belongs to a third range;

a fifth forwarding unit, configured to generate and forward third network performance alarm information when the first result does not belong to the third range;

a second determination unit configured to determine whether the second result belongs to a fourth range;

and the sixth feedback unit is used for generating and forwarding fourth network performance warning information when the second result does not belong to the fourth range.

Alternatively to this, the first and second parts may,

the first sending unit is specifically configured to send the first measurement instruction to a source proxy node to instruct the source proxy node to forward the first measurement instruction to the source probe node, and send the second measurement instruction to a destination proxy node to instruct the destination proxy node to forward the second measurement instruction to the destination probe node;

the second receiving unit is specifically configured to receive the first result forwarded by the source proxy node, where the first result is sent to the source proxy node by the source probe node.

Alternatively to this, the first and second parts may,

the third receiving unit is specifically configured to receive the second result forwarded by the destination proxy node, where the second result is sent to the destination proxy node by the destination probe node.

Alternatively to this, the first and second parts may,

the fourth receiving unit is specifically configured to receive the first alarm forwarded by the source proxy node, where the first alarm is sent to the source proxy node by the source probe node;

the fifth receiving unit is specifically configured to receive the second alarm forwarded by the destination proxy node, where the second alarm is sent to the destination proxy node by the destination probe node.

the device further comprises:

a first obtaining unit, configured to obtain a first route from the source proxy node to the source probe node, and send the first route to the source proxy node, where the first route is used for the source proxy node to forward the first measurement instruction to the source probe node

A second obtaining unit, configured to obtain a second route from the source probe node to the source proxy node, and send the second route to the source probe node, where the second route is used for the source probe node to send the first result to the source proxy node;

a third obtaining unit, configured to obtain a second route from the destination proxy node to the destination probe node, and send the second route to the destination proxy node, where the second route is used for the destination proxy node to forward the second measurement instruction to the destination probe node.

the device further comprises:

a second sending unit, configured to send first control information to the source proxy node and send second control information to the destination proxy node, where the first control information is used to instruct the source proxy node to create a first virtual machine and the source probe node in the source physical server, and the second control information is used to instruct the destination proxy node to create a second virtual machine and the destination probe node in the destination physical server, where the source probe node is one process in the first virtual machine, the first virtual machine and the source virtual client are two virtual machines located in the source physical server, the destination probe node is one process in the second virtual machine, and the second virtual machine and the destination virtual client are two virtual machines located in the destination physical server;

a specifying unit, configured to specify a first port on the source virtual switch for the source probe node and a second port on the destination virtual switch for the destination probe node, where the source probe node is capable of communicating with the destination probe node through the first port, and the destination probe node is capable of communicating with the source proxy node through the second port;

the establishing unit is used for establishing a virtual local area network comprising the first port and the second port; the virtual local area network is used for transmitting the detection request message and the detection response message.

In a fourth aspect, an embodiment of the present invention provides a method for measuring network performance. The method comprises the following steps:

a source detection node receives a measurement instruction sent by a control node, wherein the measurement instruction is sent when the control node receives a measurement request, and the measurement request is used for indicating the control node to feed back a result for identifying the performance of a link from a source virtual client to a destination virtual client;

the source detection node sends a detection request message to a target detection node based on the measurement instruction;

the source detection node receives a detection response message sent by the destination detection node, and generates a result according to the detection request message and the detection response message, wherein the result is used for identifying the performance of a link from the source detection node to the destination detection node, and the detection response message is sent by the destination detection node corresponding to the detection request message;

the source detection node sends the result to the control node so that the control node can forward the result;

Optionally, the measurement request and the measurement instruction both carry an identifier of the source virtual client, an identifier of the destination virtual client, and a type of an indicator to which a result of identifying performance of a link from the source virtual client to the destination virtual client belongs.

Optionally, the sending, by the source probe node, a probe request packet to the destination probe node based on the measurement instruction specifically includes:

the source detection node determines the sending mode of the detection request message according to the type in the measurement instruction;

and the source detection node sends the detection request message to the target detection node according to the sending mode.

Alternatively to this, the first and second parts may,

the source probe node is a process in a first virtual machine, the first virtual machine and the source virtual client being two virtual machines located on the physical server;

the destination probe node is a process in a second virtual machine, the second virtual machine and the destination virtual client being two virtual machines located on the destination physical server.

Optionally, the method further includes:

the source probe node determining whether the result belongs to a range;

when the result does not belong to the range, the source detection node generates an alarm;

and the source detection node sends the alarm to the control node so that the control node can forward the alarm.

Alternatively to this, the first and second parts may,

the method for the source detection node to receive the measurement instruction sent by the control node includes: the source detection node receives the measurement instruction forwarded by a source proxy node, wherein the measurement instruction is sent to the source proxy node by the control node;

the source detection node sends the result to the control node, specifically: and the source detection node sends the result to the source proxy node so that the source proxy node sends the result to the control node.

Alternatively to this, the first and second parts may,

the source detection node sends the alarm to the control node, specifically: and the source detection node sends the alarm to the source proxy node so that the source proxy node forwards the alarm to the control node.

Optionally, the method further includes:

the source detection node verifies the measurement instruction and sends the detection request message to the target detection node under the condition that the verification is passed;

wherein the measurement instruction is forwarded to the source probe node after the source proxy node signs.

In a fifth aspect, the present invention provides an apparatus for measuring network performance. The apparatus deploys active probing nodes. The device comprises:

a first receiving unit, configured to receive a measurement instruction sent by a control node, where the measurement instruction is sent when the control node receives a measurement request, and the measurement request is used to instruct the control node to feed back a result for identifying performance of a link from a source virtual client to a destination virtual client;

a first sending unit, configured to send a probe request packet to a target probe node based on the measurement instruction;

a second receiving unit, configured to receive a probe response packet sent by the destination probe node;

a first generating unit, configured to generate a result according to the probe request packet and the probe response packet, where the result is used to identify performance of a link from the source probe node to a destination probe node, and the probe response packet is sent by the destination probe node in response to the probe request packet;

a second sending unit, configured to send the result to the control node, so that the control node forwards the result;

Optionally, the first sending unit is specifically configured to:

determining a sending mode of the detection request message according to the type in the measurement instruction;

Alternatively to this, the first and second parts may,

Optionally, the apparatus further comprises:

a determination unit for determining whether the result belongs to a range;

a second generating unit, configured to generate an alarm when the result does not belong to the range;

and the third sending unit is used for sending the alarm to the control node so that the control node can forward the alarm.

Alternatively to this, the first and second parts may,

the second receiving unit is specifically configured to receive the measurement instruction forwarded by the source proxy node, where the measurement instruction is sent to the source proxy node by the control node;

the second sending unit is specifically configured to send the result to the source proxy node, so that the source proxy node forwards the result to the control node.

Alternatively to this, the first and second parts may,

the third sending unit is specifically configured to send the alarm to the source proxy node, so that the source proxy node forwards the alarm to the control node.

Optionally, the apparatus further comprises:

a verification unit, configured to verify a signature in the measurement instruction, where the measurement instruction is forwarded to the source probe node after being signed by the source proxy node;

and the triggering unit is used for triggering the first sending unit under the condition that the verification is passed.

In a sixth aspect, an embodiment of the present invention provides a method for measuring network performance. The method comprises the following steps:

a target detection node receives a measurement instruction sent by a control node, wherein the measurement instruction is sent when the control node receives a measurement request, and the measurement request is used for indicating the control node to feed back a result for identifying the performance of a link from a source virtual client to a target virtual client;

the target detection node receives a detection request message sent by a source detection node;

the target detection node sends a detection response message corresponding to the detection request message to the source detection node based on the measurement instruction, the detection request message and the detection response message are used by the source detection node to generate a first result and sent to the control node, the first result is used for identifying the performance of a link between the source detection node and the target detection node, and the first result is used by the control node to forward;

Optionally, the measurement request and the measurement instruction both carry an identifier of the source virtual client, an identifier of the destination virtual client, and the measurement request is used to instruct the control node to feed back a type of an indicator to which a result for identifying performance of a link from the source virtual client to the destination virtual client belongs.

Alternatively to this, the first and second parts may,

Optionally, the method further includes:

the target detection node generates a second result according to the detection request message, wherein the second result is used for identifying the performance of a link from the source detection node to the target detection node;

and the target detection node sends the second result to the control node so that the control node can forward the second result.

Optionally, the method further includes:

the destination probe node determines whether the second result belongs to a range;

when the second result does not belong to the range, the target detection node generates an alarm;

and the target detection node sends the alarm to the control node so that the control node can forward the alarm.

Alternatively to this, the first and second parts may,

the target detection node receives the measurement instruction sent by the control node, and specifically includes: the destination detection node receives the measurement instruction forwarded by a destination agent node, and the measurement instruction is sent to the destination agent node by the control node.

Alternatively to this, the first and second parts may,

the destination probe node sends the second result to the control node, specifically: and the destination detection node sends the result to the destination agent node so that the destination agent node can forward the second result to the control node.

Alternatively to this, the first and second parts may,

the target detection node sends the alarm to the control node, specifically: and the destination detection node sends the alarm to the destination agent node so that the destination agent node forwards the alarm to the control node.

Optionally, the method further includes:

the target detection node verifies the measurement instruction and receives the detection request message sent by the source detection node under the condition of passing the verification;

wherein the measurement instruction is forwarded to the destination probe node after the destination proxy node signs.

In a seventh aspect, an embodiment of the present invention provides an apparatus for measuring network performance. The device deploys a destination probe node. The device comprises:

a second receiving unit, configured to receive a probe request packet sent by a source probe node;

a returning unit, configured to send, to the source probe node, a probe response packet corresponding to the probe request packet based on the measurement instruction, where the probe request packet and the probe response packet are used by the source probe node to generate a first result and send the first result to the control node, where the first result is used to identify performance of a link between the source probe node and the destination probe node, and the first result is used by the control node to forward the first result;

Alternatively to this, the first and second parts may,

Optionally, the apparatus further comprises:

a first generating unit, configured to generate a second result according to the probe request packet, where the second result is used to identify performance of a link from the source probe node to the destination probe node;

a first sending unit, configured to send the second result to the control node, so that the control node forwards the second result.

Optionally, the apparatus further comprises:

a determination unit configured to determine whether the second result belongs to a range;

the second generating unit is used for generating an alarm when the second result does not belong to the range;

and the second sending unit is used for sending the alarm to the control node so that the control node can forward the alarm.

Alternatively to this, the first and second parts may,

the first receiving unit is specifically configured to receive the measurement instruction forwarded by the destination proxy node, where the measurement instruction is sent to the destination proxy node by the control node.

Alternatively to this, the first and second parts may,

the first sending unit is specifically configured to send the second result to the destination proxy node, so that the destination proxy node forwards the second result to the control node.

Alternatively to this, the first and second parts may,

the second sending unit is specifically configured to send the alarm to the destination agent node, so that the destination agent node forwards the alarm to the control node.

Optionally, the apparatus further comprises:

the verification unit is used for verifying the measurement instruction, and the measurement instruction is forwarded to the target detection node after being signed by the target agent node;

and the triggering unit is used for triggering the second receiving unit under the condition that the verification is passed.

In an eighth aspect, an embodiment of the present invention provides a method for measuring network performance. The method comprises the following steps:

a source proxy node receives a measurement instruction sent by a control node, wherein the measurement instruction is sent when the control node receives a measurement request, and the measurement request is used for indicating the control node to feed back a result for identifying the performance of a link from a source virtual client to a destination virtual client;

the source proxy node forwards the measurement instruction to a source detection node to instruct the source detection node to send a detection request message to a destination detection node, receive a detection response message sent by the destination detection node and generate a result according to the detection request message and the detection response message, wherein the result is used for identifying the performance of a link from the source detection node to the destination detection node, and the detection response message is sent by the destination detection node corresponding to the detection request message;

the source proxy node receives the result sent by the source detection node;

the source proxy node forwards the result to the control node so that the control node forwards the result;

Optionally, the method further includes:

the source proxy node receives an alarm sent by the source detection node, wherein the alarm is generated by the source detection node when the result is determined not to belong to the range;

and the source proxy node forwards the alarm to the control node so that the control node can forward the alarm.

Optionally, before the source proxy node forwards the measurement instruction to the source probe node, the method further includes:

the source proxy node signs the measurement instruction; the signed measurement instruction is used for verification by the source detection node and sends the detection request message to the target detection node under the condition that the verification is passed.

Optionally, the source proxy node is a process in a source physical server;

before the source proxy node receives the measurement instruction, the method further comprises:

the source proxy node receives control information sent by the control node;

the source proxy node creates a virtual machine and the source probe node in the source physical server based on the control information, wherein the source probe node is a process in the first virtual machine, and the virtual machine and the source virtual client are two virtual machines located in the source physical server.

In a ninth aspect, an embodiment of the present invention provides an apparatus for measuring network performance. The apparatus configures an active proxy node. The device comprises:

a first forwarding unit, configured to forward the measurement instruction to the source probe node to instruct the source probe node to send a probe request packet to a destination probe node, receive a probe response packet sent by the destination probe node, and generate a result according to the probe request packet and the probe response packet, where the result is used to identify performance of a link from the source probe node to the destination probe node, and the probe response packet is sent by the destination probe node in response to the probe request packet;

a second receiving unit, configured to receive the result sent by the source probe node;

a second forwarding unit, configured to forward the result to the control node, so that the control node forwards the result;

Optionally, the apparatus further comprises:

a third receiving unit, configured to receive an alarm sent by the source probe node, where the alarm is generated by the source probe node when it is determined that the result does not belong to the range;

and the third forwarding unit is used for forwarding the alarm to the control node so that the control node can forward the alarm.

Optionally, the apparatus further comprises:

a signature unit, configured to sign the measurement instruction before sending the measurement instruction to the source probe node; the signed measurement instruction is used for verification by the source detection node and sends the detection request message to the target detection node under the condition that the verification is passed.

Optionally, the source proxy node is a process in a source physical server;

the device further comprises:

a fourth receiving unit, configured to receive the control information sent by the control node;

a creating unit, configured to create, in the source physical server, a virtual machine and the source probe node based on the control information before receiving the measurement instruction, where the source probe node is one process in the first virtual machine, and the virtual machine and the source virtual client are two virtual machines located in the source physical server.

In a tenth aspect, an embodiment of the present invention provides a method for measuring network performance. The method comprises the following steps:

a target agent node receives a measurement instruction sent by a control node, wherein the measurement instruction is sent when the control node receives a measurement request, and the measurement request is used for indicating the control node to feed back a result for identifying the performance of a link from a source virtual client to a target virtual client;

the destination proxy node forwards the measurement instruction to a destination detection node to instruct the destination detection node to receive a detection request message sent by a source detection node and send a detection response message corresponding to the detection request message to the source detection node, where the detection request message and the detection response message are used by the source detection node to generate a first result and send the first result to the control node, the first result is used for identifying the performance of a link from the source detection node to the destination detection node, and the first result is used by the control node to forward;

Optionally, the method further includes:

the destination agent node receives a second result sent by the destination detection node, wherein the second result is generated by the destination detection node according to the detection request message, and the second result is used for identifying the performance of a link from the source detection node to the destination detection node;

and the destination proxy node forwards the second result to the control node so that the control node forwards the second result.

Optionally, the method further includes:

the destination agent node receives an alarm sent by the destination detection node, wherein the alarm is generated when the destination detection node determines that the second result does not belong to the range;

and the destination agent node forwards the alarm to the control node so that the control node can forward the alarm.

Optionally, before the destination proxy node forwards the measurement instruction to the destination probe node, the method further includes:

the target agent node signs the measurement instruction; the signed measurement instruction is used for verification by the source detection node and sends the detection request message to the target detection node under the condition that the verification is passed.

Optionally, the destination proxy node is a process in a destination physical server;

before the destination agent node receives the measurement instruction, the method further includes:

the target agent node receives the control information sent by the control node;

the destination agent node creates a virtual machine and the destination probe node in the destination physical server based on the control information, wherein the source probe node is a process in the first virtual machine, and the virtual machine and the source virtual client are two virtual machines located in the source physical server.

In an eleventh aspect, an embodiment of the present invention provides an apparatus for measuring network performance. The device deploys a destination proxy node. The device comprises:

a first forwarding unit, configured to forward the measurement instruction to a destination probe node to instruct the destination probe node to receive a probe request packet sent by a source probe node and send a probe response packet corresponding to the probe request packet to the source probe node, where the probe request packet and the probe response packet are used by the source probe node to generate a first result and send the first result to the control node, the first result is used to identify performance of a link from the source probe node to the destination probe node, and the first result is used by the control node to forward the first result;

Optionally, the apparatus further comprises:

a second receiving unit, configured to receive a second result sent by the destination probe node, where the second result is generated by the destination probe node according to the probe request packet, and the second result is used to identify performance of a link from the source probe node to the destination probe node;

a second forwarding unit, configured to forward the second result to the control node, so that the control node forwards the second result.

Optionally, the apparatus further comprises:

a third receiving unit, configured to receive an alarm sent by the destination probe node, where the alarm is generated by the destination probe node when it is determined that the second result does not belong to the range;

Optionally, the apparatus further comprises:

the signature unit is used for signing the measurement instruction before the measurement instruction is sent to a target detection node; the signed measurement instruction is used for verification by the source detection node and sends the detection request message to the target detection node under the condition that the verification is passed.

the device further comprises:

a creating unit, configured to create a virtual machine and the destination probe node in the destination physical server based on the control information, where the source probe node is a process in the first virtual machine, and the virtual machine and the source virtual client are two virtual machines located in the source physical server.

Compared with the prior art, the embodiment of the invention at least has the following advantages:

in the technical solution of the embodiment of the present invention, when receiving a measurement request for instructing the control node to feed back a result for identifying the performance of a link from a source virtual client to a destination virtual client, the control node sends a measurement instruction to the source probe node and the destination probe node. Under the instruction of the measurement instruction, the source detection node sends a detection request message to the destination detection node, the destination detection node sends a detection response message corresponding to the detection request message to the source detection node, and the source detection node generates a network measurement result according to the detection request message and the detection response message and sends the result to the control node. The result is used to identify the performance of the link from the source probing node to the destination probing node. The control node may forward the result as a result corresponding to the measurement request, thereby completing measurement of network performance between the source virtual client and the destination virtual client. The source virtual client, the source detection node and the source virtual switch are all located on a source physical server, the target virtual client, the target detection node and the target virtual switch are all located on a target physical server, communication between the source virtual client and the target virtual client needs to pass through the source virtual switch and the target virtual switch, and communication between the source detection node and the target detection node needs to pass through the source virtual switch and the target virtual switch. Therefore, the measuring personnel can obtain the measurement result of the network performance between the virtual clients by initiating the measurement request to the control node without manual operation on the measured virtual clients, so that the manual operation of the measuring personnel in the network performance measurement process is simplified, and the measuring personnel can measure the network performance more simply and conveniently. Particularly, in the data center of the cloud computing service provided with a large number of virtual clients, a measurement person can flexibly initiate measurement requests for different virtual clients to the control node according to requirements and obtain corresponding measurement results, so that the data center of the cloud computing service can flexibly measure network performance between the virtual clients according to requirements, and the measurement person can conveniently and flexibly know the network performance situation between the virtual clients.

In addition, in the embodiment of the present invention, since the source detection node and the source virtual client are located in the same physical server and both communicate with the outside through the source virtual switch, and the destination detection node and the destination virtual client are located in the same physical server and both communicate with the outside through the destination virtual switch, with respect to the transmission process of the detection message between the source physical router and the destination physical server, the transmission process of the detection message between the source detection node and the destination detection node and the transmission process of the message between the source virtual client and the destination virtual client do not only include the processing process of the transmission of the message between the physical routers but also include the processing process of the message inside the physical server. Therefore, the transmission process of the detection message between the source detection node and the destination detection node is closer to the transmission process of the message between the source virtual client and the destination virtual client. It can be seen that the network performance identified by the measurement result obtained by transmitting the probe packet between the source probe node and the destination probe node is closer to the real network performance between the source virtual client and the destination virtual client than the measurement result obtained by transmitting the probe packet between the source physical router and the destination physical server.

In addition, in the embodiment of the invention, the control node can send the measurement instruction to the source detection node and the target detection node and receive the measurement result through the control node, so that the measurement personnel can obtain the measurement result only by interacting with the control node without manually configuring the functions of sending, receiving and processing the detection message on the physical router, and the technical realization of the network performance measurement is easier and more flexible.

Drawings

In order to more clearly illustrate the embodiments or technical solutions in the prior art of the present invention, the drawings used in the description of the embodiments or prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in the present application, and other drawings can be obtained by those skilled in the art according to the drawings.

Fig. 1 is a schematic diagram of a system framework involved in an application scenario according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of a system for measuring network performance according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of a system for measuring network performance according to an embodiment of the present invention;

FIG. 4 is a diagram of a network architecture suitable for use with the system of the present invention;

FIG. 5 is a diagram of a network architecture suitable for use with the system of the present invention;

FIG. 6 is a flowchart illustrating a method for measuring network performance according to an embodiment of the present invention;

FIG. 7 is a flowchart illustrating a method for measuring network performance according to an embodiment of the present invention;

FIG. 8 is a flowchart illustrating a method for measuring network performance according to an embodiment of the present invention;

FIG. 9 is a flowchart illustrating a method for measuring network performance according to an embodiment of the present invention;

FIG. 10 is a flowchart illustrating a method for measuring network performance according to an embodiment of the present invention;

FIG. 11 is a flowchart illustrating a method for measuring network performance according to an embodiment of the present invention;

FIG. 12 is a flowchart illustrating a method for measuring network performance according to an embodiment of the present invention;

FIG. 13 is a schematic structural diagram of an apparatus for measuring network performance according to an embodiment of the present invention;

FIG. 14 is a schematic structural diagram of an apparatus for measuring network performance according to an embodiment of the present invention;

FIG. 15 is a schematic structural diagram of an apparatus for measuring network performance according to an embodiment of the present invention;

FIG. 16 is a schematic structural diagram of an apparatus for measuring network performance according to an embodiment of the present invention;

FIG. 17 is a schematic structural diagram of an apparatus for measuring network performance according to an embodiment of the present invention;

FIG. 18 is a block diagram of a physical server according to an embodiment of the present invention;

FIG. 19 is a diagram illustrating a physical server according to an embodiment of the present invention;

FIG. 20 is a block diagram of a physical server according to an embodiment of the present invention;

FIG. 21 is a block diagram of a physical server according to an embodiment of the present invention;

fig. 22 is a schematic structural diagram of a physical server in the embodiment of the present invention.

Detailed Description

In order to make those skilled in the art better understand the solution of the present invention, the technical solution in the embodiment of the present invention will be clearly described below with reference to the drawings in the embodiment of the present invention, and it is obvious that the described embodiment is only a part of the embodiment of the present invention, and not all embodiments.

Currently, cloud computing services have become a hot spot in network development. In the cloud computing service, a data center integrated with a large number of servers provides computing resources and network resources for users. In a data center, servers typically employ virtualization technology. Specifically, a physical server is virtualized into a plurality of logical computers by a virtualization technology, each logical computer is a Virtual Machine (VM) on the physical server, wherein each Virtual Machine can run a different operating system, and therefore, application programs of the Virtual machines can run in spaces of independent operating systems. Virtual machines on physical servers typically include virtual clients and virtual switches. Wherein the virtual client is a virtual machine for providing users with access to. Traffic needs to be transmitted between different virtual clients through the virtual switch.

The inventor finds that with the development of cloud computing services, both service providers and service users have the need to know the network performance between virtual clients. In general, a data center of a cloud computing service is provided with a large number of virtual clients, and a measurement person needs to measure network performance between different virtual clients. However, with the conventional measurement technology of network performance, the measurement personnel needs to manually write and trigger ping commands on different virtual clients according to requirements, and the operation is too cumbersome and complicated. Therefore, the conventional data center cannot flexibly measure the network performance between the virtual clients and feed back the measurement result according to the requirement, and the measurement personnel cannot easily and flexibly know the network performance condition between the virtual clients. In addition, in order to measure the network performance between the physical routers, a measurement person configures functions of sending, receiving and processing probe messages on the physical routers, and the network performance between the physical routers can be measured by transmitting the probe messages between the physical routers. However, when a message is transmitted between virtual clients, the message needs to be transmitted between physical routers, and in addition, the message also needs to be transmitted between the virtual client and the virtual switch and forwarded by the virtual switch inside the physical server, that is, the transmission process of the message between the virtual clients includes not only the processing process of the message transmitted between the physical routers but also the processing process of the message inside the physical routers. It can be seen that the measurement result obtained by transmitting the probe packet between the physical routers cannot reflect the network performance of the packet when processed inside the physical server, and therefore the measurement result cannot truly reflect the network performance between the virtual clients.

In order to solve the above problems in the prior art, embodiments of the present invention provide a method, an apparatus, and a system for measuring network performance. The network is provided with a control node (Master) and a detection node (Probe) corresponding to the virtual client, wherein the detection node and the corresponding virtual client are positioned on the same physical server and communicate with the outside through the same virtual switch. The control node provides an interface for measuring the network performance between the virtual clients according to the requirement for the measuring personnel, and is used for receiving the measurement request sent by the terminal of the measuring personnel and sending the measurement instruction to the corresponding detection node, and is used for receiving the measurement result returned by the detection node and forwarding the measurement result to the terminal of the measuring personnel. Wherein, the measurement result sent by the detection node is measured by transmitting the detection message between the detection nodes. Therefore, on one hand, a measurer can flexibly designate a virtual client to be measured according to requirements through a measurement request, and a control node can flexibly select a corresponding detection node according to the measurement request, so that a measurement result of network performance between the virtual clients meeting the requirements is obtained and fed back to the measurer, the measurer does not need to manually operate on the measured virtual client, equipment needing manual operation for measuring the network performance by the measurer is reduced, manual operation of the measurer in a network performance measurement process is simplified, and the measurer can measure the network performance more simply and conveniently. Therefore, the data center can flexibly measure the network performance between the virtual clients and feed back the measurement result according to the requirement, and measuring personnel can conveniently and flexibly know the network performance condition between the virtual clients. On the other hand, because the detection nodes and the corresponding virtual clients are located in the same physical server and are communicated with the outside through the same virtual switch, the process of transmitting the message between the detection nodes is very close to or even the same as the process of transmitting the message between the corresponding virtual clients, and the two processes both comprise the processing process of transmitting the message between the physical routers and the processing process of transmitting the message between the virtual machines in the physical server, so that the measurement result obtained by transmitting the detection message between the detection nodes corresponding to the virtual clients more closely reflects the real network performance of the network environment where the virtual machines are located.

It should be noted that the term "measurer" mentioned herein refers to a user of a function for measuring network performance. For example, the measurement staff may be a user of the virtual client, or may also be a technician of a provider of the cloud computing service. The "terminal of the measuring person" means any device which can interact with the measuring person and can interact with the control node, and the device can send a measurement request to the control node and receive a measurement result sent by the control node. The term "Virtual Machine" refers to a Virtual Machine (VM) for providing users as user terminal devices, and "user" herein refers to a user of a Virtual Machine provided by a cloud computing service. The term "Virtual Switch" refers to a Virtual machine, such as an Open Virtual Switch (OVS) or the like, for providing a message forwarding function for a Virtual client. The virtual client and the virtual switch are both deployed in the physical server, and the virtual client and the virtual switch are virtual machines providing different functions in the physical server. Further, reference herein to "physical" in the names "physical server," "physical device," etc., means that the device of the name is a physical, hardware device. As used herein, the term "virtual" in the names "virtual client", "virtual switch", and the like means that the device of the name is a device virtualized on a physical device by a virtualization technique.

For example, one of the application scenarios of the embodiment of the present invention may be applied to the network system shown in fig. 1. In the network system, the control node 101 is located in the physical server 108, the probe node 102, the virtual client 104, and the virtual switch 106 are all located in the physical server 109, and the probe node 103, the virtual client 105, and the virtual switch 107 are all located in the physical server 110. Wherein, the probe node 102 and the virtual client 104 both communicate with the outside world through the virtual switch 106, the probe node 103 and the virtual client 105 both communicate with the outside world through the virtual switch 107, the physical server 108 communicates with the outside world through the physical router 111, the physical server 109 communicates with the outside world through the physical router 112, and the physical server 110 communicates with the outside world through the physical router 113. In the process of network performance measurement, the control node 101 receives a measurement request for instructing the control node to feed back a result for identifying the performance of the link of the virtual client 104 to the virtual client 105, transmits a first measurement instruction to the probe node 102, and transmits a second measurement instruction to the probe node 103. The probe node 102 sends a probe request message to the probe node 103 based on the first measurement instruction. The probe node 103 receives the probe request packet sent by the probe node 102 and sends a probe response packet corresponding to the probe request packet to the probe node 102 based on the second test instruction. The probe node 102 receives the probe request packet sent by the probe node 103, generates a result for identifying the performance of the link from the probe node 102 to the probe node 103 according to the probe request packet and the probe response packet, and sends the result to the control node 101. The control node 101 forwards the result with the result. It can be understood that, in the above information interaction manner, the probe node 102 corresponds to a source probe node, and the probe node 103 corresponds to a destination probe node. Accordingly, virtual client 103 corresponds to a source virtual client, and virtual client 104 corresponds to a destination virtual client. As can be seen from fig. 1, the transmission path of the probe packet between the probe node 102 and the probe node 103 is closer to the transmission path of the packet between the virtual client 104 and the virtual client 105 than the transmission path of the probe packet between the physical router 112 and the physical router 113, and therefore, the performance of the link from the virtual client 102 to the virtual client 103 can be more accurately reflected by the measurement result obtained by transmitting the probe packet between the probe node 102 and the probe node 103.

In practical applications, the virtual client 103 and the virtual client 105 involved in the measurement request may be located on different physical servers, i.e. the physical server 109 and the physical server 110 may be two different physical servers. Alternatively, the virtual client 103 and the virtual client 105 involved in the measurement request may be located on the same physical server, that is, the physical server 109 and the physical server 110 may be the same physical server. Further, the control node 101 may be located on any physical server in the network system. That is, the physical server 108 having the control node 101 may be the same physical server as the physical server 109, may be the same physical server as the physical server 110, or may be another physical server different from both the physical server 109 and the physical server 110.

It should be noted that the foregoing application scenarios are only shown for facilitating understanding of the principles of the present invention, and are not intended to limit the technical solutions provided by the embodiments of the present invention.

The following describes in detail the implementation of the method, apparatus and system for measuring network performance according to the present invention by embodiments with reference to the accompanying drawings.

Fig. 2 is a schematic structural diagram of a system for measuring network performance according to an embodiment of the present invention. In this embodiment, the system may include, for example:

a control node 201, configured to receive a measurement request, send a first measurement instruction to a source probe node 202, send a second measurement instruction to a destination probe node 203, receive a first result sent by the source probe node 202, and forward the first result, where the measurement request is used to instruct the control node to feed back a result for identifying performance of a link from a source virtual client to a destination virtual client;

the source probe node 202 is configured to receive a first measurement instruction sent by the control node, send a probe request packet to a destination probe node 203 based on the first measurement instruction, receive a probe response packet sent by the destination probe node 203, generate a first result according to the probe request packet and the probe response packet, and send the first result, where the first result is used to identify performance of a link between the source probe node 202 and the destination probe node 203;

the destination probe node 203 is configured to receive a second measurement instruction sent by the control node 201, receive the probe request packet sent by the source probe node 201, and send the probe response packet corresponding to the probe request packet to the source probe node 202 based on the second measurement instruction.

Wherein the source probe node 202, the source virtual client and the source virtual switch are located in a source physical server, the destination probe node 203, the destination virtual client and the destination virtual switch are located in a destination physical server, the source physical server and the destination physical server are two physical servers, the source probe node 202 is configured to communicate with the destination probe node 203 through the source virtual switch, the destination probe node 203 is configured to communicate with the source probe node 202 through the destination virtual switch, the source virtual client is configured to communicate with the destination virtual client through the source virtual switch, and the destination virtual client is configured to communicate with the source virtual client through the destination virtual switch. Specifically, the port of the source probe node 202 and the port of the source virtual client may both be connected to a source virtual switch, and the port of the destination probe node 203 and the port of the destination probe node may both be connected to a destination virtual switch. The port of the source probe node 202 is used for the source probe node to communicate with the outside, the port of the source virtual client is used for the source virtual client to communicate with the outside, the port of the destination probe node 203 is used for the destination node to communicate with the outside, and the port of the destination virtual client is used for the destination virtual client to communicate with the outside.

In the system provided in this embodiment, when a measurer needs to measure the network performance between the source virtual machine and the destination virtual machine, the measurer may send a measurement request to the control node 201 through a terminal of the measurer. After receiving the measurement request, the control node 201 determines a source probe node 202 corresponding to the source virtual client and a destination probe node 203 corresponding to the destination virtual client according to the measurement request. The control node 201 sends a first measurement instruction to the source probing node 202 and a second measurement instruction to the destination probing node 203. After receiving the first measurement instruction, the source probe node 202 sends a probe request packet to the destination probe node 203 based on the first measurement instruction. After receiving the second measurement instruction and the probe request packet, the destination probe node 203 sends a probe response packet corresponding to the probe request packet to the source probe node 202 based on the second measurement instruction. The source probe node 202 generates a first result for identifying the performance of the link from the source probe node 202 to the destination probe node 203 according to the probe request packet and the probe response packet, and sends the first result to the control node 201. After receiving the first result, the control node 201 forwards the first result to the terminal of the measurement person as a result of identifying the performance of the link from the source virtual client to the destination virtual client, so that the measurement person obtains the first result through the terminal of the measurement person. The measurement request may be from a user using the virtual client, that is, a user using the cloud computing service, or the user measurement request may also be from an operator user providing the cloud computing service, where the operator user deploys a data center of the cloud computing service for the user using the cloud computing service. Furthermore, the control node 201 may be located on any one of the physical servers in the network. For example, if there is a physical server acting as a controller in the network, the control node 201 may be located on the physical server acting as the controller.

In a specific embodiment, the measurement request may carry an identifier of the source virtual client, an identifier of the destination virtual machine, and a service identifier for indicating a network performance measurement service. The control node 201 may determine, according to the service identifier in the measurement request, that the measurement request is used for requesting a service for network performance measurement. The control node 201 may determine, according to the identifier of the source virtual client and the identifier of the destination virtual client in the measurement request, that the measurement request is directed to the source virtual client and the destination virtual client, so as to determine a source probe node 202 corresponding to the source virtual client and a destination probe node 203 corresponding to the destination virtual client. It can be understood that the measurement personnel may specify the source virtual client and the destination virtual client that need to measure the network performance according to the requirement, that is, the measurement request may carry the identifier of the source virtual client and the identifier of the destination virtual client that the measurement personnel specify according to the requirement, so that the control node 201 may measure the network performance between the specified source virtual client and the specified destination virtual client according to the requirement.

In addition, the measurement request may also carry a type of the indicator to which the result for identifying the performance of the link from the source virtual client to the destination virtual client belongs. The control node 201 may determine the type of the indicator to which the result of the measurement request measurement belongs according to the type in the measurement request. If the system shown in fig. 2 is capable of measuring results of a plurality of different types of indicators of network performance, the measurement request may carry the type of the indicator specified according to the requirement, so that the control node 201 may measure the result of the indicator of the specified type according to the requirement. Wherein the index is used to reflect performance of a link from the source virtual client to the destination virtual client. The type of the index can be time delay, jitter, packet loss rate or load rate, etc. .

Further, after receiving the measurement request, the control node 201 may extract the identifier of the source virtual client, the identifier of the destination virtual client, the service identifier, and the type from the measurement request, and generate the first measurement instruction and the second measurement instruction based on the identifier of the source virtual client, the identifier of the destination virtual client, the service identifier, and the type. Thus, the first measurement instruction and the second measurement instruction may also carry the identifier of the source virtual client, the identifier of the destination virtual machine, the service identifier, and the type. It should be noted that, in order to measure the results of different types of metrics, the source probe node 202 needs to send the probe request packet to the destination probe node 203 in different sending manners. Therefore, when the measurement request carries the type, the source probe node 202 may perform the following operations in order to implement the sending of the probe request packet to the destination probe node based on the first measurement instruction: determining a sending mode of the detection request message according to the type in the first measurement instruction; and sending the detection request message to the target detection node according to the sending mode. The sending method may represent the sending times, sending intervals, and the like of the probe request packet.

In a specific embodiment, the source probe node 202 may generate results of different types of indicators according to different receiving conditions of the probe response packet. For example, if the source probe node 202 receives the probe response packet within a predetermined time after sending the probe request packet, the result of the network delay of the link from the source probe node 202 to the destination probe node 203 may be generated according to the receiving time of the probe response packet. For another example, if the source probe node 202 cannot receive the probe response packet within a predetermined time after sending the probe request packet, the current probe event may be used as a packet loss event, and a result of a network packet loss rate of a link from the source probe node 202 to the destination probe node 203 is generated.

In this embodiment, various embodiments may be used to deploy the source probe node 202 and the destination probe node 203.

As an example, the source probe node 202 may be deployed in the source virtual client and the destination probe node is deployed in the destination virtual client. Specifically, the source probe node 202 may be a process in the source virtual client, a port of the source probe node 202 is equivalent to a port of the source virtual client, and a manner of communicating with the outside world by the source probe node 202 is a manner of communicating with the outside world by the source virtual client. Accordingly, the destination probe node 203 may be a process in the destination virtual client, a port of the destination probe node 203 is equivalent to a port of the destination virtual client, and a manner of communicating with the outside world by the destination probe node 203 is a manner of communicating with the outside world by the destination virtual client. Therefore, the transmission path of the probe request message and the probe response message between the source probe node 202 and the destination probe node 203 is the transmission path of the message between the source virtual client and the destination virtual client. As can be seen, the network performance reflected by the first result measured by the probe request message and the probe response message is equivalent to the real performance of the link from the source virtual client to the destination virtual client.

As another example, the source probe node 202 may be deployed in a first virtual machine, a port of the source probe node 202 is a port of the first virtual machine, the first virtual machine and the source virtual client are two virtual machines located in a source physical server, and the first virtual machine is only used for deploying the source probe node 202; the destination probe node 203 may be deployed in a second virtual machine, a port of the destination probe node 203 is a port of the second virtual machine, the first virtual machine and the source virtual client are two virtual machines located in the source physical server, and the second virtual machine is only used for deploying the destination probe node. Specifically, on a source physical server where the source probing node 202, the source virtual client, and the source virtual switch are deployed at the same time, a first virtual machine only for deploying the source probing node 202 may be deployed. That is, the first virtual machine is another virtual machine on the source physical server, which is independent of the source virtual client and the source virtual switch, the source probe node is a process in the first virtual machine, and the first virtual machine is only used for completing the functions of the source probe node 202. Wherein the port of the first virtual machine and the port of the source virtual client are both connected to the source virtual switch. Accordingly, on a destination physical server where the destination probe node 203, the destination virtual client, and the destination virtual switch are deployed at the same time, a second virtual machine only for deploying the destination probe node 203 may be deployed. That is, the second virtual machine is another virtual machine on the destination physical server, which is independent of the destination virtual client and the destination virtual switch, the source probe node is a process in the first virtual machine, and the second virtual machine is only used for completing the function of the destination probe node 203. Wherein the port of the second virtual machine and the port of the destination virtual client are both connected to the destination virtual switch. Therefore, the transmission path of the probe request message and the probe response message between the source probe node 202 and the destination probe node 203 is close to the transmission path of the message between the source virtual client and the destination virtual client. As can be seen, the network performance reflected by the first result measured by the probe request message and the probe response message is close to the real performance of the link from the source virtual client to the destination virtual client.

In practical application, measurement personnel often pay more attention to abnormal conditions of network performance. The system shown in fig. 2 may feed back to the measurer an abnormal situation of the network performance based on the measured first result. As an example, the source probing node 202 may initiate an alarm based on the first result. Specifically, in the system, the source probe node 202 may be further configured to determine whether the first result belongs to a first range, and generate and send a first alarm to the control node 201 when the first result does not belong to the first range; the control node 201 may be further configured to receive the first alarm sent by the meta-probe node, and forward the first alarm. As another example, the controlling node 201 may initiate an alarm based on the first result. Specifically, in the system, the control node 201 is further configured to determine whether the first result belongs to a third range, and generate a third alarm and forward the third alarm when the first result does not belong to the third range.

Wherein the first range is a result range set in advance in the source probe node 202 for the first result. The first range corresponds to a normal state of network performance. If the first result belongs to the first range, the source probe node 202 determines that the network performance is normal. If the first result does not belong to the first range, the source probe node 202 determines that the network performance is abnormal. The third range is a result range set in advance in the control node 201 for the first result. The third range corresponds to a normal state of network performance. If the first result belongs to the first range, the control node 201 determines that the network performance is normal. If the first result does not belong to the third range, the control node 201 determines that the network performance is abnormal. It is understood that the first range and the third range may be the same or different.

In the process of transmitting the probe request packet and the probe response packet between the source probe node 202 and the destination probe node 203, except that the source probe node 202 may measure the performance of the link from the source probe node 202 to the destination probe node 203 according to the receiving condition of the probe response packet, the destination probe node 203 may also measure the performance of the link from the source probe node 202 to the destination probe node 203 according to the receiving condition of the probe request packet. Specifically, in the system, the destination probe node 203 is further configured to generate a second result according to the probe request packet, and send the second network performance parameter, where the second result is used to identify the performance of the link from the source probe node 202 to the destination probe node 203; the control node 201 is further configured to receive the second result and forward the second result.

Further, if the destination probe node 203 measures the second result, the system shown in fig. 2 may feed back an abnormal situation of the network performance to the measurement personnel based on the measured second result. As an example, the destination probing node 203 may initiate an alarm based on the second result. Specifically, in the system, the destination probe node 203 is further configured to determine whether the second result is in a second range, and generate and send a second alarm to the control node 201 when the second result does not belong to the second range; the control node 201 is further configured to receive the second alarm and forward the second alarm. As another example, control node 201 may initiate an alarm based on the second result. Specifically, in the system, the control node 201 is further configured to determine whether the second result belongs to a fourth range, and generate and forward a fourth alarm when the second result does not belong to the fourth range.

Wherein the second range is a result range set in advance in the destination probe node 203 for the second result. The second range corresponds to a normal state of network performance. If the second result belongs to the second result range, the destination probe node 203 determines that the network performance is normal. If the second result does not belong to the second range, the destination probe node 203 determines that the network performance is abnormal. The fourth range is a result range set in advance in the control node 201 for the second result. The fourth range corresponds to a normal state of network performance. If the second result belongs to the fourth range, the control node 201 determines that the network performance is normal. If the second result does not belong to the fourth range, the control node 201 determines that the network performance is abnormal. It is understood that the second range and the fourth range may be the same or different.

In order to better adapt to the network of the cloud computing data center, the system of this embodiment may further add an Agent node (Broker or Agent) between the control node and the probe node, and information interacted between the control node and the probe node is forwarded by the Agent node. As shown in fig. 3, the system of this embodiment may further include, on the basis of fig. 2:

a source proxy node 301, configured to receive the first measurement instruction sent by the control node 201 and forward the first measurement instruction to the source probe node 202, and receive the first result sent by the source probe node 202 and forward the first result to the control node 201;

a destination proxy node 302, configured to receive the second measurement instruction sent by the control node 201 and forward the second measurement instruction to the destination probe node.

In the system shown in fig. 3, after receiving the measurement request, the control node 201 determines a source probe node 202 corresponding to the source virtual client and a destination probe node 203 corresponding to the destination virtual client according to the measurement request. The control node 201 sends the first measurement instruction to a source proxy node 301 corresponding to the source probe node 201 and sends the second measurement instruction to a destination proxy node 302 corresponding to the destination probe node 203. After receiving the first measurement instruction, the source proxy node 301 determines, according to the first measurement instruction, the source probe node 202 corresponding to the first measurement instruction, and forwards the first measurement instruction to the source probe node 202. After receiving the second measurement instruction, the destination proxy node 302 determines the destination probe node 203 corresponding to the second measurement instruction according to the second measurement instruction, and forwards the measurement node to the destination probe node 203. After receiving the first measurement instruction, the source probe node 202 sends a probe request packet to the destination probe node 203 based on the first measurement instruction. After receiving the second measurement instruction and the probe request packet, the destination probe node 203 sends a probe response packet corresponding to the probe request packet to the source probe node 202 based on the second measurement instruction. The source probe node 202 generates a first result for identifying the performance of the link from the source probe node 202 to the destination probe node 203 according to the probe request packet and the probe response packet, and sends the first result to the source proxy node 301. After receiving the first result, the source proxy node 301 forwards the first result to the control node 201. After receiving the first forwarding, the control node 201 forwards the first result.

In a specific embodiment, after the source probe node 202 receives the first measurement instruction, a first preparation work of network performance measurement may be performed, and a first ready message may be sent to the source proxy node 301 after the first preparation work is completed. The source proxy node 301 sends a first start measurement instruction to the source probing node 202 after receiving the first ready information. After receiving the first start instruction, the source probe node 202 sends the probe request packet to the destination probe node 203 and receives the probe response packet sent by the destination probe packet 203. The first preparation work includes, for example, the source probe node 202 opening a port for sending the probe request packet to the destination probe node 203 and receiving the probe response packet sent by the destination probe node 203. Accordingly, after the destination probe node 203 receives the second measurement instruction, a second preparation of network performance measurement may be performed, and a second ready message may be sent to the destination agent node 302 after the second preparation is completed. The destination agent node 302 sends a first start measurement instruction to the destination probing node 203 after receiving the second ready information. After receiving the first start instruction, the destination probe node 203 receives the probe request packet sent by the source probe node 202 and returns the probe response packet to the source probe node 202. The second preparation work includes, for example, the destination probe node 203 opening a port for receiving the probe request packet sent by the source probe node 202 and returning the probe response packet to the destination probe node 203.

It is understood that in the system shown in fig. 3, besides the first measurement instruction, the second measurement instruction and the first result, other information interacted between the control node and the probe node may also be forwarded through the proxy node.

For example, if the destination probe node 203 measures the second result, the destination probe node 203 may send the second result to the control node 201 through the destination proxy node 302. Specifically, in the system shown in fig. 3, the destination proxy node 302 may be further configured to receive the second result sent by the destination probe node 203 and forward the second result to the control node 201.

For another example, if the source detecting node 201 initiates the first alarm based on the first result, the source detecting node 202 may send the first alarm to the control node 201 through the source proxy node 301. Specifically, in the system shown in fig. 3, the source proxy node 301 may be further configured to receive the first alarm sent by the source probe node 202 and forward the first alarm to the control node 201.

For another example, if the destination probe node 203 initiates the second result alarm based on the second result, the destination probe node 203 may send the second alarm to the control node 201 through the destination proxy node 302. Specifically, in the system shown in fig. 3, the destination proxy node 302 may be further configured to receive the second alarm sent by the destination probe node 203 and forward the second alarm to the control node 201.

In some network environments, the information transfer between the source proxy node 301 and the source probe node 202 may not be isolated from other information transfers. In such a network environment, the source proxy node 301 may sign information that needs to be sent to the source probe node 202, and the source probe node 202 may verify the signature of the received information, thereby ensuring the security of information transmission between the source proxy node 301 and the source probe node 202. Specifically, in the system shown in fig. 3, the source proxy node 301 may be further configured to sign the first measurement instruction before forwarding the first measurement instruction; the source probe node 202 may be further configured to verify the signed first measurement instruction after receiving the first measurement instruction, and send a probe request packet to the destination probe node 203 when the verification is passed.

Accordingly, in some network environments, the information transfer between the destination proxy node 302 and the destination probe node 203 may not be isolated from other information transfers. In such a network environment, the destination proxy node 302 may sign the information that needs to be sent to the destination probe node 203, and the destination probe node 203 may verify the signature of the received information, thereby ensuring the security of information transmission between the destination proxy node 302 and the destination probe node 203. Specifically, in the system shown in fig. 3, the destination proxy node 302 may be further configured to sign the second measurement instruction before forwarding the second measurement instruction; the destination probe node 303 may be further configured to verify the signed second measurement instruction after receiving the second measurement instruction, and send the probe response packet corresponding to the probe request packet to the source probe node 201 based on the second measurement instruction when the second measurement instruction is verified.

In this embodiment, various embodiments may be used to deploy the source proxy node 301 and the destination proxy node 302.

As an example, only one single proxy node may be deployed within the network corresponding to the control node 201, i.e., the proxy nodes in communication with the control node 201 include only the single proxy node. The information transmitted between the control node 201 and all probing nodes under the control of the control node 201 is forwarded via the only one proxy node. The only one proxy node may be deployed on any one physical server in the network that can communicate with the control node 201 and all the probe nodes under the control of the control node 201. For example, the only one proxy node may be deployed together with the control node 201 on a physical server as a controller. As another example, the only one proxy node may be deployed on the same physical server with one or more virtual clients. As another example, the only one proxy node may be deployed on an independent physical server, where the independent physical server has only the only one proxy node deployed, and the independent physical server does not have the control node 201, any virtual client, and any probe node deployed.

It is understood that if only one proxy node is deployed in the network corresponding to the control node 201, the source proxy node 301 and the destination proxy node 302 are actually the same proxy node located in the same physical server, or the source proxy node 301 and the destination proxy node 302 are both the only one proxy node. Because only one physical server in the network is deployed with the only one proxy node, most or all of the probe nodes corresponding to the virtual clients are not deployed on the physical server on which the only one proxy node is deployed. It can be seen that if the source proxy node 301 and the destination proxy node 302 are the same proxy node located on the same physical server, the source probe node 202 and the source proxy node 301 may not be on the same physical server, and the destination probe node 203 and the destination proxy node 302 may not be on the same physical server. In order to enable communication between the source probe node 202 and the source proxy node 301, in the system shown in fig. 3, the control node 201 may be further configured to, before sending the first measurement instruction, obtain a first route from the source proxy node 301 to the source probe node 202 and a second route from the source probe node 202 to the source proxy node 301, send the first route to the source proxy node 301, and send the second route to the source probe node 202, where the first route is used for the source proxy node 301 to send the first measurement instruction to the source probe node 202, and the second route is used for the source probe node 02 to send the first result to the source proxy node 301. In addition, the first route and the second route may also be used for transmitting other information between the source probe node 202 and the source proxy node 301, for example, the second route may be used for transmitting the aforementioned first alarm. Accordingly, in order to enable communication between the destination probe node 203 and the destination proxy node 302, in the system shown in fig. 3, the control node 201 may be further configured to, before sending the second measurement instruction, obtain a third route from the destination proxy node 302 to the destination probe node 203 and a fourth route from the destination probe node 203 to the destination proxy node 302, send the third route to the destination proxy node 302, and send the fourth route to the destination probe node 203, where the third route is used for the destination proxy node 301 to send the second measurement instruction to the destination probe node 203, and the fourth route is used for the destination probe node 203 to send the second result to the destination proxy node. In addition, the third route and the fourth route may also be used for transmitting other information between the destination probe node 203 and the destination proxy node 301, for example, the fourth route may be used for transmitting the aforementioned second alarm.

If the source proxy node 301 and the destination proxy node 302 are the only one proxy node, in the foregoing embodiment related to the first ready information and the second ready information, the only one proxy node may send the first measurement starting instruction to the source probing node 202 and send the second measurement starting instruction to the destination probing node 203 after the first ready information and the second ready information are both received.

As another example, a proxy node may be respectively deployed on each physical server on which a virtual client and a probe node are deployed in the network, that is, the proxy node corresponding to the probe node and the probe node are deployed on the same physical server. If the detection node and the agent node are deployed on the same physical server, the detection node and the agent node can communicate with each other. If the detection node and the agent node are deployed in different physical servers, communication between the detection node and the agent node is not possible. Since the probe nodes under control of the control node 201 are typically deployed on a plurality of different physical servers, the control node 201 may typically communicate with agent nodes deployed on a plurality of physical servers.

It can be understood that, if a proxy node is respectively deployed on each physical server on which a virtual client and a probe node are deployed in the network, the source proxy node 301 and the source probe node 202 are both deployed in the same physical server, and the destination proxy node 302 and the destination probe node 203 are both deployed in the same physical server. In other words, the source proxy node 301 is a process in a source physical server, and the destination proxy node 302 is a process in a destination physical server, wherein the source physical server represents a physical server on which the source probe node 202 is deployed, and the destination physical server represents a physical server on which the destination probe node 203 is deployed.

If the source proxy node 301 and the source probe node 202 are both deployed in the source physical server and the destination proxy node 302 and the destination probe node 203 are both deployed in the destination physical server, the source probe node 202 and the source proxy node 301 may communicate with each other through a socket, and the destination probe node 203 and the destination proxy node 302 may communicate with each other through a socket. Specifically, in the source physical server, the source probe node 202 may be deployed in a source Namespace (Namespace), and the source proxy node 301 may be deployed outside the source Namespace. At this time, the source proxy node 301 and the source probe node 202 may communicate with each other through a socket. In the destination physical server, the destination probe node 203 may be deployed in the destination Namespace, and the destination proxy node 302 may be deployed outside the destination Namespace. At this time, the destination proxy node 302 and the destination probe node 203 may communicate with each other through a socket. The source Namespace and the target Namespace can be Linux Network Namespace specifically. The source Namespace may be a Namespace opened in a virtual machine in which the source probe node 202 is deployed, and if the source Namespace is a Namespace in the source virtual client or a Namespace in the first virtual machine. The destination Namespace may be a Namespace opened in a virtual machine in which the destination probe node 203 is deployed, and if the destination Namespace is the Namespace in the destination virtual client or the Namespace in the second virtual machine. The socket body may be a UNIX domain socket and the socket body may be a UNIX domain socket. It is understood that Linux NetworkNamespace can isolate the probe request message and the probe response message transmitted between the source probe node 202 and the destination probe node 203 from other transmission information.

If the source proxy node 301 and the source probe node 202 are both deployed in the source physical server and the destination proxy node 302 and the destination probe node 203 are both deployed in the destination physical server, the source probe node 202 may be dynamically created by the source proxy node 301 and the destination probe node 203 may be dynamically created by the destination proxy node 302. Specifically, in the system shown in fig. 3, the control node 201 may be further configured to send first control information to the source proxy node 301 and second control information to the destination proxy node 302 before sending the first measurement instruction and the second measurement instruction, designate a first port on the source virtual switch for the source probe node 202 and designate a second port on the destination virtual switch for the destination probe node 203, and establishing a Virtual Local Area Network (VLAN) including the first port and the second port, the source probing node 202 is able to communicate with the destination probing node 203 through the first port, the destination probe node 203 can communicate with the source probe node 202 through the second port, and the virtual local area network is configured to transmit the probe request packet and the probe response packet; the source proxy node 301 may be further configured to receive the first control information, and create the first virtual machine and the source probe node 202 in the source physical server based on the first control information; the destination proxy node 301 may be further configured to receive the second control information, and create the second virtual machine and the destination probe node 203 in the destination physical server based on the second control information. The VLAN may be a dedicated network for measurement, that is, the VLAN transmits only the probe request message and the probe response message, and does not transmit other information. Furthermore, after the source proxy node 301 has created said source probe node 202, the source proxy node 301 may send a third ready information to the control node 201. The control node 201 may send said first measurement instruction to the source proxy node 301 after receiving said third ready information. Accordingly, after the destination proxy node 302 has created the destination probe node 203, the destination proxy node 302 may send a fourth ready information to the control node 201. The control node 201 may send said second measurement instruction to the destination agent node 302 after receiving said fourth ready information.

If the source proxy node 301 and the source probe node 202 are both deployed in the source physical server and the destination proxy node 302 and the destination probe node 203 are both deployed in the destination physical server, the source proxy node 301 and the destination proxy node 302 may be the same proxy node deployed in the same physical server or may be different proxy nodes deployed in different physical servers. Specifically, if the source virtual client and the destination virtual client are deployed in different physical servers, the source probe node 202 and the destination probe node 203 are deployed in different physical servers, that is, the source physical server and the destination physical server are different physical servers, and at this time, the source proxy node 301 and the destination proxy node 302 are different proxy nodes deployed in different physical servers. If the source virtual client and the destination virtual client are deployed in the same physical server, the source probe node 202 and the destination probe node 203 are deployed in the same physical server, that is, the source physical server and the destination physical server are the same physical server, and at this time, the source proxy node 301 and the destination proxy node 302 are the same proxy node deployed in the same physical server.

In this embodiment, the system shown in fig. 3 can be applied to networks of different architectures.

As an example, the system shown in fig. 3 may be applied to the network architecture shown in fig. 4. The network architecture shown in fig. 4 includes a physical server 401 as a controller, a physical server 402 with virtual clients deployed, a control network 403, and a forwarding network 404. The control node 201 may be deployed on a physical server 401. The network architecture shown in fig. 4 may deploy a proxy node capable of communicating with the control node 201 on only one physical server 402, where the source proxy node 301 and the destination proxy node 302 are the same proxy node deployed on the same physical server 402. The network architecture shown in fig. 4 may respectively deploy a proxy node on each physical server 402, which may communicate with the control node 201, in which case the source proxy node 301 and the destination proxy node 302 may be the same proxy node deployed on the same physical server 402, or may be different proxy nodes deployed on different physical servers 402. The probe node is deployed on the same physical server 402 as its corresponding virtual client. In particular, the source probe node 202 is deployed on the same physical server 402 as the source virtual client. The destination probe node 203 is deployed on the same physical server 402 as the virtual client. The control network 403 is used for transmitting management or control information, such as the first measurement command, the second measurement command, the first network performance parameter, and the like, between the physical server 401 and the physical server 402. The data network is used for transmitting information between the virtual machines on the physical server 402 and transmitting information between the virtual machines and the user on the physical server 402, such as the probe request message and the probe response message.

As another example, the system shown in fig. 3 may be applied to the network architecture shown in fig. 5. The network architecture shown in fig. 5 includes a physical server 501 as a controller, a physical server 502 with virtual clients deployed, and a forwarding network 503. The control node 201 may be deployed on a physical server 501. The network architecture shown in fig. 5 may deploy a proxy node capable of communicating with the control node 201 on only one physical server 502, where the source proxy node 301 and the destination proxy node 302 are the same proxy node deployed on the same physical server 502. The network architecture shown in fig. 4 may respectively deploy a proxy node on each physical server 502, where the proxy node can communicate with the control node 201, and the source proxy node 301 and the destination proxy node 302 may be the same proxy node deployed on the same physical server 502 or different proxy nodes deployed on different physical servers 502. The probe node is deployed on the same physical server 502 as its corresponding virtual client. In particular, the source probe node 202 is deployed on the same physical server 502 as the source virtual client. The destination probe node 203 is deployed on the same physical server 502 as the virtual client. The forwarding network 503 is used for transmitting information related to management or control between the physical server 501 and the physical server 502, such as the aforementioned first measurement instruction, the aforementioned second measurement instruction, the aforementioned first result, and the like. The forwarding network 503 is also used for transmitting information between the virtual machines on the physical server 502 and the user, such as the aforementioned probe request message, the aforementioned probe response message, and the like.

Through the technical scheme of the embodiment, on one hand, a measurer can flexibly designate a virtual client to be measured according to requirements through a measurement request, and a control node can flexibly select a corresponding detection node according to the measurement request, so that a measurement result of network performance between the virtual clients meeting the requirements is obtained and fed back to the measurer, and the measurer does not need to manually operate on the measured virtual client, thereby reducing equipment which needs to be manually operated when the measurer measures the network performance, simplifying the manual operation of the measurer in the network performance measurement process, and enabling the measurer to more simply measure the network performance. Therefore, the data center can flexibly measure the network performance between the virtual clients and feed back the measurement result according to the requirement, and measuring personnel can conveniently and flexibly know the network performance condition between the virtual clients. On the other hand, because the detection nodes and the corresponding virtual clients are located in the same physical server and are communicated with the outside through the same virtual switch, the process of transmitting the message between the detection nodes is very close to or even the same as the process of transmitting the message between the corresponding virtual clients, and the two processes both comprise the processing process of transmitting the message between the physical routers and the processing process of transmitting the message between the virtual machines in the physical server, so that the measurement result obtained by transmitting the detection message between the detection nodes corresponding to the virtual clients more closely reflects the real network performance of the network environment where the virtual machines are located.

In order to make the specific application of the embodiments of the present invention more clearly understood by those skilled in the art, the following describes the embodiments of the present invention in two exemplary application scenarios.

In a first exemplary application scenario, a source probe node is deployed in a source virtual client, and a destination virtual client deploys a destination probe node. Only one agent node capable of communicating with the control node is deployed in the network system, namely the source agent node and the destination agent node are the same agent node. In the first exemplary application scenario, fig. 6 is a flowchart illustrating a method for measuring network performance according to an embodiment of the present invention. In this embodiment, the method may include, for example:

601. the control node receives a measurement request.

602. And the control node acquires the related information of the source virtual client and the target virtual client according to the measurement request.

Wherein the related information may include an identification of the source virtual client and an identification of the destination virtual client, and may further include an identification of the source probe node in the source virtual client and an identification of the destination probe node in the destination virtual client.

603. The control node establishes a first route from the proxy node to the source probe node, a second route from the source probe node to the proxy node, a third route from the proxy node to the destination probe node, and a fourth route from the destination probe node to the proxy node.

604. And the control node sends a first measurement instruction and a second measurement instruction to the proxy node.

605. And the proxy node signs the first measurement instruction and forwards the signed first measurement instruction to the source detection node.

606. And after receiving the signed first measurement instruction, the source detection node verifies the signed first measurement instruction. In case the authentication is passed 607 may be entered.

It is understood that if the verification in 606 is not passed, the source probe node may terminate the flow, or may feed back error information to the proxy node.

607. The source detection node executes first preparation work of network performance measurement, and returns first ready information to the agent node after the first preparation work is completed.

608. And the proxy node signs the second measurement instruction and forwards the signed second measurement instruction to the target detection node.

609. And after receiving the signed second measurement instruction, the target detection node verifies the signed second measurement instruction. In the case of a verification pass, 610 may be entered.

It is understood that if the verification fails in 609, the destination probe node may terminate the flow, or may feed back an error message to the proxy node.

610. And the destination detection node executes second preparation work of network performance measurement and returns second ready information to the agent node after the second preparation work is finished.

It should be noted that 605-607 and 608-610 are processes without any execution order limitation after 604 is executed. For example, the present embodiment may first perform the processes 605 to 607 and then perform the processes 608 to 610, or the present embodiment may also first perform the processes 608 to 610 and then perform the processes 605 to 607, or the present embodiment may also perform the processes 605 to 607 and the processes 608 to 610 simultaneously.

611. And after receiving the first ready information and the second ready information, the proxy node sends a first measurement starting instruction to the source detection node and sends a second measurement starting instruction to the destination detection node.

612. And after receiving the first measurement starting instruction, the source detection node sends a detection request message to the target detection node.

613. And after receiving the second test starting instruction, the destination detection node receives the detection request message sent by the source detection node and sends a detection response message corresponding to the detection request message to the source detection node.

614. And the source detection node receives the detection response message sent by the target detection node, generates a result for identifying the performance of the link from the source detection node to the target detection node according to the detection request message and the detection response message, and sends the result to the proxy node.

615. And after receiving the result, the proxy node forwards the result to the control node.

616. The control node forwards the result after receiving the result.

617. And the source detection node determines whether the result belongs to a preset range.

Wherein the preset range represents a result range of normal network performance.

618. And after determining that the result does not belong to the preset range, the source detection node generates an alarm and sends the alarm to the proxy node.

619. And after receiving the alarm, the agent node forwards the alarm to the control node.

620. And after receiving the alarm, the control node forwards the alarm.

In the technical solution of this embodiment, by deploying a control node and a proxy node in the network system and deploying a probe node in the virtual client, the network system can flexibly measure the network performance between different virtual clients according to the user requirements. Moreover, the path for transmitting the detection message between the detection nodes is the path for transmitting the message between the virtual clients, so that the measurement result can accurately reflect the real network performance between the virtual clients.

In a second exemplary application scenario, a source probe node is deployed in a first virtual machine, a port of the first virtual machine and a port of a source virtual client are connected to a source virtual switch, a destination probe node is deployed in a second virtual machine, and a port of the second virtual machine and a port of a destination virtual client are connected to a destination virtual switch. In the network system, a proxy node is respectively deployed on each physical server deployed with a virtual client, that is, a source proxy node and a source detection node are both deployed on a source physical server, and a destination proxy node and a destination detection node are both deployed on a destination physical server. In the second exemplary application scenario, fig. 7 is a flowchart illustrating a method for measuring network performance according to an embodiment of the present invention. In this embodiment, the method may include, for example:

701. the control node receives a measurement request.

702. And the control node acquires the related information of the source virtual client and the target virtual client according to the measurement request.

Wherein, the related information can be referred to the embodiment shown in fig. 6.

703. The control node establishes a VLAN for measurement, a first port is allocated to the VLAN on a source virtual switch of a source physical server, and a second port is allocated to the VLAN on a destination virtual switch of a destination physical server.

704. And the control node sends first control information to the source proxy node and second control information to the destination proxy node.

705. The source proxy node creates the first virtual machine and the source probe node in the source physical server after receiving the first control information.

The first virtual machine opens up a source Namespace, and the source detection node is deployed in the source Namespace.

706. The source agent node returns first ready information to the control node.

It will be appreciated that the first ready information corresponds to the third ready information in the foregoing system embodiment.

707. And after receiving the second control information, the destination agent node creates the second virtual machine and the destination probe node in the destination physical server.

And opening a target Namespace in the second virtual machine, wherein the target detection node is deployed in the target Namespace.

708. And the destination agent node returns second ready information to the control node.

It will be appreciated that the second ready information corresponds to the fourth ready information in the foregoing system embodiment.

709. The control node assigns the first port to the first virtual machine and the second port to the second virtual machine.

Wherein the source probe node is capable of communicating with the destination probe node through the first port, and the destination probe node is capable of communicating with the source probe node through the second port.

710. And the control node sends a first measurement instruction to the source proxy node and sends a second measurement instruction to the destination proxy node.

711. And after receiving the first measurement instruction, the source proxy node forwards the first measurement instruction to the source detection node.

712. And after receiving the second measurement instruction, the destination proxy node forwards the second measurement instruction to the source detection node.

713. And after receiving the first measurement instruction, the source detection node sends the detection request message to the target detection node.

714. And after receiving the second measurement instruction, the destination detection node receives the detection request message sent by the source detection node, and sends a detection response message corresponding to the detection request message to the source detection node.

715. And the source detection node receives the detection response message sent by the target detection node, generates a result for identifying the performance of the link from the source detection node to the target detection node according to the detection request message and the detection response message, and sends the result to the source proxy node.

716. And after receiving the result, the source proxy node forwards the result to the control node.

717. The control node forwards the result after receiving the result.

718. And the source detection node determines whether the result belongs to a preset range.

Wherein the preset range of network performance represents a result range of normal network performance.

719. And after determining that the result does not belong to the preset range, the source detection node generates an alarm and sends the alarm to the proxy node.

720. And after receiving the alarm, the source proxy node forwards the alarm to the control node.

721. And after receiving the alarm, the control node forwards the alarm.

In the technical solution of this embodiment, by deploying a control node in the network system, respectively deploying a proxy node in each physical server where the virtual client is located, and deploying a probe node in another virtual machine connected to the same virtual switch as the virtual client, the network system can flexibly measure the network performance between different virtual clients according to the requirements. Moreover, the path for transmitting the detection message between the detection nodes is closer to the path for transmitting the message between the virtual clients than the path for transmitting the message between the physical routers, so that the measurement result can more accurately reflect the real network performance between the virtual clients.

Fig. 8 is a flowchart illustrating a method for measuring network performance according to an embodiment of the present invention. In this embodiment, for example, the method may specifically include:

801. receiving a measurement request by a control node, wherein the measurement request is used for instructing the control node to feed back a result for identifying the performance of a link from a source virtual client to a destination virtual client;

802. the control node sends a first measurement instruction to a source detection node and sends a second measurement instruction to a destination detection node, wherein the first measurement instruction is used for instructing the source detection node to send a detection request message to the destination detection node, receiving a detection response message corresponding to the detection request message and generating a first result according to the detection request message and the detection response message, the first result is used for marking the performance of a link from the source detection node to the destination detection node, and the second measurement instruction is used for instructing the destination detection node to send the detection response message corresponding to the detection request message to the source detection node;

803. the control node receives the first result sent by the source detection node and forwards the first result;

Optionally, the method further includes:

the control node determining whether the first result belongs to a third range;

Alternatively to this, the first and second parts may,

the 802 specifically includes: the control node sends the first measurement instruction to a source proxy node to instruct the source proxy node to forward the first measurement instruction to the source probe node; the control node sends the second measurement instruction to a destination proxy node to instruct the destination proxy node to forward the second measurement instruction to the destination detection node;

the 803 is specifically: the control node receives the first result forwarded by the source proxy node, the first result being sent by the source probe node to the source proxy node.

Optionally, the receiving, by the control node, the second result sent by the destination probe node specifically includes: the control node receives the second result forwarded by the destination proxy node, the second result being sent to the destination proxy node by the destination probe node.

Alternatively to this, the first and second parts may,

prior to the 802, the method further comprises:

It should be noted that the control node in the embodiment of the present invention corresponds to the control node 201 in the embodiment shown in fig. 2. For various specific embodiments of the method executed by the control node in this embodiment, reference may be made to the detailed description of the embodiment shown in fig. 2, which is not described herein again.

In the technical scheme of this embodiment, a user can flexibly designate a virtual client machine that needs to measure network performance according to own requirements through a measurement request, and a control node can flexibly issue a measurement instruction to a corresponding detection node and receive a returned measurement result according to the requirements indicated by the measurement request when receiving the measurement request, so that a data center can flexibly measure and feed back network performance between the virtual client machines according to user requirements. In addition, the user can also flexibly indicate other requirements such as the type of network performance parameters and the like through the measurement request, and the control node can flexibly obtain measurement results meeting various requirements of the user according to the indication of the measurement request and feed the measurement results back to the user. Therefore, the data center can flexibly meet various requirements of users on network performance measurement.

Fig. 9 is a flowchart illustrating a method for measuring network performance according to an embodiment of the present invention. In this embodiment, for example, the method specifically includes:

901. a source detection node receives a measurement instruction sent by a control node, wherein the measurement instruction is sent when the control node receives a measurement request, and the measurement request is used for indicating the control node to feed back a result for identifying the performance of a link from a source virtual client to a destination virtual client;

902. the source detection node sends a detection request message to a target detection node based on the measurement instruction;

903. the source detection node receives a detection response message sent by the destination detection node, and generates a result according to the detection request message and the detection response message, wherein the result is used for identifying the performance of a link from the source detection node to the destination detection node, and the detection response message is sent by the destination detection node corresponding to the detection request message;

904. the source detection node sends the result to the control node so that the control node can forward the result;

Optionally, the 902 specifically includes:

Optionally, the source probe node is a process in the source virtual client, and the destination probe node is a process in the destination virtual client.

Optionally, the source probe node is a process in a first virtual machine, and the first virtual machine and the source virtual client are two virtual machines located on the physical server; the destination probe node is a process in a second virtual machine, the second virtual machine and the destination virtual client being two virtual machines located on the destination physical server.

Optionally, the method further includes:

the source probe node determining whether the result belongs to a range;

Alternatively to this, the first and second parts may,

the 901 specifically includes: the source detection node receives the measurement instruction forwarded by a source proxy node, wherein the measurement instruction is sent to the source proxy node by the control node;

the 904 specifically comprises: and the source detection node sends the result to the source proxy node so that the source proxy node sends the result to the control node.

Alternatively to this, the first and second parts may,

Optionally, the method further includes:

It should be noted that the source node in the embodiment of the present invention is equivalent to the source node 202 in the embodiment shown in fig. 2. For various specific implementations of the method executed by the source probe node in this embodiment, reference may be made to the detailed description about the source probe node 202 in the embodiment shown in fig. 2, and details are not described here again.

In the technical solution of this embodiment, a probe node is deployed for a virtual client in a network system, where a port of the probe node and a port of the virtual client are both connected to the same virtual switch. Therefore, the path for transmitting the packet between the probing nodes and the path for transmitting the packet between the virtual clients are closer to each other than the path for transmitting the packet between the physical routers, and thus, the measurement result of the network performance obtained by transmitting the packet between the probing nodes is closer to the network performance between the virtual clients. In addition, the measurement instruction received by the source detection node is issued by the control node according to the user requirement indicated by the measurement request, and thus, the user can flexibly designate the virtual client machines needing to measure the network performance according to the own requirement through the measurement request, and therefore, the data center can flexibly measure and feed back the network performance between the virtual client machines according to the user requirement.

Fig. 10 is a flowchart illustrating a method for measuring network performance according to an embodiment of the present invention. In this embodiment, for example, the method may specifically include:

1001. a target detection node receives a measurement instruction sent by a control node, wherein the measurement instruction is sent when the control node receives a measurement request, and the measurement request is used for indicating the control node to feed back a result for identifying the performance of a link from a source virtual client to a target virtual client;

1002. the target detection node receives a detection request message sent by a source detection node;

1003. the target detection node sends a detection response message corresponding to the detection request message to the source detection node based on the measurement instruction, the detection request message and the detection response message are used by the source detection node to generate a first result and sent to the control node, the first result is used for identifying the performance of a link between the source detection node and the target detection node, and the first result is used by the control node to forward;

Alternatively to this, the first and second parts may,

Optionally, the method further includes:

Optionally, the 1001 is specifically: the destination detection node receives the measurement instruction forwarded by a destination agent node, and the measurement instruction is sent to the destination agent node by the control node.

Optionally, the destination probe node sends the second result to the control node, specifically: and the destination detection node sends the result to the destination agent node so that the destination agent node can forward the second result to the control node.

Optionally, the sending, by the destination probe node, the alarm to the control node specifically includes: and the destination detection node sends the alarm to the destination agent node so that the destination agent node forwards the alarm to the control node.

Optionally, the method further includes:

It should be noted that the destination probe node in the embodiment of the present invention is equivalent to the destination probe node 203 in the embodiment shown in fig. 2. For various specific embodiments of the method executed by the destination probe node in this embodiment, reference may be made to the detailed description about the destination probe node 203 in the embodiment shown in fig. 2, and details are not described here again.

In the technical solution of this embodiment, a probe node is deployed for a virtual client in a network system, where a port of the probe node and a port of the virtual client are both connected to the same virtual switch. Therefore, the path for transmitting the message between the detection nodes and the path for transmitting the message between the virtual clients are closer to each other than the path for transmitting the message between the physical routers, and therefore, the measurement result of the network performance obtained by transmitting the message between the detection nodes is closer to the real network performance between the virtual clients. In addition, the measurement instruction received by the target detection node is issued by the control node according to the user requirement indicated by the user measurement request, and thus, the user can flexibly designate the virtual client machines needing to measure the network performance according to the own requirement through the measurement request, and therefore, the data center can flexibly measure and feed back the network performance between the virtual client machines according to the user requirement.

Fig. 11 is a flowchart illustrating a method for measuring network performance according to an embodiment of the present invention. In this embodiment, for example, the method may specifically include:

1101. a source proxy node receives a measurement instruction sent by a control node, wherein the measurement instruction is sent when the control node receives a measurement request, and the measurement request is used for indicating the control node to feed back a result for identifying the performance of a link from a source virtual client to a destination virtual client;

1102. the source proxy node forwards the measurement instruction to a source detection node to instruct the source detection node to send a detection request message to a destination detection node, receive a detection response message sent by the destination detection node and generate a result according to the detection request message and the detection response message, wherein the result is used for identifying the performance of a link from the source detection node to the destination detection node, and the detection response message is sent by the destination detection node corresponding to the detection request message;

1103. the source proxy node receives the result sent by the source detection node;

1104. the source proxy node forwards the result to the control node so that the control node forwards the result;

Optionally, the method further includes:

Optionally, before 1102, the method further includes:

Optionally, the source proxy node is a process in a source physical server;

the source proxy node receives control information sent by the control node;

It should be noted that the source proxy node in the embodiment of the present invention is equivalent to the source proxy node 301 in the embodiment shown in fig. 2. For various specific implementations of the method executed by the source proxy node in this embodiment, reference may be made to the detailed description of the embodiment shown in fig. 2, which is not described herein again.

In the technical scheme of this embodiment, the measurement instruction received by the source proxy node is issued by the control node according to the user requirement indicated by the measurement request, and the source proxy node needs to send the measurement instruction to the source detection node meeting the user requirement according to the indication of the measurement instruction. In addition, for the network performance parameters measured by the source detection node, the source proxy node can return to the control node after sorting, statistics and analysis. The information interaction between the control node and the source detection node is forwarded through the source agent node, and the source agent node can ensure the safety of the information interaction in a signature mode. The source proxy node can dynamically create a source detection node in the source physical server according to the control of the control node, so that the source detection node does not occupy the resource of the source physical server under the condition of no measurement requirement.

Fig. 12 is a flowchart illustrating a method for measuring network performance according to an embodiment of the present invention. In this embodiment, for example, the method may specifically include:

1201. a target agent node receives a measurement instruction sent by a control node, wherein the measurement instruction is sent when the control node receives a measurement request, and the measurement request is used for indicating the control node to feed back a result for identifying the performance of a link from a source virtual client to a target virtual client;

1202. the destination proxy node forwards the measurement instruction to a destination detection node to instruct the destination detection node to receive a detection request message sent by a source detection node and send a detection response message corresponding to the detection request message to the source detection node, where the detection request message and the detection response message are used by the source detection node to generate a first result and send the first result to the control node, the first result is used for identifying the performance of a link from the source detection node to the destination detection node, and the first result is used by the control node to forward;

Optionally, the method further includes:

Optionally, before the step 1202, the method further includes:

prior to the 1201, the method further comprises:

It should be noted that the destination agent node in the embodiment of the present invention is equivalent to the destination agent node 302 in the embodiment shown in fig. 2. For various specific embodiments of the method executed by the destination proxy node in this embodiment, reference may be made to the detailed description of the embodiment shown in fig. 2, which is not described herein again.

In the technical scheme of this embodiment, the measurement instruction received by the destination agent node is issued by the control node according to the user requirement indicated by the user measurement request, and the destination agent node needs to send the measurement instruction to the destination probe node meeting the user requirement according to the indication of the measurement instruction. In addition, for the network performance parameters measured by the target detection node, the target proxy node can return to the control node after sorting, statistics and analysis. The information interaction between the control node and the target detection node is forwarded through the target agent node, and the target agent node can ensure the safety of the information interaction in a signature mode. The target proxy node can dynamically create a target detection node in the target physical server according to the control node, so that the target detection node does not occupy the resource of the target physical server under the condition of no measurement requirement.

Fig. 13 is a schematic structural diagram of an apparatus for measuring network performance according to an embodiment of the present invention. In this embodiment, the apparatus 1300 is deployed with a control node. The apparatus 1300 may specifically include, for example:

a first receiving unit 1301, configured to receive a measurement request, where the measurement request is used to instruct the control node to feed back a result used for identifying performance of a link from a source virtual client to a destination virtual client;

a first sending unit 1302, configured to send a first measurement instruction to a source probe node and send a second measurement instruction to a destination probe node, where the first measurement instruction is used to instruct the source probe node to send a probe request packet to the destination probe node, receive a probe response packet corresponding to the probe request packet, and generate a first result according to the probe request packet and the probe response packet, the first result is used to identify performance of a link from the source probe node to the destination probe node, and the second measurement instruction is used to instruct the destination probe node to send the probe response packet corresponding to the probe request packet to the source probe node;

a second receiving unit 1303, configured to receive the first result sent by the source probe node;

a first forwarding unit 1304 for forwarding the first result;

For example, the apparatus 1300 shown in fig. 13 may be used to perform the method shown in fig. 8. Specifically, the first receiving unit 1301 may be configured to perform 801 in fig. 8. The first sending unit 1302 may be configured to perform 802 in fig. 8. The second receiving unit 1303 and the first feedback unit 1304 may be used to perform 803 in fig. 8.

Optionally, the apparatus 1300 further includes:

and the second forwarding unit is used for forwarding the second result.

Optionally, the apparatus 1300 further includes:

a third forwarding unit, configured to forward the first alarm;

a fifth receiving unit, configured to receive a second alarm sent by the destination probe node, where the second alarm is generated and sent by the destination probe node when it is determined that the second alarm does not belong to a second range;

and the fourth forwarding unit is used for forwarding the second alarm.

Optionally, the apparatus 1300 further includes:

Alternatively to this, the first and second parts may,

the first sending unit 1302 is specifically configured to send the first measurement instruction to a source proxy node to instruct the source proxy node to forward the first measurement instruction to the source probe node, and send the second measurement instruction to a destination proxy node to instruct the destination proxy node to forward the second measurement instruction to the destination probe node;

the second receiving unit 1303 is specifically configured to receive the first result forwarded by the source proxy node, where the first result is sent to the source proxy node by the source probe node.

Optionally, the third receiving unit is specifically configured to receive the second result forwarded by the destination proxy node, where the second result is sent to the destination proxy node by the destination probe node.

Alternatively to this, the first and second parts may,

the apparatus 1300 further comprises:

It should be noted that the apparatus 1300 in the embodiment of the present invention is deployed with the control node 201 in the embodiment shown in fig. 2. For various specific implementations of the apparatus 1400 in this embodiment, reference may be made to the detailed description of the control node 201 in the embodiment shown in fig. 2, which is not described herein again.

Through the apparatus 1300 of this embodiment, a user may flexibly specify a virtual client that needs to measure network performance according to a own requirement through a measurement request, when the first receiving unit 1301 receives the measurement request, the first sending unit 1302 may flexibly issue a measurement instruction to a corresponding probe node according to a user requirement indicated by the measurement request, the second receiving unit 1303 may receive a returned network performance parameter as a measurement result, and the first forwarding unit 1304 may feed back the measurement result to the user, so that the data center may flexibly measure and feed back network performance between virtual clients according to the user requirement. In addition, the user may also flexibly indicate other requirements such as the type of network performance parameters through the measurement request, and the apparatus 1300 may flexibly obtain the measurement result meeting various requirements of the user according to the indication of the measurement request and feed back the measurement result to the user. Therefore, the data center can flexibly meet various requirements of users on network performance measurement.

Fig. 14 is a schematic structural diagram of an apparatus for measuring network performance according to an embodiment of the present invention. In this embodiment, the apparatus 1400 is deployed with an active probing node. The apparatus 1400 may specifically include, for example:

a first receiving unit 1401, configured to receive a measurement instruction sent by a control node, where the measurement instruction is sent when the control node receives a measurement request, where the measurement request is used to instruct the control node to feed back a result used for identifying performance of a link from a source virtual client to a destination virtual client;

a first sending unit 1402, configured to send a probe request packet to a target probe node based on the measurement instruction;

a second receiving unit 1403, configured to receive the probe response message sent by the destination probe node;

a first generating unit 1404, configured to generate a result according to the probe request packet and the probe response packet, where the result is used to identify performance of a link from the source probe node to a destination probe node, and the probe response packet is sent by the destination probe node in response to the probe request packet;

a second sending unit 1405, configured to send the result to the control node, so that the control node forwards the result;

For example, the apparatus 1400 shown in fig. 14 may be used to perform the method shown in fig. 9. Specifically, the first receiving unit 1401 may be used to execute 901 in fig. 9. The first sending unit 1402 may be configured to perform 902 in fig. 9. The second receiving unit 1403 and the first generating unit 1404 may be used to perform 903 in fig. 9. The second transmission unit 1405 may be used to perform 904 in fig. 9.

Optionally, the first sending unit 1402 is specifically configured to:

Alternatively to this, the first and second parts may,

Optionally, the apparatus 1400 further includes:

a determination unit for determining whether the result belongs to a range;

Alternatively to this, the first and second parts may,

the second receiving unit 1403 is specifically configured to receive the measurement instruction forwarded by the source proxy node, where the measurement instruction is sent to the source proxy node by the control node;

the second sending unit 1405 is specifically configured to send the result to the source proxy node, so that the source proxy node forwards the result to the control node.

Optionally, the third sending unit is specifically configured to send the alarm to the source proxy node, so that the source proxy node forwards the alarm to the control node.

Optionally, the apparatus 1400 further includes:

a triggering unit, configured to trigger the first sending unit 1402 if the verification passes.

It should be noted that the apparatus 1400 in the embodiment of the present invention is deployed with the source probe node 202 in the embodiment shown in fig. 2. For various specific embodiments of the apparatus 1400 in this embodiment, reference may be made to the detailed description of the source probe node 202 in the embodiment shown in fig. 2, which is not described herein again.

In the apparatus 1400 of this embodiment, a probe node is deployed for a virtual client in a network system, wherein a port of the probe node and a port of the virtual client are both connected to the same virtual switch. Therefore, the path for transmitting the packet between the probing nodes and the path for transmitting the packet between the virtual clients are closer to each other than the path for transmitting the packet between the physical routers, and thus, the measurement result of the network performance obtained by transmitting the packet between the probing nodes is closer to the network performance between the virtual clients. In addition, the measurement instruction received by the first receiving unit 1401 is issued by the control node according to the user requirement indicated by the measurement request, and thus, the user can flexibly specify the virtual client machines which need to measure the network performance according to the own requirement through the measurement request, and therefore, the data center can flexibly measure and feed back the network performance between the virtual client machines according to the user requirement.

Fig. 15 is a schematic structural diagram of an apparatus for measuring network performance according to an embodiment of the present invention. In this embodiment, the apparatus 1500 deploys purposeful probing nodes. The apparatus 1500 may specifically include, for example:

a first receiving unit 1501, configured to receive a measurement instruction sent by a control node, where the measurement instruction is sent when the control node receives a measurement request, and the measurement request is used to instruct the control node to feed back a result used for identifying the performance of a link from a source virtual client to a destination virtual client;

a second receiving unit 1502, configured to receive a probe request packet sent by a source probe node;

a returning unit 1503, configured to send, to the source probe node, a probe response packet corresponding to the probe request packet based on the measurement instruction, where the probe request packet and the probe response packet are used by the source probe node to generate a first result and send the first result to the control node, where the first result is used to identify performance of a link between the source probe node and the destination probe node, and the first result is used by the control node to forward the first result;

For example, the apparatus 1500 shown in FIG. 15 may be used to perform the method shown in FIG. 10. Specifically, the first receiving unit 1501 may be used to perform 1001 in fig. 10. The second receiving unit 1402 may be configured to perform 1002 in fig. 10. The return unit 1503 may be used to execute 1003 in fig. 10.

Alternatively to this, the first and second parts may,

Optionally, the apparatus 1500 further includes:

Optionally, the first receiving unit 1501 is specifically configured to receive the measurement instruction forwarded by the destination agent node, where the measurement instruction is sent to the destination agent node by the control node.

Optionally, the first sending unit is specifically configured to send the second result to the destination proxy node, so that the destination proxy node forwards the second result to the control node.

Optionally, the second sending unit is specifically configured to send the alarm to the destination agent node, so that the destination agent node forwards the alarm to the control node.

Optionally, the apparatus 1500 further includes:

It should be noted that the apparatus 1500 in the embodiment of the present invention is deployed with the destination probe node 203 in the foregoing embodiment shown in fig. 2. For various specific implementations of the apparatus 1500 in this embodiment, reference may be made to the detailed description of the target probing node 203 in the embodiment shown in fig. 2, which is not described herein again.

In the apparatus 1500 of this embodiment, a probe node is deployed for a virtual client in a network system, where a port of the probe node and a port of the virtual client are both connected to the same virtual switch. Therefore, the path for transmitting the message between the detection nodes and the path for transmitting the message between the virtual clients are closer to each other than the path for transmitting the message between the physical routers, and therefore, the measurement result of the network performance obtained by transmitting the message between the detection nodes is closer to the real network performance between the virtual clients. In addition, the measurement instruction received by the first receiving unit 1501 is issued by the control node according to the user requirement indicated by the user measurement request, and thus, the user can flexibly specify the virtual client machines needing to measure the network performance according to the own requirement through the measurement request, and therefore, the data center can flexibly measure and feed back the network performance between the virtual client machines according to the user requirement.

Fig. 16 is a schematic structural diagram of an apparatus for measuring network performance according to an embodiment of the present invention. In this embodiment, the apparatus 1600 is configured with an active proxy node, and the apparatus 1600 may specifically include:

a first receiving unit 1601, configured to receive a measurement instruction sent by a control node, where the measurement instruction is sent when the control node receives a measurement request, where the measurement request is used to instruct the control node to feed back a result for identifying performance of a link from a source virtual client to a destination virtual client;

a first forwarding unit 1602, configured to forward the measurement instruction to the source probe node to instruct the source probe node to send a probe request packet to a destination probe node, receive a probe response packet sent by the destination probe node, and generate a result according to the probe request packet and the probe response packet, where the result is used to identify performance of a link from the source probe node to the destination probe node, and the probe response packet is sent by the destination probe node corresponding to the probe request packet;

a second receiving unit 1603, which receives the result sent by the source probe node;

a second forwarding unit 1604, configured to forward the result to the control node, so that the control node forwards the result;

For example, the device 1600 shown in FIG. 16 may be used to perform the method shown in FIG. 11. Specifically, the first receiving unit 1601 may be used to execute 1101 in fig. 11. The first forwarding unit 1602 may be used to perform 1102 in fig. 11. The second receiving unit 1603 may be used to perform 1103 in fig. 11. Second forwarding unit 1604 may be used to perform 1104 in fig. 11.

Optionally, the apparatus 1600 further includes:

Optionally, the source proxy node is a process in a source physical server;

the apparatus 1600 further comprises:

It should be noted that the apparatus 1600 in the embodiment of the present invention is deployed with the source proxy node 301 in the embodiment shown in fig. 2. For various specific embodiments of the apparatus 1600 in this embodiment, reference may be made to the detailed description of the source proxy node 301 in the embodiment shown in fig. 2, which is not described herein again.

In the apparatus 1600 of this embodiment, the measurement instruction received by the first receiving unit 1601 is issued by the control node according to the user requirement indicated by the measurement request, and the first sending unit 1602 needs to send the measurement instruction to the source probe node meeting the user requirement according to the indication of the measurement instruction, and as a result, the user can flexibly specify the virtual client that needs to measure the network performance according to the own requirement through the measurement request, and therefore, the data center can flexibly measure and feed back the network performance between the virtual clients according to the user requirement. In addition, for the network performance parameters measured by the source probe node, the apparatus 1600 may return to the control node via the second sending unit 1604 after sorting, counting and analyzing. The information interaction between the control node and the source detection node is forwarded through the source agent node, and the source agent node can ensure the safety of the information interaction in a signature mode. The apparatus 1600 may dynamically create a source probe node in the source physical server under the control of the control node, such that the source probe node does not occupy the resources of the source physical server without measurement requirements.

Fig. 17 is a schematic structural diagram of an apparatus for measuring network performance according to an embodiment of the present invention. In this embodiment, the apparatus 1700 deploys a destination agent node, and the apparatus 1700 may specifically include:

a first receiving unit 1701, configured to receive a measurement instruction sent by a control node, where the measurement instruction is sent when the control node receives a measurement request, and the measurement request is used to instruct the control node to feed back a result for identifying performance of a link from a source virtual client to a destination virtual client;

a first forwarding unit 1702, configured to forward the measurement instruction to a destination probe node, so as to instruct the destination probe node to receive a probe request packet sent by a source probe node and send a probe response packet corresponding to the probe request packet to the source probe node, where the probe request packet and the probe response packet are used by the source probe node to generate a first result and send the first result to the control node, the first result is used to identify performance of a link from the source probe node to the destination probe node, and the first result is used by the control node to forward the first result;

For example, the apparatus 1700 shown in fig. 17 may be used to perform the method shown in fig. 12. Specifically, the first receiving unit 1701 may be used to perform 1201 in fig. 12. The first forwarding unit 1702 may be used to perform 1202 in fig. 12.

Optionally, the apparatus 1700 further includes:

the apparatus 1700 further comprises:

It should be noted that the apparatus 1700 in the embodiment of the present invention is deployed with the destination agent node 302 in the embodiment shown in fig. 2. For various specific embodiments of the apparatus 1700 in this embodiment, reference may be made to the foregoing detailed description of the destination proxy node 302 in the embodiment shown in fig. 2, which is not described herein again.

In the apparatus 1700 of this embodiment, the measurement instruction received by the first receiving unit 1701 is issued by the control node according to the user requirement indicated by the measurement request, and the first forwarding unit 1702 needs to send the measurement instruction to the target probe node meeting the user requirement according to the indication of the measurement instruction, and thus, the user can flexibly specify the virtual client needing to measure the network performance according to the own requirement through the measurement request, and therefore, the data center can flexibly measure and feed back the network performance between the virtual clients according to the user requirement. In addition, the device 1700 may return the measurement result of the network performance of the target probe node to the control node after sorting, statistics and analysis. The information interaction between the control node and the target detection node is forwarded through the device 1700, and the signature unit can ensure the safety of the information interaction in a signature mode. The apparatus 1700 may dynamically create a destination probe node in the destination physical server according to the control node, so that the destination probe node does not occupy the resource of the destination physical server without measurement requirement.

Fig. 18 is a schematic structural diagram of a physical server in the embodiment of the present invention. In this embodiment, the physical server 1800 is deployed with the control node 201 in the embodiment shown in fig. 2, and may be used to execute the method in the embodiment shown in fig. 8. The physical server 1800 includes: a processor 1801, a memory 1802, a network interface 1803, and a bus system 1804.

The bus system 1804 is used to couple together the various hardware components of the physical server 1800.

The network interface 1803 is used to implement communication connection between the physical server 1800 and at least one other network element, and may use internet, wide area network, local network, metropolitan area network, or the like.

The memory 1802 is used to store program instructions and data.

The processor 1801 is configured to read instructions and data stored in the memory 1802, and perform the following operations:

receiving a measurement request for instructing the control node to feed back a result for identifying a performance of a link of a source virtual client to a destination virtual client;

sending a first measurement instruction to a source detection node and sending a second measurement instruction to a destination detection node, wherein the first measurement instruction is used for instructing the source detection node to send a detection request message to the destination detection node, receiving a detection response message corresponding to the detection request message and generating a first result according to the detection request message and the detection response message, the first result is used for identifying the performance of a link from the source detection node to the destination detection node, and the second measurement instruction is used for instructing the destination detection node to send the detection response message corresponding to the detection request message to the source detection node;

receiving the first result sent by the source detection node, and forwarding the first result;

Optionally, the processor 1801 may further perform the following operations:

receiving the second result sent by the target detection node, and forwarding the second result;

Optionally, the processor 1801 may further perform the following operations:

receiving a first alarm sent by the source detection node, and forwarding the first alarm;

receiving a second alarm sent by the target detection node, and forwarding the second alarm;

Optionally, the processor 1801 may further perform the following operations:

determining whether the first result belongs to a third range;

when the first result does not belong to the third range, generating a third alarm and forwarding;

determining whether the second result falls within a fourth range;

and when the second result does not belong to the fourth range, generating a fourth alarm and forwarding.

Optionally, in order to send the first measurement instruction and the second measurement instruction, the processor 1801 may perform the following operations:

sending the first measurement instruction to a source proxy node to instruct the source proxy node to forward the first measurement instruction to the source probe node;

and sending the second measurement instruction to a destination proxy node to instruct the destination proxy node to forward the second measurement instruction to the destination detection node.

Optionally, to receive the first result, the processor 1801 may perform the following operations:

receiving the first result forwarded by the source proxy node, the first result being sent by the source probe node to the source proxy node.

Optionally, to receive the second result, the processor 1801 may perform the following operations:

receiving the second result forwarded by the destination proxy node, the second result being sent to the destination proxy node by the destination probe node.

Optionally, to receive the first alarm, the processor 1801 may perform the following operations:

receiving the first alarm forwarded by the source proxy node, wherein the first alarm is sent to the source proxy node by the source probe node;

optionally, to receive the second alarm, the processor 1801 may perform the following operations:

receiving the second alarm forwarded by the destination agent node, wherein the second alarm is sent to the destination agent node by the destination probe node.

Optionally, the source proxy node and the destination proxy node may be the same proxy node;

before sending the first measurement instruction and the second measurement instruction, the processor 1801 may further perform the following operations:

acquiring a first route from the source agent node to the source detection node and a second route from the source detection node to the source agent node, sending the first route to the source agent node, and sending the second route to the source detection node, wherein the first route is used for the source agent node to forward the first measurement instruction to the source detection node, and the second route is used for the source detection node to send the first result to the source agent node;

and acquiring a third route from the destination agent node to the destination detection node, and sending the third route to the destination agent node, wherein the third route is used for forwarding the second measurement instruction to the destination detection node by the destination agent node.

Optionally, the source proxy node may be a process in a source physical server, and the destination proxy node may be a process in a destination physical server;

before sending the first measurement instruction and the second measurement instruction, the processor 1801 may perform the following operations:

sending first control information to the source proxy node and second control information to the destination proxy node, the first control information being used for instructing the source proxy node to create a first virtual machine and the source probe node in the source physical server, the second control information being used for instructing the destination proxy node to create a second virtual machine and the destination probe node in the destination physical server, wherein the source probe node is one process in the first virtual machine, the first virtual machine and the source virtual client are two virtual machines located in the source physical server, the destination probe node is one process in the second virtual machine, and the second virtual machine and the destination virtual client are two virtual machines located in the destination physical server;

assigning a first port on the source virtual switch to the source probing node and a second port on the destination virtual switch to the destination probing node, the source probing node being capable of communicating with the destination probing node through the first port, the destination probing node being capable of communicating with the source proxy node through the second port;

establishing a virtual local area network comprising the first port and the second port; the virtual local area network is used for transmitting the detection request message and the detection response message.

Fig. 19 is a schematic structural diagram of a physical server according to an embodiment of the present invention. In this embodiment, the physical server 1900 is deployed with the source probe node 202 in the embodiment shown in fig. 2, and may be used to execute the method in the embodiment shown in fig. 9. The physical server 1900 includes: a processor 1901, a memory 1902, a network interface 1903, and a bus system 1904.

The bus system 1904 is used to couple the various hardware components of the physical server 1900 together.

The network interface 1903 is used to implement communication connection between the physical server 1900 and at least one other network element, and may use the internet, a wide area network, a local area network, a metropolitan area network, or the like.

The memory 1902 is used to store program instructions and data.

The processor 1901 is configured to read instructions and data stored in the memory 1902, and perform the following operations:

receiving a measurement instruction sent by a control node, wherein the measurement instruction is sent when the control node receives a measurement request, and the measurement request is used for indicating the control node to feed back a result for identifying the performance of a link from a source virtual client to a destination virtual client;

sending a detection request message to a target detection node based on the measurement instruction;

receiving a detection response message sent by the destination detection node, and generating a result according to the detection request message and the detection response message, wherein the result is used for identifying the performance of a link from the source detection node to the destination detection node, and the detection response message is sent by the destination detection node corresponding to the detection request message;

sending the result to the control node so that the control node can forward the result;

Optionally, to send the probe request packet, the processor 1901 may perform the following operations:

Optionally, the processor 1901 may further perform the following operations:

determining whether the result falls within a range;

when the result does not belong to the range, generating an alarm;

and sending the alarm to the control node so that the control node can forward the alarm.

Optionally, to receive the measurement instruction, the processor 1901 may perform the following operations:

receiving the measurement instruction forwarded by a source proxy node, the measurement instruction being sent to the source proxy node by the control node.

Optionally, to send the result, the processor 1901 may perform the following operations:

and sending the result to the source proxy node so that the source proxy node sends the result to the control node.

Optionally, to send the alert, the processor 1901 may perform the following operations:

and sending the alarm to the source proxy node so that the source proxy node forwards the alarm to the control node.

Optionally, the processor 1901 may further perform the following operations:

verifying the measurement instruction, and sending the detection request message to the target detection node under the condition that the verification is passed;

Fig. 20 is a schematic structural diagram of a physical server according to an embodiment of the present invention. In this embodiment, the physical server 2000 is deployed with the destination probe node 203 in the embodiment shown in fig. 2, and may be configured to execute the method in the embodiment shown in fig. 10. The physical server 2000 includes: a processor 2001, a memory 2002, a network interface 2003, a bus system 2004.

The bus system 2004 is used to couple the various hardware components of the physical server 2000 together.

The network interface 2003 is used to implement communication connection between the physical server 2000 and at least one other network element, and may use the internet, a wide area network, a local area network, a metropolitan area network, or the like.

The memory 2002 is used to store program instructions and data.

The processor 2001 is configured to read instructions and data stored in the memory 2002, and perform the following operations:

receiving a detection request message sent by a source detection node;

sending a probe response message corresponding to the probe request message to the source probe node based on the measurement instruction, where the probe request message and the probe response message are used by the source probe node to generate a first result and send the first result to the control node, where the first result is used to identify performance of a link between the source probe node and the destination probe node, and the first result is used by the control node to forward the first result;

Optionally, the source probe node is a process in a first virtual machine, and the first virtual machine and the source virtual client are two virtual machines located in the source physical server; the destination probe node is a process in a second virtual machine, and the second virtual machine and the destination virtual machine are two virtual machines located in the destination physical server.

Optionally, the processor 2001 may further perform the following operations:

generating a second result according to the detection request message, wherein the second result is used for identifying the performance of a link from the source detection node to the destination detection node;

and sending the second result to the control node so that the control node can forward the second result.

Optionally, the processor 2001 may further perform the following operations:

determining whether the second result falls within a range;

when the second result does not belong to the range, generating an alarm;

Optionally, to receive the measurement instruction, the processor 2001 may perform the following operations:

receiving the measurement instruction forwarded by a destination agent node, wherein the measurement instruction is sent to the destination agent node by the control node.

Optionally, to send the second result, the processor 2001 may perform the following operations:

and sending the result to the destination agent node so that the destination agent node can forward the second result to the control node.

Optionally, to send the alarm, the processor 2001 may perform the following operations:

and sending the alarm to the target agent node so that the target agent node can forward the alarm to the control node.

Optionally, the processor 2001 may further perform the following operations:

verifying the measurement instruction, and receiving the detection request message sent by the source detection node under the condition that the verification is passed;

Fig. 21 is a schematic structural diagram of a physical server in the embodiment of the present invention. In this embodiment, the physical server 2100 is deployed with the source proxy node 301 in the embodiment shown in fig. 2, and may be used to execute the method in the embodiment shown in fig. 11. The physical server 2100 includes: a processor 2101, a memory 2102, a network interface 2103, a bus system 2104.

The bus system 2104 is used to couple the various hardware components of the physical server 2100 together.

The network interface 2103 is used to implement a communication connection between the physical server 2100 and at least one other network element, and may use the internet, a wide area network, a local area network, a metropolitan area network, or the like.

The memory 2102 is used to store program instructions and data.

The processor 2101 is configured to read the instructions and data stored in the memory 2102 and perform the following operations:

forwarding the measurement instruction to a source detection node to instruct the source detection node to send a detection request message to a destination detection node, receive a detection response message sent by the destination detection node, and generate a result according to the detection request message and the detection response message, wherein the result is used for identifying the performance of a link from the source detection node to the destination detection node, and the detection response message is sent by the destination detection node corresponding to the detection request message;

receiving the result sent by the source detection node;

forwarding the result to the control node so that the control node forwards the result;

Optionally, the processor 2100 may further perform the following operations:

receiving an alarm sent by the source detection node, wherein the alarm is generated by the source detection node when the result is determined not to belong to the range;

and forwarding the alarm to the control node so that the control node can forward the alarm.

Optionally, before sending the measurement instruction, the processor 2100 may further perform the following operations:

signing the measurement instruction; the signed measurement instruction is used for verification by the source detection node and sends the detection request message to the target detection node under the condition that the verification is passed.

Optionally, the source proxy node is a process in a source physical server;

prior to receiving the measurement instruction, the processor 2100 may further perform the following:

receiving control information sent by the control node;

creating a virtual machine and the source probe node in the source physical server based on the control information, wherein the source probe node is a process in the first virtual machine, and the virtual machine and the source virtual client are two virtual machines located in the source physical server.

Fig. 22 is a schematic structural diagram of a physical server in the embodiment of the present invention. In this embodiment, the physical server 2200 is deployed with the destination proxy node 302 in the embodiment shown in fig. 2, and may be used to execute the method in the embodiment shown in fig. 12. The physical server 2200 includes: a processor 2201, a memory 2202, a network interface 2203, and a bus system 2204.

The bus system 2204 is used to couple the various hardware components of the physical server 2200 together.

The network interface 2203 is configured to implement communication connection between the physical server 2200 and at least one other network element, and may use the internet, a wide area network, a local network, a metropolitan area network, or the like.

The memory 2202 is used to store program instructions and data.

The processor 2201 is configured to read instructions and data stored in the memory 2202, and perform the following operations:

forwarding the measurement instruction to a destination detection node to instruct the destination detection node to receive a detection request message sent by a source detection node and send a detection response message corresponding to the detection request message to the source detection node, where the detection request message and the detection response message are used by the source detection node to generate a first result and are sent to the control node, the first result is used to identify performance of a link from the source detection node to the destination detection node, and the first result is used by the control node to forward;

Optionally, the processor 2201 may further perform the following operations:

receiving a second result sent by the destination probe node, where the second result is generated by the destination probe node according to the probe request packet, and the second result is used to identify the performance of a link from the source probe node to the destination probe node;

and forwarding the second result to the control node so that the control node forwards the second result.

Optionally, the processor 2201 may further perform the following operations:

receiving an alarm sent by the target detection node, wherein the alarm is generated by the target detection node when the second result is determined not to belong to the range;

Optionally, before sending the measurement instruction, the processor 2201 may further perform the following operations:

before receiving the measurement instruction, the processor 2201 may further perform the following operations:

receiving control information sent by the control node;

creating a virtual machine and the destination probe node in the destination physical server based on the control information, wherein the source probe node is a process in the first virtual machine, and the virtual machine and the source virtual client are two virtual machines located in the source physical server.

In the embodiments of the present invention, the "first" in the names of the "first measurement instruction", the "first network performance parameter", and the like is used merely as a name, and does not represent the first in sequence. The same applies to "second", "third", "fourth" and "fifth", etc.

It should be noted that the processor in the embodiment of the present invention may be an integrated circuit chip having signal processing capability. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware in a processor or instructions in the form of software. The instructions may be implemented and controlled by a processor to perform the methods disclosed by the embodiments of the invention. The processor may also be a general purpose processor, a Digital Signal Processor (DSP), an application specific integrated circuit (application specific integrated circuit), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic device, or discrete hardware components.

The general purpose processor may be a microprocessor or the processor may be any conventional processor, decoder, etc. The steps of a method disclosed in connection with the embodiments of the present invention may be directly implemented by a hardware processor, or may be implemented by a combination of hardware and software modules in the processor. The software module may be located in ram, flash memory, rom, prom, or eprom, registers, etc. storage media as is well known in the art.

In addition, it should be noted that the bus system may include a power bus, a control bus, and a status signal bus in addition to the data bus. For clarity of illustration, however, the various buses are labeled as bus systems in fig. 8 and 9.

As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that all or part of the steps in the above embodiment methods can be implemented by software plus a general hardware platform. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which may be stored in a storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network communication device such as a media gateway, etc.) to execute the method according to the embodiments or some parts of the embodiments.

It should be noted that, in the present specification, all the embodiments are described in a progressive manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, the method embodiments and apparatus embodiments are substantially similar to the system embodiments and therefore are described in a relatively simple manner, and reference may be made to some of the descriptions of the system embodiments for related points. The above-described embodiments of the apparatus and system are merely illustrative, wherein modules described as separate parts may or may not be physically separate, and parts shown as modules may or may not be physical modules, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.

The above description is only for the preferred embodiment of the present invention and is not intended to limit the scope of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principle of the present invention, and such improvements and modifications should be considered as the protection scope of the present invention.

Claims

1. A system for measuring network performance, the system comprising:

2. The system according to claim 1, wherein the measurement request, the first measurement instruction, and the second measurement instruction each carry an identifier of the source virtual client, an identifier of the destination virtual client, and a type of the indicator to which the result for identifying the performance of the link from the source virtual client to the destination virtual client belongs.

3. The system according to claim 2, wherein said sending a probe request packet to the destination probe node based on the first measurement instruction comprises:

4. The system of claim 1,

5. The system of claim 1,

6. The system of claim 1,

7. The system of claim 1,

8. The system of claim 7,

9. The system of any one of claims 1 to 8, further comprising:

10. The system of claim 7 or 8, further comprising:

and the destination proxy node is used for receiving the second result sent by the destination detection node and forwarding the second result to the control node.

11. The system of claim 6, further comprising:

and the source proxy node is used for receiving the first alarm sent by the source detection node and forwarding the first alarm to the control node.

12. The system of claim 8, further comprising:

and the destination proxy node is used for receiving the second alarm sent by the destination detection node and forwarding the second alarm to the control node.

13. The system of claim 9,

14. The system of claim 9, wherein the source proxy node and the destination proxy node are the same proxy node;

15. The system of claim 9, wherein the source proxy node is a process in the source physical server and the destination proxy node is a process in the destination physical server.

16. The system of claim 15,

the source proxy node is further configured to receive the first control information, and create a first virtual machine in the source physical server and create the source probe node in the first virtual machine based on the first control information;

the destination agent node is further configured to receive the second control information, and create a second virtual machine in the destination physical server and create the destination probe node in the second virtual machine based on the second control information.

17. A method of measuring network performance, comprising:

18. The method according to claim 17, wherein the measurement request, the first measurement instruction, and the second measurement instruction each carry an identifier of the source virtual client, an identifier of the destination virtual client, and a type of an indicator to which the result for identifying the link from the source virtual client to the destination virtual client belongs.

19. The method of claim 17, further comprising:

the control node receives a second result sent by the target detection node and forwards the second result;

20. The method of claim 19, further comprising:

21. The method of claim 19, further comprising:

the control node determining whether the first result belongs to a third range;

22. The method according to any one of claims 17 to 21,

23. The method of claim 22,

24. The method of claim 23,

25. The method of claim 22, wherein the source proxy node and the destination proxy node are the same proxy node;

26. The method of claim 22, wherein the source proxy node is a process in a source physical server and the destination proxy node is a process in a destination physical server;

27. An apparatus for measuring network performance, the apparatus being deployed with a control node, the apparatus comprising:

a first forwarding unit, configured to forward the first result;

28. The apparatus of claim 27, wherein the measurement request, the first measurement instruction, and the second measurement instruction each carry an identifier of the source virtual client, an identifier of the destination virtual client, and a type of an indicator to which the result for identifying the link from the source virtual client to the destination virtual client belongs.

29. The apparatus of claim 27, further comprising:

a third receiving unit, configured to receive a second result sent by the destination probe node, where the second result is generated by the destination probe node according to the probe request packet, and the second result is used to identify performance of a link from the source probe node to the destination probe node;

and the second forwarding unit is used for forwarding the second result.

30. The apparatus of claim 29, further comprising:

a third forwarding unit, configured to forward the first alarm;

and the fourth forwarding unit is used for forwarding the second alarm.

31. The apparatus of claim 29, further comprising:

32. The apparatus of any one of claims 27 to 31,

33. The apparatus of any one of claims 29 to 31,

34. The apparatus of claim 30,

35. The apparatus of claim 32, wherein the source proxy node and the destination proxy node are the same proxy node;

the device further comprises:

36. The apparatus of claim 32, wherein the source proxy node is a process in a source physical server and the destination proxy node is a process in a destination physical server;

the device further comprises:

the establishing unit is used for establishing a virtual local area network comprising the first port and the second port;

the virtual local area network is used for transmitting the detection request message and the detection response message.