WO2018045821A1

WO2018045821A1 - Methods and devices for sending, receiving, and transmitting cloud host data

Info

Publication number: WO2018045821A1
Application number: PCT/CN2017/092571
Authority: WO
Inventors: 沈娟; 李国超; 崔孝飞; 杨万里
Original assignee: 北京京东尚科信息技术有限公司; 北京京东世纪贸易有限公司
Priority date: 2016-09-06
Filing date: 2017-07-12
Publication date: 2018-03-15
Also published as: CN107800645B; CN107800645A

Abstract

The present invention relates to the field of cloud computing, and provides methods and devices for sending, receiving, and transmitting cloud host data. The method for receiving cloud host data of the present invention comprises: obtaining destination address information of downlink traffic of a cloud host, the destination address information being public network address information of the cloud host; determining a target queue according to the destination address information on the basis of a corresponding relationship between a queue and a public network address; and performing downlink transmission by means of the target queue on the basis of bandwidth limitation of the target queue. By means of the method, a queue can be allocated for downlink traffic of a cloud host according to a destination address, and downlink traffic bandwidth can be limited by using bandwidth limitation of the queue, so that downlink traffic of the cloud host is limited based on a public network address, to avoid that network bandwidth is occupied by an individual user due to traffic of excessive surge, thus improving the stability of the cloud host.

Description

Cloud host data transmitting, receiving, transmitting method and device

Technical field

The present invention relates to the field of cloud computing, and in particular, to a cloud host data transmitting, receiving, and transmitting method and apparatus.

Background technique

At present, in the IaaS (Infrastructure as a Service) in the field of OpenStack cloud computing, the cloud computing service platform provides virtualized hosts to the users through cloud computing virtualization hosts, and needs to be customized according to users. The situation limits the upstream and downstream bandwidth of the host it creates, preventing the influx of excessive traffic and causing the network bandwidth to be occupied by individual users. In addition, economic benefits can be generated by charging a corresponding fee based on the bandwidth requested by the user.

In the prior art, the OpenStack G version supports setting QoS (Quality of Service) through Nova Flavor. Setting the QoS through the Flavor is not very flexible, which requires maintaining a large number of Flavors to ensure different QoS requirements. The OpenStack L version supports the ML2-QoS function. This feature can be used to configure traffic QoS for L2-port and L2-net. ML2-QoS can only limit the traffic of Layer 2 networks. It cannot distinguish between intranet traffic and public network traffic. Making full use of the hardware resources of the internal network, resulting in a great waste of resources.

Summary of the invention

An object of the present invention is to propose a scheme for bandwidth limitation of public network traffic of a cloud host.

According to an aspect of the present invention, a method for receiving a cloud host data includes: obtaining destination address information of a downlink traffic of a cloud host, where the destination address information is public network address information of the cloud host; and based on a correspondence between the queue and the public network address, Determining a target queue according to the destination address information; performing downlink transmission through the target queue based on the bandwidth limitation of the target queue.

Optionally, the destination address information of the downlink traffic of the cloud host is: the intermediate functional block device (IFB) acquires the downlink traffic that enters the network card, and obtains the destination address information; and the correspondence between the queue and the public network address, Determining the target queue according to the destination address information includes: determining, by the IFB, the target queue according to the correspondence between the queue and the public network address, wherein the target queue has bandwidth limitation information; and performing downlink transmission through the target queue according to the bandwidth limitation of the target queue includes: downlink traffic The network card is returned under the bandwidth limitation of the target queue, and the downlink transmission is performed through the network card.

Optionally, the cloud host data receiving method further includes: connecting the network card of the cloud host to the IFB; adding a TC (Traffic Control) rule in the IFB, setting a queue through the TC rule, and specifying bandwidth limit information of each queue. The correspondence between the queue and the public network address is established in the IFB.

Optionally, the cloud host data receiving method further includes: if the target queue is not determined according to the destination address information based on the correspondence between the queue and the public network address, selecting a predetermined default queue as the target queue.

Optionally, the cloud host data receiving method further includes: when deleting the public network address of the cloud host, deleting the target queue that matches the public network address, and deleting the correspondence between the public network address and the queue.

In this way, the queue can be allocated to the downstream traffic of the cloud host according to the destination address, and the bandwidth limitation of the queue can be used to limit the downlink traffic bandwidth, thereby realizing the downlink traffic restriction of the cloud host based on the public network address, thereby preventing excessive flooding. The traffic causes the network bandwidth to be occupied by individual users, which improves the stability of the cloud host.

According to another aspect of the present invention, a cloud host data sending method is provided, including: obtaining source address information of a cloud host uplink traffic, the source address information being public network address information of the cloud host; and performing source address information and flow table information Matching, determining a target queue, where the flow table information includes a correspondence between the public network address information and the queue; and the uplink transmission is performed through the target queue based on the bandwidth limitation of the target queue.

Optionally, obtaining source address information of the uplink traffic of the cloud host includes: acquiring, by the bridge of the cloud host, source address information of the uplink traffic; matching the source address information with the flow table information, and determining that the target queue includes: the source address information of the bridge Matching the flow table information in the bridge to determine the target queue in the QoS function module, wherein the exit of the queue is bound to the uplink port of the bridge, and each queue has bandwidth limitation information; based on the bandwidth limitation of the target queue, The uplink transmission of the target queue includes: calling the quality of service QoS function module, and performing uplink transmission on the uplink port under the bandwidth limitation of the target queue of the QoS function module.

Optionally, the cloud host data sending method further includes: binding a bridge of the cloud host to the QoS function module; generating a queue in the QoS function module, setting a flow control TC classification for each queue, and specifying bandwidth limitation information by using the TC classification; Add flow table information to the bridge.

Optionally, the cloud host data sending method further includes: if the source address information and the flow table information do not match successfully, selecting a predetermined default queue as the target queue.

Optionally, the cloud host data sending method further includes: when deleting the public network address of the cloud host, deleting the target queue in the flow table information matching the public network address, and deleting the flow table information including the public network address.

In this way, the queue can be allocated to the upstream traffic of the cloud host according to the source address, and the bandwidth limitation of the queue can be used to limit the uplink traffic bandwidth, thereby realizing the uplink traffic restriction of the cloud host based on the public network address. Preventing influx of excessive traffic causes network bandwidth to be occupied by individual users, improving the stability of the cloud host.

According to still another aspect of the present invention, a cloud host data transmission method is provided, including any one of the cloud host data receiving methods mentioned above; and/or any one of the cloud host data transmitting methods mentioned above.

In this way, the queue can be allocated to the downstream traffic of the cloud host according to the destination address, and the bandwidth limitation of the queue can be used to limit the bandwidth of the downlink traffic; and the queue can be allocated to the upstream traffic of the cloud host according to the source address, and the bandwidth limitation of the queue can be utilized. The limitation of the upstream traffic bandwidth is implemented, thereby realizing the bidirectional traffic bandwidth limitation of the cloud host based on the public network address, preventing the influx of excessive traffic and causing the network bandwidth to be occupied by individual users, thereby improving the stability of the cloud host.

According to still another aspect of the present invention, a cloud host data receiving apparatus is provided, including: a destination address obtaining module, configured to acquire destination address information of a cloud host downlink traffic, and the destination address information is a public network address information of the cloud host; The determining module is configured to determine a target queue according to the destination address information based on the correspondence between the queue and the public network address, and the downlink transmission module is configured to perform downlink transmission by using the target queue based on the bandwidth limitation of the target queue.

Optionally, the destination address obtaining module is configured to obtain the downlink traffic entering the network card at the IFB, and obtain the destination address information; the target queue determining module is configured to determine the target queue according to the correspondence between the IFB and the public network address, where the target queue is configured. The bandwidth transmission information is used; the downlink transmission module is configured to return the downlink traffic to the network card under the bandwidth limitation of the target queue, and perform downlink transmission through the network card.

Optionally, the cloud host data receiving device further includes: a network card connection module, configured to connect the network card of the cloud host with the IFB; and a downlink queue generating module, configured to add a TC rule in the IFB, set a queue through the TC rule, and specify each The bandwidth limitation information of the queues; the correspondence generation module is configured to establish a correspondence between the queues and the public network addresses in the IFB.

Optionally, the target queue determining module is further configured to select a predetermined default queue as the target queue when the target queue is not determined according to the destination address information based on the correspondence between the queue and the public network address.

Optionally, the cloud host data receiving apparatus further includes: a deleting module, configured to delete the target queue that matches the public network address, and delete the correspondence between the public network address and the queue, when the public network address of the cloud host is deleted.

Such a device can allocate a queue for the downlink traffic of the cloud host according to the destination address, and use the bandwidth limitation of the queue to implement the limitation of the downlink traffic bandwidth, thereby implementing the downlink traffic restriction of the cloud host based on the public network address, and preventing the influx of excessive traffic. The network bandwidth is occupied by individual users, which improves the stability of the cloud host.

In addition, according to an aspect of the present invention, a cloud host data transmitting apparatus is provided, including: obtaining a source address The module is configured to obtain source address information of the uplink traffic of the cloud host, and the source address information is public network address information of the cloud host, and the target queue determining module is configured to match the source address information with the flow table information to determine a target queue, where The flow table information includes a correspondence between the public network address information and the queue, and the uplink transmission module is configured to perform uplink transmission through the target queue based on the bandwidth limitation of the target queue.

Optionally, the source address obtaining module is configured to obtain source address information of the uplink traffic on the bridge of the cloud host; the target queue determining module is configured to match the source address information with the flow table information in the bridge, and determine the QoS function module. The target queue, wherein the outlet of the queue is bound to the uplink port of the bridge, and each queue has bandwidth limitation information; the uplink transmission module is configured to invoke the quality of service QoS function module, and the uplink traffic is in the bandwidth of the target queue of the QoS function module. Uplink transmission is performed via the uplink port under the restriction.

Optionally, the cloud host data sending apparatus further includes: a bridge binding module, configured to bind the bridge of the cloud host to the QoS function module; and an uplink queue generating module, configured to generate a queue in the QoS function module, for each The queue sets the flow control TC classification, and specifies the bandwidth limitation information by the TC classification; the flow table adding module is used to add the flow table information to the bridge.

Optionally, the target queue determining module is further configured to: when the source address information and the flow table information do not match successfully, select a predetermined default queue as the target queue.

Optionally, the cloud host data sending apparatus further includes: a deleting module, configured to: when deleting the public network address of the cloud host, delete the target queue in the flow table information matching the public network address, and delete the flow table including the public network address information.

Such a device can allocate a queue for the upstream traffic of the cloud host according to the source address, and use the bandwidth limitation of the queue to limit the bandwidth of the upstream traffic, thereby implementing the uplink traffic restriction of the cloud host based on the public network address, and preventing the influx of excessive traffic. The network bandwidth is occupied by individual users, which improves the stability of the cloud host.

In addition, according to still another aspect of the present invention, a cloud host data transmission apparatus is provided, comprising the cloud host data receiving apparatus of any of the above mentioned; and/or cloud host data of any of the above mentioned Sending device.

Such a device can allocate a queue for the downlink traffic of the cloud host according to the destination address, and use the bandwidth limitation of the queue to implement the limitation of the downlink traffic bandwidth; and can allocate a queue according to the source address for the uplink traffic of the cloud host, and implement the bandwidth limitation by using the bandwidth of the queue. The limitation of the upstream traffic bandwidth, which realizes the bidirectional traffic bandwidth limitation of the cloud host based on the public network address, prevents the influx of excessive traffic, and the network bandwidth is occupied by individual users, thereby improving the stability of the cloud host.

According to still another aspect of the present invention, a cloud host data transmission apparatus is provided, including: a memory; A processor coupled to the memory, the processor being configured to perform the method of any of the above mentioned based on the instructions stored in the memory.

Such a device can allocate a queue for the downlink traffic of the cloud host according to the destination address, and use the bandwidth limitation of the queue to implement the limitation of the downlink traffic bandwidth; and can allocate a queue according to the source address for the uplink traffic of the cloud host, and implement the bandwidth limitation by using the bandwidth of the queue. The limitation of the upstream traffic bandwidth, which realizes the bidirectional traffic bandwidth limitation of the cloud host based on the public network address, prevents the influx of excessive traffic, and the network bandwidth is occupied by individual users, thereby improving the stability of the cloud host. Because the rate limit is based on the public network address, the internal network traffic and the public network traffic can be distinguished, so that the hardware resources of the internal network can be fully utilized while the speed limit is increased, thereby improving the performance of the cloud host.

Additionally, in accordance with one aspect of the invention, a computer readable storage medium having stored thereon computer program instructions that, when executed by a processor, implements the steps of any of the above methods.

Such a computer readable storage medium implements a bidirectional traffic bandwidth limitation of a cloud host based on a public network address by executing program instructions thereon, thereby improving the stability of the cloud host. Because the rate limit is based on the public network address, the internal network traffic and the public network traffic can be distinguished, so that the hardware resources of the internal network can be fully utilized while the speed limit is increased, thereby improving the performance of the cloud host.

DRAWINGS

The drawings described herein are intended to provide a further understanding of the invention, and are intended to be a part of the invention. In the drawing:

FIG. 1 is a flowchart of an embodiment of a cloud host data receiving method according to the present invention.

2 is a flow chart of another embodiment of a cloud host data receiving method of the present invention.

FIG. 3 is a flowchart of still another embodiment of a cloud host data receiving method according to the present invention.

4 is a flow chart of an embodiment of a method for transmitting cloud host data according to the present invention.

FIG. 5 is a flowchart of another embodiment of a cloud host data sending method according to the present invention.

FIG. 6 is a flowchart of still another embodiment of a cloud host data sending method according to the present invention.

FIG. 7 is a schematic diagram of an embodiment of a cloud host data transmission method according to the present invention.

FIG. 8 is a schematic diagram of an embodiment of a cloud host data receiving apparatus according to the present invention.

FIG. 9 is a schematic diagram of another embodiment of a cloud host data receiving apparatus of the present invention.

FIG. 10 is a schematic diagram of an embodiment of a cloud host data transmitting apparatus according to the present invention.

FIG. 11 is a schematic diagram of another embodiment of a cloud host data transmitting apparatus according to the present invention.

FIG. 12 is a schematic diagram of an embodiment of a cloud host data transmission apparatus according to the present invention.

FIG. 13 is a schematic diagram of another embodiment of a cloud host data transmission apparatus according to the present invention.

FIG. 14 is a schematic diagram of still another embodiment of a cloud host data transmission apparatus of the present invention.

detailed description

The technical solution of the present invention will be further described in detail below through the accompanying drawings and embodiments.

A flowchart of one embodiment of the cloud host data receiving method of the present invention is shown in FIG.

In step 101, the destination address information of the downlink traffic of the cloud host is obtained, and the destination address information is the public network address information of the cloud host.

In step 102, based on the correspondence between the queue and the public network address, the target queue is determined according to the destination address information. Each queue is pre-configured with public network address information of traffic that can be transmitted, and each queue has a bandwidth limit.

In step 103, downlink transmission is performed through the target queue based on the bandwidth limitation of the target queue. Since the target queue has a bandwidth limit, when the downstream traffic is transmitted through the target queue, it also needs to be transmitted under the bandwidth limit of the queue.

In this way, the queue can be allocated to the downstream traffic of the cloud host according to the destination address, and the bandwidth limitation of the queue can be used to limit the downlink traffic bandwidth, thereby realizing the downlink traffic restriction of the cloud host based on the public network address, thereby preventing excessive flooding. The traffic causes the network bandwidth to be occupied by individual users, which improves the stability of the cloud host. Because the rate limit is based on the public network address, the internal network traffic and the public network traffic can be distinguished, so that the hardware resources of the internal network can be fully utilized while the speed limit is increased, thereby improving the performance of the cloud host.

A flowchart of another embodiment of the cloud host data receiving method of the present invention is shown in FIG. 2.

In step 201, the IFB obtains the downlink traffic entering the network card, and the IFB obtains the destination address information. In one embodiment, the IFB can be invoked with an ifb0 virtual NIC loaded in the IFB. The virtual network card ifb0 can obtain the destination address information of the downlink traffic by using the loaded TC rule.

In step 202, the IFB determines the target queue according to the correspondence between the queue and the public network address. In one embodiment, multiple queues can be configured in the IFB, and the target queue has bandwidth limit information. The target queue of the downstream traffic can be determined by the correspondence between the queue and the address. In one embodiment, a TC rule is set on a port of the virtual network card ifb0, a plurality of queues are set in the TC rule, and bandwidth limitation information of a plurality of queues is specified.

In step 203, the downlink traffic is returned to the network card by the bandwidth limitation of the target queue, and is performed by the network card. Downlink transmission.

In the Linux operating system, TC is used for flow control of the Linux kernel, mainly by establishing a queue at the output port to implement kernel flow control. IFB can be regarded as a virtual network card with only TC filtering function. It does not change the direction of the data packet. For outgoing packets, it is redirected to the IFB. After the IF filtering by the IFB, it is still through the redirection. The network card sends out, and the data packet received by a network card is redirected to the IFB. After being filtered by the IFB TC, the network card before being redirected continues to receive and process. Through such a method, the IFB function in Linux can be used to improve, and the bandwidth limitation of the downlink traffic of the cloud host can be realized, which is advantageous for implementing the application.

A flowchart of still another embodiment of the cloud host data receiving method of the present invention is shown in FIG.

In step 301, the network host of the cloud host is connected to the IFB, so that when the downstream traffic arrives, the traffic is redirected to the IFB. In one embodiment, the IFB driver can be loaded and the virtual network card ifb0 can be started.

In step 302, a TC rule is added to the IFB, a queue is set by the TC rule, and bandwidth limit information for each queue is specified. In one embodiment, multiple queues can be set according to downstream bandwidth requirements. In one embodiment, a root queue may be first set to the ifb0 network card, and then several sub-queues may be added, each sub-queue corresponding to one TC classification. Each TC class sets the maximum rate, the lowest rate and priority that can be used, and so on.

In step 303, a correspondence between the queue and the public network address is established in the IFB. In one embodiment, the queue can be specified based on the public network address. If the destination public network address of the downstream traffic is different, the specified queue is also different, and the queue sets the maximum bandwidth by the TC classification, thereby completing the export bandwidth limitation of ifb0.

In step 304, the downlink traffic entering the network card is forwarded to the intermediate function block device IFB, and the destination address information is obtained by the IFB. In an embodiment, the downlink data entering the network card may be redirected to the root queue of ifb0, and the destination address information is obtained by ifb0.

In step 305, the IFB determines the target queue according to the correspondence between the queue and the public network address.

In step 306, the downlink traffic is returned to the network card under the bandwidth limitation of the target queue, and the downlink transmission is performed via the network card.

Through such a method, the IFB function in Linux can be improved and applied to the current limit of the downlink traffic of the cloud host, which is stable and reliable, and is beneficial to practical applications. At the same time, the bandwidth is set according to the public network IP address, which is convenient for setting and management.

In one embodiment, the bandwidth modification function can be opened, and the interface of the bandwidth and the interface of the bandwidth modification are specified when the user applies for the public network address, so that the bandwidth limitation of the queue can be easily modified. In one embodiment, traffic can be passed Controlling the billing, the higher the bandwidth selected, the more the charge, resulting in economic benefits.

In one embodiment, a default queue can be set, with the default queue having a lower priority than the other queues. When the downlink traffic arrives, the public network address is matched in the order of priority from high to low. When the match fails, the default queue is determined as the target queue and is transmitted under the bandwidth limit of the default queue.

In this way, it is possible to prevent the abnormality of the downlink traffic from being transmitted in the public network address, and the abnormal situation such as congestion, which can ensure the normal operation of the system and improve the robustness of the system.

In an embodiment, when the cloud host deletes the public network address, the queue corresponding to the public network address can be deleted, and the corresponding relationship is deleted, thereby effectively utilizing resources, and preventing invalid information from occupying system resources and reducing efficiency. It is beneficial to maintain the efficient operation of the cloud host.

A flowchart of one embodiment of the cloud host data transmitting method of the present invention is shown in FIG.

In step 401, the source address information of the cloud host uplink traffic is obtained, and the source address information is the public network address information of the cloud host.

In step 402, the source address information is matched with the flow table information to determine a target queue, where the flow table information includes a correspondence between the public network address information and the queue. Each queue is pre-configured with public network address information of traffic that can be transmitted, and each queue has a bandwidth limit.

In step 403, uplink transmission is performed through the target queue based on the bandwidth limitation of the target queue. Since the target queue has a bandwidth limitation, when the upstream traffic is transmitted through the target queue, it also needs to be transmitted under the bandwidth limit of the queue.

In this way, the queue can be allocated to the upstream traffic of the cloud host according to the source address, and the bandwidth limitation of the queue can be used to limit the upstream traffic bandwidth, thereby realizing the uplink traffic restriction of the cloud host based on the public network address, thereby preventing excessive flooding. The traffic causes the network bandwidth to be occupied by individual users, which improves the stability of the cloud host. Because the rate limit is based on the public network address, the internal network traffic and the public network traffic can be distinguished, so that the hardware resources of the internal network can be fully utilized while the speed limit is increased, thereby improving the performance of the cloud host.

A flowchart of another embodiment of the cloud host data transmitting method of the present invention is shown in FIG.

In step 501, the bridge of the cloud host obtains source address information of the uplink traffic.

In step 502, the bridge matches the source address information with the flow table information in the bridge to determine the target queue in the QoS function module. In an embodiment, multiple queues may be configured in the QoS function module, and the egress of the queue is bound to the uplink port of the bridge, and each queue has bandwidth limitation information.

In step 503, the QoS function module is invoked, and the uplink traffic is in the target queue of the QoS function module. Uplink transmission is performed via the uplink port under the bandwidth limitation.

The QoS function module is a functional interface provided by OpenvSwitch. OpenvSwitch is a virtual switching software. It is mainly used in the virtual machine VM environment. As a virtual switch, it supports multiple virtualization technologies. It can support the control of upstream traffic in the virtual machine domain, but it has limited effect on downstream traffic. not ideal. By integrating the QoS function module into the cloud platform, the uplink traffic control function can be well applied, the implementation efficiency is improved, the implementation effect is ensured, and the application is facilitated.

In an embodiment, the bandwidth modification function can be opened, and the interface of the bandwidth and the interface of the bandwidth modification are specified when the user applies for the public network address, so that the bandwidth limitation of the queue can be easily modified. In one embodiment, billing can be performed by flow control, and the higher the bandwidth selected, the more the charge, resulting in economic benefits.

A flowchart of still another embodiment of the cloud host data transmitting method of the present invention is shown in FIG. 6.

In step 601, the bridge of the cloud host is bound to the QoS function module, and when the uplink traffic arrives, the QoS function module is called.

In step 602, a queue is generated in the QoS function module, a TC classification is set for each queue, and bandwidth limitation information is specified by the TC classification. In one embodiment, multiple queues can be set according to upstream bandwidth requirements. In an embodiment, the QoS root queue may be added to the NIC first, and the total bandwidth that can be used is set to be the maximum bandwidth of the NIC; further, multiple subqueues are added under the root queue, and one TC classification corresponding to each subqueue is set. The TC classifies the maximum bandwidth, the minimum guaranteed bandwidth, etc., and each sub-queue can correspond to different bandwidths.

In step 603, the flow table information is added to the bridge, where the flow table information includes the correspondence between the public network address of the cloud host and the queue. In one embodiment, a flow table can be added to the bridge br-ex, the flow table matches the data packet according to the source address, and the matching data packet passes through the target queue. Because the queue sets the highest rate, the flow restriction can be achieved. . When there are multiple public network IPs, multiple such flow tables are added, and each flow label points to the corresponding QoS queue.

In step 604, the bridge of the cloud host obtains source address information of the uplink traffic.

In step 605, the source address information is matched with the flow table information in the bridge to determine the target queue.

In step 606, the QoS function module is invoked to perform uplink transmission on the uplink port under the bandwidth limitation of the target queue of the QoS function module.

Through such a method, the QoS function module in the OVS can be improved and applied to the bandwidth limitation of the uplink traffic of the cloud host, which is stable and reliable, and is beneficial for practical applications.

In one embodiment, a default queue can be set, with the default queue having a lower priority than the other queues. When going up When the traffic arrives, the public network address is matched according to the priority. When the match fails, the default queue is determined as the target queue and is transmitted under the bandwidth limit of the default queue.

In an embodiment, when the cloud host deletes the public network address, the queue corresponding to the public network address can be deleted, and the flow table entry related to the public network address is deleted, thereby effectively utilizing resources and preventing invalidation. The reduced efficiency caused by the information occupied by the system resources is conducive to maintaining the efficient operation of the cloud host.

The cloud host data transmission method of the present invention may include any one of the cloud host data transmission methods and the cloud host data reception method mentioned above. In one embodiment, as shown in FIG.

In 701, the downstream traffic arrives at the network card.

In 702, the NIC redirects traffic to the IFB. The IFB extracts the destination address of the downstream traffic, and then determines the target queue according to the corresponding relationship between the preset queue and the public network address, and implements the bandwidth limitation by using the TC function.

In 703, the IFB returns downstream traffic. Since the downlink traffic bandwidth is limited in the IFB, the traffic returning to the NIC can also achieve the downstream traffic bandwidth limitation.

At 704, the bridge receives upstream traffic.

In 705, the bridge performs address matching based on the flow table information according to the source address of the upstream traffic, determines the target queue, invokes the QoS function module, and uses the bandwidth limitation of the target queue in the QoS function module to implement the limitation of the upstream traffic bandwidth.

In 706, the uplink traffic that is limited by the target queue bandwidth is sent to the external network through the uplink network interface.

Currently OpenStack does not provide a good two-way speed limit method. In this way, the traffic restriction is performed on the IP address of the public network, and the host can achieve the purpose of limiting the current by using the public IP address with the bandwidth set, thereby increasing the flexibility of the bandwidth limitation of the cloud host, which is well implemented based on the public. The two-way speed limit function of the cloud host of the network address can distinguish between intranet traffic and public network traffic, so as to fully utilize the hardware resources of the intranet and improve the performance of the cloud host while limiting the speed.

A schematic diagram of one embodiment of the cloud host data receiving apparatus of the present invention is shown in FIG. The destination address obtaining module 801 can obtain the destination address information of the downlink traffic of the cloud host, and the destination address information is the public network address information of the cloud host. The target queue determination module 802 can determine the target queue based on the destination address information based on the correspondence between the queue and the public network address. Each queue is pre-configured with public network address information of traffic that can be transmitted, and each queue has a bandwidth limit. In one embodiment, the target queue can be performed by means of flow table matching. It is determined that the flow table includes the correspondence between the public network address and the queue. The downlink transmission module 803 can perform downlink transmission through the target queue based on the bandwidth limitation of the target queue. Since the target queue has a bandwidth limit, when the downstream traffic is transmitted through the target queue, it also needs to be transmitted under the bandwidth limit of the queue.

Such a device can allocate a queue for the downlink traffic of the cloud host according to the destination address, and use the bandwidth limitation of the queue to implement the limitation of the downlink traffic bandwidth, thereby implementing the downlink traffic restriction of the cloud host based on the public network address, and preventing the influx of excessive traffic. The network bandwidth is occupied by individual users, which improves the stability of the cloud host. Because the rate limit is based on the public network address, the internal network traffic and the public network traffic can be distinguished, so that the hardware resources of the internal network can be fully utilized while the speed limit is increased, thereby improving the performance of the cloud host.

In an embodiment, the destination address obtaining module 801 can obtain the downlink traffic entering the network card at the IFB, and obtain the destination address information by the IFB. The target queue determination module 802 can determine the target queue based on the correspondence between the IFB and the public network address. In an embodiment, multiple queues may be configured in the IFB, and the target queue has bandwidth limitation information, and the target queue of the downlink traffic may be determined by the correspondence between the queue and the address. The downlink transmission module 803 can return the downlink traffic to the network card under the bandwidth limitation of the target queue, and perform downlink transmission through the network card.

Such a device is improved by using the IFB function in Linux to realize the bandwidth limitation of the downstream traffic of the cloud host, which is advantageous for implementing the application.

A schematic diagram of another embodiment of the cloud host data receiving apparatus of the present invention is shown in FIG. The structure and function of the destination address obtaining module 901, the target queue determining module 902, and the downlink transmitting module 903 are similar to those in the embodiment of FIG. The cloud host data receiving device further includes a network card connection module 904, a downlink queue generation module 905, and a correspondence relationship generation module 906. The network card connection module 904 can connect the network card of the cloud host with the IFB, so that when the downlink traffic arrives, it is convenient to redirect the traffic to the IFB. The downlink queue generation module 905 can add a TC rule in the IFB, set a queue through the TC rule, and specify bandwidth limitation information for each queue. In one embodiment, multiple queues can be set according to downstream bandwidth requirements. The correspondence generation module 906 can establish a correspondence between the queue and the public network address in the IFB.

Such a device can improve the IFB function in Linux and apply it to the current limit of the downlink traffic of the cloud host, which is stable and reliable, and is beneficial for practical applications.

In one embodiment, the queue includes a default queue, and the default queue has a lower priority than the other queues. When the downlink traffic arrives, the target queue determining module can match the public network address in descending order of priority. When the match fails, the default queue is determined as the target queue, which is performed under the bandwidth limit of the default queue. transmission.

Such a device can prevent the abnormality of the downlink traffic from being transmitted in the public network address, and the abnormal situation such as congestion, which can ensure the normal operation of the system and improve the robustness of the system.

In an embodiment, the cloud host data receiving device further includes a deleting module, which is capable of deleting a queue corresponding to the public network address when the cloud host deletes the public network address, and deleting the corresponding relationship, thereby realizing efficient use of resources. Preventing the failure information from consuming the system resources and reducing the efficiency is beneficial to maintaining the efficient operation of the cloud host.

A schematic diagram of one embodiment of the cloud host data transmitting apparatus of the present invention is shown in FIG. The source address obtaining module 1001 is configured to obtain source address information of the cloud host uplink traffic, where the source address information is public network address information of the cloud host. The target queue determining module 1002 is configured to match the source address information with the flow table information to determine a target queue, where the flow table information includes a correspondence between the public network address information and the queue. Each queue is pre-configured with public network address information of traffic that can be transmitted, and each queue has a bandwidth limit. The uplink transmission module 1003 is configured to perform uplink transmission through the target queue based on the bandwidth limitation of the target queue. Since the target queue has a bandwidth limitation, when the upstream traffic is transmitted through the target queue, it also needs to be transmitted under the bandwidth limit of the queue.

Such a device can allocate a queue for the upstream traffic of the cloud host according to the source address, and use the bandwidth limitation of the queue to limit the bandwidth of the upstream traffic, thereby implementing the uplink traffic restriction of the cloud host based on the public network address, and preventing the influx of excessive traffic. The network bandwidth is occupied by individual users, which improves the stability of the cloud host. Because the rate limit is based on the public network address, the internal network traffic and the public network traffic can be distinguished, so that the hardware resources of the internal network can be fully utilized while the speed limit is increased, thereby improving the performance of the cloud host.

In an embodiment, the source address obtaining module 1001 can obtain source address information of the uplink traffic at the bridge of the cloud host. The target queue determination module 1002 can match the source address information with the flow table information in the bridge through the bridge to determine the target queue in the QoS function module. In an embodiment, multiple queues may be configured in the QoS function module, and the egress of the queue is bound to the uplink port of the bridge, and each queue has bandwidth limitation information. The uplink transmission module 1003 can invoke the QoS function module to perform uplink transmission on the uplink port under the bandwidth limitation of the target queue of the QoS function module.

Such a device integrates the QoS function module into the cloud platform, which can well apply its upstream flow control function, improve the efficiency of implementation, ensure the realization effect, and is beneficial to practical applications.

A schematic diagram of another embodiment of the cloud host data transmitting apparatus of the present invention is shown in FIG. The structure, function, and function of the source address obtaining module 1101, the target queue determining module 1102, and the uplink transmitting module 1103 Similar in the embodiment of 10. The cloud host data sending apparatus further includes a bridge binding module 1104, an uplink queue generating module 1105, and a flow table adding module 1106. The bridge binding module 1104 can bind the bridge of the cloud host to the QoS function module, and when the uplink traffic arrives, the QoS function module is called. The uplink queue generation module 1105 can generate a queue in the QoS function module, set a TC classification for each queue, and specify bandwidth limitation information by the TC classification. In one embodiment, multiple queues can be set according to upstream bandwidth requirements. The flow table adding module 1106 can add flow table information to the bridge, where the flow table information includes the correspondence between the public network address of the cloud host and the queue.

Such a device improves the QoS function module in the OpenvSwitch and applies it to the bandwidth limitation of the uplink traffic of the cloud host, is stable and reliable, and is beneficial for practical applications.

In one embodiment, a default queue can be set, with the default queue having a lower priority than the other queues. When the uplink traffic arrives, the target queue determining module performs matching of the public network address according to the priority level. When the match fails, the default queue is determined as the target queue and is transmitted under the bandwidth limit of the default queue.

Such a device can prevent the downlink traffic from being transmitted, and the abnormal situation such as congestion, which can prevent the normal operation of the system and improve the robustness of the system.

In an embodiment, the cloud host data sending device further includes a deleting module, which can delete a queue corresponding to the public network address when the cloud host deletes the public network address, and delete the flow table entry related to the public network address. Thereby, the efficient use of resources is realized, and the efficiency reduction caused by the failure information occupying system resources is prevented, and the efficient operation of the cloud host is maintained.

The cloud host data transmission device of the present invention may include any one of the cloud host data transmitting devices and the cloud host data receiving device mentioned above. In one embodiment, as shown in FIG. 12, the cloud host data transmission device includes a cloud host data receiving device 1200 and a cloud host data transmitting device 1210. The cloud host data receiving device 1200 includes a destination address obtaining module 1201, a target queue determining module 1202, and a downlink transmitting module 1203. The cloud host data transmitting device 1210 includes a source address obtaining module 1211, a target queue determining module 1212, and an uplink transmitting module 1213. The cloud host data receiving apparatus 1200 can implement a downlink traffic bandwidth limitation based on a public network address; the cloud host data transmitting apparatus 1210 can implement an uplink traffic bandwidth limitation based on a public network address.

A schematic diagram of another embodiment of the cloud host data transmission apparatus of the present invention is shown in FIG. 13, and includes a memory 1301 and a processor 1302. among them:

Memory 1301 can be a magnetic disk, flash memory, or any other non-volatile storage medium. The memory 1301 is used to store running instructions, queues, bandwidth limits, flow table information, and the like of the system.

The processor 1302 is coupled to the memory 1301 and can be implemented as one or more integrated circuits, such as a microprocessor or a microcontroller. The processor 1302 is configured to execute instructions stored in the memory, thereby implementing data operations, transmissions, and the like.

In one embodiment, as shown in FIG. 14, 1400 is a cloud host data transmission device, including a memory 1401 and a processor 1402. Processor 1402 can include

processors

1402a, 1402b...1402n. Processors 1402a-1402n are coupled to memory 1401 via BUS bus 1403. Systems based on distributed architectures enable fast calculations and improved operational efficiency. The cloud host data transmission device 1400 can also be connected to the external storage device 1405 through the storage interface 1404 to invoke external data or transfer data to the external storage device 1405, and can also be connected to the network or another computer system through the network interface 1406 (not labeled Out). It will not be described in detail here.

In this embodiment, by storing the data instruction in the memory and processing the above instruction by the processor, the bidirectional traffic bandwidth limitation of the cloud host based on the public network address can be implemented, and the intrusion of excessive traffic is prevented, so that the network bandwidth is occupied by individual users, thereby improving The stability of the cloud host.

In still another embodiment, a computer readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the steps of a method in a cloud host data transfer method corresponding to an embodiment. Those skilled in the art will appreciate that embodiments of the present application can be provided as a method, apparatus, or computer program product. Thus, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment in combination of software and hardware. Moreover, the application can take the form of a computer program product embodied on one or more computer-usable non-transitory storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) .

The present invention has been described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (system) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or FIG. These computer program instructions can be provided to a processor of a general purpose computer, a special purpose computer, an embedded processor, or other programmable data processing device to produce a machine such that a processor through a computer or other programmable data processing device The executed instructions produce means for implementing the functions specified in one or more blocks of the flowchart or in a block or blocks of the flowchart.

The computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device. The apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.

These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operational steps are performed on a computer or other programmable device to produce computer-implemented processing for execution on a computer or other programmable device. The instructions provide steps for implementing the functions specified in one or more of the flow or in a block or blocks of a flow diagram.

So far, the present invention has been described in detail. In order to avoid obscuring the concepts of the present invention, some details known in the art are not described. Those skilled in the art can fully understand how to implement the technical solutions disclosed herein according to the above description.

The method and apparatus of the present invention may be implemented in a number of ways. For example, the methods and apparatus of the present invention can be implemented in software, hardware, firmware, or any combination of software, hardware, and firmware. The above-described sequence of steps for the method is for illustrative purposes only, and the steps of the method of the present invention are not limited to the order specifically described above unless otherwise specifically stated. Moreover, in some embodiments, the invention may also be embodied as a program recorded in a recording medium, the program comprising machine readable instructions for implementing the method according to the invention. Thus, the invention also covers a recording medium storing a program for performing the method according to the invention.

It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to be limiting; although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that The invention is not limited to the spirit of the technical solutions of the present invention, and should be included in the scope of the technical solutions claimed in the present invention.

Claims

A cloud host data receiving method includes:

Obtaining destination address information of the downlink traffic of the cloud host, where the destination address information is public network address information of the cloud host;

Determining a target queue according to the destination address information based on a correspondence between the queue and the public network address;

Downlink transmission is performed through the target queue based on a bandwidth limit of the target queue.
The method of claim 1 wherein

The obtaining destination address information of the downlink traffic of the cloud host includes:

The intermediate function block device IFB acquires the downlink traffic entering the network card, and acquires the destination address information;

Determining, according to the correspondence between the queue and the public network address, determining the target queue according to the destination address information includes:

Determining, by the IFB, the target queue according to a correspondence between a queue and the public network address, where the target queue has the bandwidth limitation information;

The downlink transmission by using the target queue based on the bandwidth limitation of the target queue includes:

And downlinking the downlink traffic to the network card under the bandwidth limitation of the target queue, and performing downlink transmission by using the network card.
The method of claim 2 further comprising:

Connecting a network card of the cloud host to the IFB;

Adding a TC rule to the IFB, setting a queue by using the TC rule, and specifying bandwidth limitation information of each of the queues;

Establishing a correspondence between the queue and the public network address in the IFB.
The method according to any one of claims 1 to 3, further comprising:

If the target queue is not determined according to the destination address information based on the correspondence between the queue and the public network address, the predetermined default queue is selected as the target queue.
The method according to any one of claims 1 to 3, further comprising: when deleting the public network address of the cloud host, deleting a target queue that matches the public network address, and deleting the public network address and Correspondence of the queues.
A cloud host data sending method includes:

Obtaining source address information of the cloud host uplink traffic, where the source address information is public network address information of the cloud host;

Matching the source address information with the flow table information to determine a target queue, where the flow table information includes a correspondence between the public network address information and the queue;

Uplink transmission is performed through the target queue based on a bandwidth limit of the target queue.
The method of claim 6 wherein

The source address information of obtaining the uplink traffic of the cloud host includes:

The bridge of the cloud host acquires source address information of the uplink traffic;

The matching the source address information with the flow table information, and determining the target queue includes:

The bridge matches the source address information with the flow table information in the bridge to determine a target queue in the QoS function module, where the outlet of the queue is bound to an uplink port of the bridge, and each The queue has bandwidth limit information;

The uplink transmission by using the target queue based on the bandwidth limitation of the target queue includes:

The service quality QoS function module is invoked, and the uplink traffic is uplinked by the uplink port under the bandwidth limitation of the target queue of the QoS function module.
The method of claim 7 further comprising:

Binding the bridge of the cloud host to the QoS function module;

Generating a queue in the QoS function module, setting a flow control TC classification for each of the queues, and specifying bandwidth limitation information by using the TC classification;

Add flow table information to the bridge.
A method according to any one of claims 6 to 8, further comprising:

If the source address information and the flow table information do not match successfully, the predetermined default queue is selected as the target queue.
A method according to any one of claims 6 to 8, further comprising:

When the public network address of the cloud host is deleted, the target queue in the flow table information matching the public network address is deleted, and the flow table information including the public network address is deleted.
A cloud host data transmission method, comprising the cloud host data receiving method according to any one of claims 1 to 5; and/or

The cloud host data transmitting method according to any one of claims 6 to 10.
A cloud host data receiving device includes:

a destination address obtaining module, configured to obtain destination address information of a downlink traffic of the cloud host, where the destination address information is public network address information of the cloud host;

a target queue determining module, configured to determine a target queue according to the destination address information based on a correspondence between the queue and the public network address;

And a downlink transmission module, configured to perform downlink transmission by using the target queue based on a bandwidth limitation of the target queue.
The device according to claim 12, wherein

The destination address obtaining module is configured to acquire the downlink traffic entering the network card by using the intermediate function block device IFB, and obtain the destination address information;

The target queue determining module is configured to determine the target queue according to a correspondence between the IFB and the public network address, where the target queue has the bandwidth limitation information;

The downlink transmission module is configured to return the downlink traffic to the network card by using a bandwidth limitation of the target queue, and perform downlink transmission by using the network card.
The apparatus of claim 13 further comprising:

a network card connection module, configured to connect the network card of the cloud host with the IFB;

a downlink queue generating module, configured to add a TC rule in the IFB, set a queue by using the TC rule, and specify bandwidth limitation information of each of the queues;

The correspondence generation module is configured to establish a correspondence between the queue and the public network address in the IFB.
The apparatus according to any one of claims 12 to 14, wherein

The target queue determining module is further configured to: when the target queue is not determined according to the destination address information based on the correspondence between the queue and the public network address, select a predetermined default queue as the target queue.
The apparatus according to any one of claims 12 to 14, further comprising:

The deleting module is configured to: when the public network address of the cloud host is deleted, delete the target queue that matches the public network address, and delete the correspondence between the public network address and the queue.
A cloud host data sending device includes:

The source address obtaining module is configured to obtain source address information of the cloud host uplink traffic, where the source address information is public network address information of the cloud host;

a target queue determining module, configured to match the source address information and the flow table information to determine a target queue, where the flow table information includes a correspondence between the public network address information and the queue;

An uplink transmission module, configured to perform uplink transmission by using the target queue based on a bandwidth limit of the target queue.
The device according to claim 17, wherein

The source address obtaining module is configured to acquire source address information of the uplink traffic on a bridge of a cloud host;

The target queue determining module is configured to match the source address information with the flow table information in the bridge to determine a target queue in the QoS function module, where the queue is bounded by the network The upstream port of the bridge, and each queue has bandwidth limitation information;

The uplink transmission module is configured to invoke a quality of service QoS function module, and perform uplink transmission on the uplink port by using the uplink traffic in the bandwidth limitation of the target queue of the QoS function module.
The apparatus of claim 18, further comprising:

a bridge binding module, configured to bind the bridge of the cloud host to the QoS function module;

An uplink queue generating module, configured to generate a queue in the QoS function module, set a flow control TC classification for each of the queues, and specify bandwidth limitation information by using the TC classification;

A flow table adding module is configured to add flow table information to the bridge.
The method according to any one of claims 17 to 19, wherein the target queue determining module is further configured to: when the source address information and the flow table information do not match successfully, select a predetermined default queue as the target queue.
The method according to any one of claims 17 to 19, further comprising:

And deleting the target queue in the flow table information matching the public network address, and deleting the flow table information including the public network address, when the public network address of the cloud host is deleted.
A cloud host data transmission device, comprising the cloud host data receiving device according to any one of claims 12 to 16; and/or

The cloud host data transmitting apparatus according to any one of claims 17 to 21.
A cloud host data transmission device includes:

Memory;

A processor coupled to the memory, the processor being configured to perform the method of any one of claims 1 to 10 based on an instruction stored in the memory.
A computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, perform the steps of the method of any one of claims 1 to 10.