WO2018024063A1 - Data transmission quality detection method and device - Google Patents

Abstract

The present invention relates to the technical field of communications; provided are a data transmission quality detection method and device, used to detect data transmission quality between different virtual machines (VM) in an NFV system. The method comprises: staining a plurality of packets having a same transmission path sent by a first VM, detecting and counting the number of stained packets at different nodes on the transmission path, comparing the numbers of stained packets detected at the different nodes to obtain differences, and determining the network quality between the different nodes according to the differences.

Description

Data transmission quality detecting method and device

This application claims priority to Chinese Patent Application No. 201610634978.2, entitled "A Data Transmission Quality Detection Method and Apparatus", filed on August 4, 2016, the entire contents of which are incorporated herein by reference. In the application.

Technical field

The present invention relates to the field of communications technologies, and in particular, to a data transmission quality detecting method and apparatus.

Background technique

The traditional telecommunication system is composed of various dedicated hardware devices, and different applications use different hardware devices. As the scale of the network grows, the system becomes more and more complex, which brings many challenges, including the development of new services, the operation and maintenance of the system, and resource utilization. In response to these challenges and the use of information technology (English full name: Information Technology, referred to as: IT) industry virtualization technology and cloud computing technology, in 2012, telecom operators jointly released network function virtualization (English full name: Network Function Virtualization, referred to as : NFV) White Paper, announcing the establishment of the NFV Industry Standards Organization (English name: Industry Standard Group, ISG) at the European Telecommunications Standards Institute (English name: ETSI) to develop NFV requirements and technical frameworks. Promote the development of NFV.

After the NFV of the telecom equipment, a large number of virtual machines are interconnected through the Internet Protocol of the Cloud Platform (English name: Internet Protocol, IP for short) to jointly provide telecommunication services. Different from the traditional telecommunication equipment, after the NFV of the telecommunication equipment, the network becomes complicated. The virtual network function of the NFV (English name: Virtual Network Function, VNF for short) is the internal virtual machine (English name: Virtual Machine, VM for short). The communication traffic is merged with the external traffic channel, and the probability of IP packet loss, jitter, delay, etc. affecting the quality of network communication increases.

Since the quality of the network communication between the VMs has a great impact on the normal operation of the service, when the above problems occur between the two virtual machines and affect the normal operation of the service, the data transmission quality between the VMs needs to be detected, and the boundary is quickly delimited. The problem occurs in order to take appropriate measures (such as virtual machine migration, reconstruction, etc.) to quickly restore the business. However, there is currently no good detection scheme to detect the quality of data transmission between VMs under NFV.

Summary of the invention

Embodiments of the present invention provide a data transmission quality detecting method and apparatus to detect data transmission quality between different VMs in an NFV system.

In order to achieve the above object, embodiments of the present invention adopt the following technical solutions:

In a first aspect, an embodiment of the present invention provides a data transmission quality detection method, which is applied to a network function virtualization NFV system, and the method may include:

A plurality of data packets with the same transmission path sent by the first virtual machine VM are dyed, and different nodes on the transmission path detect and count the number of the dyed data packets, and compare the difference of the number of the dyed data packets detected by different nodes. According to the difference, the network quality between different nodes is judged.

Specifically, if the number of the dyed data packets detected by the first node is greater than the number of the dyed data packets detected by the second node The quantity determines that there is a packet loss problem between the first node and the second node, wherein the first node may be the first VM, and the second node may be any other node on the transmission path.

In this way, the difference in the number of dyed packets on different nodes can be compared by dyeing, identifying, and counting the data packets, and whether the packet loss problem occurs between the networks is determined according to the difference.

The data packet may be an IP data packet or an Ethernet data packet.

Optionally, in an implementation manner of the first aspect, when the data packet is an IP data packet, coloring the IP data packet may include:

The dye flag is filled in the time-to-live TTL field of the IP packet header.

When the packet is an Ethernet packet, the Ethernet packet is dyed and can include:

The dye flag is filled in the preset offset field of the Ethernet packet header.

The dyeing identifier may be a preset value, and the value may be used to identify the dyed data packet; the preset offset field may be set as needed.

Optionally, in another implementation manner of the first aspect, the foregoing transmission path may be: a path from the first VM to the second VM through the first virtual switch, a second VM, a first virtual switch, and a One VM is on the same host;

Alternatively, the path from the first VM to the second virtual switch to the second virtual switch, and the path from the second virtual switch to the second VM, the first VM and the first virtual switch are located at the first host, the second VM and the second virtual switch Located on the second host.

The first VM and the second VM are located in a virtual network function VNF entity in the NFV system; the first virtual switch and the second virtual switch are located in a virtual network of the network function virtualization infrastructure NFVI in the NFV system.

In the actual application, the data packets sent by the VM are sequential, and it is difficult for each node to delimit the start time and the final occurrence time of the data packet, which is prone to the problem of statistical statistics of the dyed data packet. Therefore, in order to avoid the transmission path In another implementation manner of the first aspect, the coloring of the multiple data packets sent by the first virtual machine VM may include:

The first VM sends data packets in a plurality of consecutive time periods, wherein the data packets sent in the same time period have the same dyeing identifier, and the data packets sent in the adjacent time segments have different dyeing identifiers.

The time period can be divided according to the requirements, which is not limited by the embodiment of the present invention.

Optionally, before the plurality of data packets sent by the first virtual machine VM are dyed, receiving a detection task for indicating that a plurality of data packets sent by the first VM are stained, where the detection task may include a time period And, according to the time period of the received detection task, the time when the first VM sends the data packet may be divided into multiple consecutive time segments, and the data packets in each time segment are dyed, so that each node on the transmission path The number of dyed data packets sent in each time period can be counted, and the network quality between different nodes in the time period is determined by comparing the number of dyed data packets between different nodes in each time period.

In a second aspect, the embodiment of the present invention further provides a data transmission quality detecting method, which is applied to a network function virtualization NFV system, and the method may include:

The first data packet of the plurality of data packets with the same transmission path sent by the first virtual machine VM is dyed, and different time nodes on the transmission path detect and count the time stamp of the first color data packet, and compare different node detections. The difference in the time stamp of the first dyed data packet determines the network quality between different nodes according to the difference.

Specifically, if the difference between the time stamp of the first coloring data packet detected between the first node and the second node is greater than the time stamp of the first coloring data packet detected between the third node and the fourth node, Determining between the first node and the second node The transmission delay is greater than the transmission delay between the third node and the fourth node.

In this way, the difference between the time stamps of the first dyed data packets on different nodes can be compared by dyeing, identifying, and counting the data packets, and the delay problem between the networks is determined according to the difference.

The data packet may be an IP data packet or an Ethernet data packet.

Optionally, in an implementation manner of the second aspect, when the data packet is an IP data packet, the color data packet may be:

The dye flag is filled in the time-to-live TTL field of the IP packet header.

When the packet is an Ethernet packet, the Ethernet packet is dyed and can include:

The dye flag is filled in the preset offset field of the Ethernet packet header.

The first coloring identifier may be a preset value, and the value may be used to identify the first coloring data packet; the preset offset field may be set as needed.

Optionally, in another implementation manner of the second aspect, the foregoing transmission path may be: a path from the first VM to the second VM through the first virtual switch, a second VM, a first virtual switch, and a One VM is on the same host;

Alternatively, the path from the first VM to the second virtual switch to the second virtual switch, and the path from the second virtual switch to the second VM, the first VM and the first virtual switch are located at the first host, the second VM and the second virtual switch Located on the second host.

The first VM and the second VM are located in a virtual network function VNF entity in the NFV system; the first virtual switch and the second virtual switch are located in a virtual network of the network function virtualization infrastructure NFVI in the NFV system.

In the actual application, the data packets sent by the VM are sequential, and it is difficult for each node to delimit the start time and the final occurrence time of the data packet, which is prone to the problem of statistical statistics of the dyed data packet. Therefore, in order to avoid the transmission path In another implementation manner of the second aspect, the first data packet sent by the first virtual machine VM may be:

Data packets sent by the first VM in a plurality of consecutive time periods are dyed, wherein the first data packet sent in the same time period is dyed with the first coloring identifier, and the first data packet sent in the same time period except the first data packet All other data packets are dyed with the second staining identifier, and the first data packet sent in the adjacent time period has different dyeing identifiers.

It should be noted that the foregoing time period may be divided according to requirements, which is not limited by the embodiment of the present invention.

Optionally, before the first data packet of the plurality of data packets sent by the first virtual machine VM is dyed, receiving a detection task for indicating that the plurality of data packets sent by the first VM are dyed, where The detecting task may include a time period, and the time that the first VM sends the data packet may be divided into multiple consecutive time segments according to the time period in the received detecting task, and the data packet in each time segment is dyed, so that Each node on the transmission path can count the time stamp of the first coloring data packet sent in each time period, and compare the time stamps of the first coloring data packet between different nodes in each time period to determine the time between different nodes in the time segment. Transmission delay problem.

In a third aspect, the embodiment of the present invention further provides a data transmission quality detecting apparatus, which is configured to execute the foregoing method, and the apparatus may include: a dyeing module, a statistics module, and a comparison determining module.

The dyeing module is located in the first virtual machine VM in the network function virtualization NFV system, and the dyeing module is configured to perform dyeing on the plurality of data packets with the same transmission path sent by the first VM;

The statistics module is located in different nodes on the transmission path, and the statistics module is configured to detect and count the number of dye packets received by the node where the node is located;

a comparison judgment module for comparing differences in the number of dye packets detected by the statistical modules in different nodes, The network quality between different nodes is judged based on the difference.

For the specific execution process of the foregoing dyeing module, the statistic module, and the comparison judging module, reference may be made to the corresponding process in the method described in the first aspect, and details are not described herein again.

In this way, the difference in the number of dyed packets on different nodes can be compared by dyeing, identifying, and counting the data packets, and whether the packet loss problem occurs between the networks is determined according to the difference.

Further, in order to implement the detection of the delay problem in the data transmission process, the dyeing module is further configured to perform dyeing on the first data packet of the plurality of data packets having the transmission path sent by the first virtual machine VM;

The statistics module is further configured to detect and count a time stamp of the first dyed data packet received by the node where the node is located;

The comparison judgment module is further configured to compare the difference of the time stamps of the first dyed data packets detected by the statistical modules in different nodes, and determine the network quality between the different nodes according to the difference.

For the specific execution process of the foregoing dyeing module, the statistic module, and the comparison judging module, reference may be made to the corresponding processes in the method described in the second aspect, and details are not described herein again.

In this way, the difference between the time stamps of the first dyed data packets on different nodes can be compared by dyeing, identifying, and counting the data packets, and the delay problem between the networks is determined according to the difference.

It can be understood that the foregoing data transmission quality detecting apparatus can separately perform the detection of the packet loss problem in the data transmission process, separately perform the detection of the delay problem in the data transmission process, or combine the above schemes, and simultaneously adopt the above The method implements the detection of packet loss and delay.

In addition, it should be noted that the data transmission quality detecting apparatus described in the third aspect may also be deployed as an independent device in the NFV system, and when the data transmission quality detecting device is used as a separate device, the dyeing module and the statistical module in the device. And the comparison judging module may be a separately set processor, or may be integrated in a processor of the data transmission quality detecting device, or may be stored in a memory of the data transmission quality detecting device in the form of program code. The function of the above dyeing module, the statistical module, and the comparison judging module is called and executed by one of the processors of the data transmission quality detecting device. The processor described herein may be a central processing unit (English name: Central Processing Unit, CPU for short), or a specific integrated circuit (English name: Application Specific Integrated Circuit, ASIC for short), or configured to implement One or more integrated circuits of embodiments of the present application.

As can be seen from the above, the embodiment of the present invention provides a data transmission quality detecting method and apparatus, which dyes multiple data packets with the same transmission path sent by the first virtual machine VM, and detects and statistically dyes at different nodes on the transmission path. The number of packets compares the difference in the number of dyed packets detected by different nodes, and judges the network quality between different nodes according to the difference. In this way, the difference in the number of data packets sent between different nodes can be compared by dyeing, identifying, and counting the data packets, and then the packet loss problem between the nodes can be determined according to the difference, so that corresponding measures can be taken to restore the entire service.

DRAWINGS

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the description of the prior art will be briefly described below. Obviously, the drawings in the following description are only It is a certain embodiment of the present invention, and other drawings can be obtained from those skilled in the art without any creative work.

1 is a schematic diagram of an NFV system architecture;

2 is a flowchart of a method for detecting data transmission quality according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of several data transmissions according to an embodiment of the present invention; FIG.

FIG. 4 is a flowchart of a method for detecting data transmission quality according to an embodiment of the present invention;

FIG. 5 is a structural diagram of a data transmission quality detecting apparatus according to an embodiment of the present invention;

5A is a structural diagram of an NFV system including a data transmission quality detecting apparatus according to an embodiment of the present invention;

FIG. 6 is a flowchart of a method for detecting data transmission quality according to an embodiment of the present invention.

detailed description

The technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the drawings in the embodiments of the present invention. It is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

It should be noted that the term “and/or” in this document is merely an association relationship describing an association object, indicating that there may be three relationships, for example, A and/or B, which may indicate that A exists separately, and A exists at the same time. And B, there are three cases of B alone. In addition, the character "/" in this article generally indicates that the contextual object is an "or" relationship.

In order to cope with future competition and challenges, operators avoid being pipelined. In response to the current development trend of virtualization and cloud computing, operators have proposed NFV systems. FIG. 1 is a schematic diagram of an NFV system architecture. As shown in FIG. 1 , the NFV system 10 may include: NFV management and orchestration device (English name: NFV Management and Orchestration, referred to as: NFV MANO) 101, NFV infrastructure layer (English full name) : NFV Infrastructure, referred to as: NFVI) 102, multiple virtual network functions (English full name: Virtual Network Function, referred to as: VNF) 103, multiple network element management (English full name: Element Management, referred to as: EM) 104, network services, VNF and Infrastructure Description (VNF and Infrastructure Description) 105, and Business Support Management Device (English name: Operation-Support System/Business Support System, abbreviated as OSS/BSS) 106.

Among them, NFV-MANO101 can include: NFV orchestrator (English full name: NFV Orchestrator, referred to as: NFVO) 1011, one or more VNF managers (English full name: VNF Manager, referred to as: VNFM) 1012, and virtualization infrastructure management (English name: Virtualized Infrastructure Manager, referred to as: VIM) 1013; NFV MANO101 is mainly used to monitor and manage VNF103 and NFVI102; NFVO1011 in NFV MANO101 is mainly used to receive resource related requests from one or more VNFM1012, and Send configuration information to VNFM1012 to collect status information of VNF103. In addition, NFVO1011 is also used to communicate with VIM1013 to implement resource allocation and/or reservation and exchange configuration and status information of virtualized hardware resources. VNFM1012 is mainly used to manage one. Or multiple VNFs 103, such as: instantiate, update, query, scale, and/or terminate VNF103, etc.; VIM1013 mainly performs resource management functions, such as managing the allocation of infrastructure resources (such as adding resources to virtual containers) and operating functions (such as Collect NFVI failure information).

The NFVI 102 can include computing hardware 1021, storage hardware 1022, network hardware 1023, virtualization layer, virtual computing 1024, virtual storage 1025, and virtual network 1026. The virtualization layer in NFVI 102 can abstract hardware resources from the physical layer and decouple VNF 103 to provide virtualized resources to VNF 103. Virtual computing 1024 and virtual storage 1025 can be provided to VNF 103 in the form of virtual machines, and/or other virtual containers. For example, one or more VNFs 103 can be deployed in a virtual machine (English full name: Virtual Machine, Abbreviation: VM). The virtualization layer abstracts network hardware 1023 to form a virtual network 1026, which may include a virtual switch that can be used to provide connection communication between VMs on the VNF 103 and other VMs.

As shown in Figure 1, under the NFV system, the communication traffic between VMs is merged with the external traffic channel, the hardware resource layer traffic is shared by all virtual machines, and the network hardware (such as network card and other hardware) is also shared. When the number is large, the probability of IP packet loss, jitter, delay, etc. affecting the quality of data transmission increases. In this case, in order not to affect the normal transmission of data, quality detection of data transmission between VMs is required to quickly The location of the demarcation problem occurs so that appropriate measures (such as VM migration, reconstruction, etc.) can be taken to quickly restore the business.

To this end, the embodiment of the present invention is based on the NFV system shown in FIG. 1, and the data packet in the service flow is dyed at the originating node of the service flow. When the service flow flows through each node according to the transmission path, each node pair The dyeing data packet is identified and counted, and the data and time stamp of the dyed data packet are obtained, and then the change of each node of the transmission path can be timely marked according to the number of dyed data packets, and the packet loss situation between different nodes is determined. And the transmission delay, and the location of the problem is delimited according to the judgment result.

For convenience of description, the following embodiment 1 shows and describes in detail the data transmission quality detecting method provided by the present invention, wherein the steps shown may also be in addition to the NFV system, such as a set of executable instructions. Executed in a computer device. Moreover, although logical sequences are shown in the figures, in some cases the steps shown or described may be performed in a different order than the ones described herein.

Embodiment 1

FIG. 2 is a flowchart of a method for detecting data transmission quality according to an embodiment of the present invention. The method is applicable to an NFV system as shown in FIG. 1 for detecting data transmission quality between different VMs in an NFV system; As shown in FIG. 2, the method may include the following steps:

S101: Sampling a plurality of data packets sent by the first virtual machine VM, the multiple data packets having the same transmission path.

The first VM may be any VM in the VNF under the NFV system, and is an originating node that sends multiple data packets.

The data packet may be an IP data packet, or may be a private Layer 2 data packet, such as an Ethernet data packet, which is not limited in this embodiment of the present invention. The multiple data packets may be included in the same service flow, and the service flow is a service flow with a quality problem, a quality check to be performed, or a service flow specified for the user.

For example, before step S101, the quality condition of the multiple service flows sent by the first VM may be detected, and if the quality problem of the first service flow is found, the multiple data in the first service flow sent by the first VM may be Packing for dyeing;

Or, before the step S101, receiving the quintuple information of the first service flow delivered by the user, and dyeing, according to the received quintuple information, the plurality of data packets in the first service flow sent by the first VM, where The quintuple information may include: a source IP address, a destination IP address, a protocol type, a source port number, and a destination port number of the service flow.

Usually, a service flow has a specific transmission path along which data packets in the service flow are transmitted. In the NFV system, the transmission path may be: a path from the first VM through the first virtual switch to the second VM, where the second VM, the first virtual switch, and the first VM may be in the same host;

The path may be a path from the first VM to the second virtual switch, and then from the second virtual switch to the second VM, where the first VM and the first virtual switch are located in the first host, and the second VM and the second The second virtual switch may be located in the second host, and the first host and the second host may be in the same VNF or in different VNFs;

The second VM may also be located in the VNF of the NFV system; the first virtual switch and the second virtual switch It can be located in a virtual network in the NFVI of the NFV system.

For example, FIG. 3 is a schematic diagram of several data transmissions provided by an embodiment of the present invention. As shown in FIG. 3( a ), a VM 1 in the host 1 can transmit multiple data packets to the host 1 through a virtual switch in the host 1 . VM2; as shown in Figure 3(b), host 1 and host 2 are in the same VNF network. At this time, VM1 in host 1 can transmit the data packet to virtual in host 2 through virtual switch 1 in host 1. Switch 2, virtual switch 2 then transmits the data packet to VM2 in host 2; as shown in Figure 3 (c), host 1 and host 2 are in different VNF networks. At this time, VM1 in host 1 can pass through host 1 The virtual switch 1 inside transfers the data packet to the virtual switch 2 in the host 2, and the virtual switch 2 transmits the data packet to the VM2 in the host 2.

It can be understood that although the above describes only the transmission path between two VMs, the transmission path between two or more VMs may also be set by referring to the foregoing manner, that is, the VMs in the same host pass through the virtual in the host. The switch transmits data packets, and the VMs in different hosts transmit data packets through the virtual switches in the respective hosts, and details are not described herein.

Among them, in the process of dyeing data packets, the dyeing bits are different for different types of data packets. Optionally, when the data packet is an IP data packet, according to the special format of the IP data packet, the dyeing identifier may be filled in the time of the IP data packet header (Time To Live, TTL for short), the dyeing identifier Used to identify the IP packets that are being dyed.

When the data packet is an Ethernet data packet, according to the special format of the Ethernet data packet, a dyeing identifier can be filled in the preset offset field of the Ethernet data packet header, and the dyeing identifier is used to identify the dyed Ethernet data packet. .

The coloring identifier may be a value that is set in advance, and the value may be used to identify the coloring data packet; the embodiment of the present invention does not limit this.

Normally, for an IP packet, since the TTL field in the IP packet is a standard field defined by the protocol, the field is given a larger value (such as 255) when the IP packet is sent, and then the IP. Each time a packet passes through a virtual switch, the TTL value is decremented by 1. When the TTL value is reduced to 0, the IP packet will be discarded. Therefore, in the case of no dyeing, the TTL value of the originating node IP packet is 255, assuming After the virtual switch is passed N times in the middle of the transmission path, the destination node receives the IP packet with a TTL value of 255-N. That is, the TTL value ranges from 255 to 255-N on the transmission path, so when starting After the task is detected, the TTL value of the IP data packet can be set to less than 255-N at the originating node, that is, the value of the dye flag is set to (255-N), and the IP data packet can be distinguished as the dye data packet.

The preset offset field can be set as needed, which is not limited in this embodiment of the present invention. Optionally, the offset field may be pre-determined before step S101 to dye the Ethernet data packet according to the offset field in step S101. For example, if an offset field of 8 bits is specified in advance, when the Ethernet packet is colored, the dye flag is filled at a position shifted by 8 bits from the Ethernet packet header.

S102: Detect and count the number of the dyed data packets in different nodes on the transmission path, where the dyed data packets are the data packets that are stained in the plurality of data packets.

The different nodes on the transmission path may include: a VM, a virtual switch, and the like on the transmission path.

Optionally, the dye bits of the padding identifier may be notified to each node on the transmission path in advance, and the dye bit may be a TTL field or a preset offset field as described in step S101;

Each node may identify the dyed bits of each received data packet according to the notification. If the dyed bit is filled with the dyed mark, the data packet is determined to be a dyed data packet, and the number of the dyed data packets is counted. Otherwise, Then it is determined that the data packet is not dyed. For example, if a node on the transmission path receives a packet, it detects 90 numbers. According to the dyeing mark filled in the dyeing position of the package, it is determined that 90 dyeing packets are received.

S103: Compare differences in the number of dyed data packets detected by different nodes, and determine network quality between different nodes according to the difference.

Optionally, the first VM is used as the first node, and any other node on the transmission path is used as the second node. At this time, if the number of the dyed data packets detected by the first node is greater than the dyed data detected by the second node. The number of packets determines that there is a packet loss problem between the first node and the second node.

For example, in the transmission path shown in FIG. 3(b), the dye data packet detected by each node is as shown in Table 1 below, and the number of dyed IP data packets detected at point P1 is 100, and the number of dyed IP data packets is detected at point P2. For the 100, P3, P4 points, the number of dyed IP packets is 90, which means that 10 IP packets are lost between P2 and P3, indicating that there is IP packet loss in the network between P2 and P3. And exclude the possibility of IP packet loss between P1 and P2, P3 and P4. According to the judgment result, maintenance personnel can quickly locate the problem and take necessary measures to prevent business damage.

Table 1

Statistical point	Number of dyed IP packets
P1	100
P2	100
P3	90
P4	90

In this way, by dyeing the data packet and the difference in the number of dyed data packets between the nodes on the transmission path between the VMs, it is determined that the network has a packet loss problem between the nodes, and the location of the fast demarcation problem is prevented, thereby preventing business damage.

In the actual application, the data packets sent by the VM are sequential, and it is difficult for each node to delimit the start time and the final occurrence time of the data packet, which is prone to the problem of statistical errors in the coloring data packet. Therefore, in the embodiment of the present invention, In order to avoid statistical errors of each node on the transmission path, in a feasible solution of the present invention, the data packets sent by the VM may be divided according to a preset time period, and the data packets in each time period are performed. Dyeing, detection and statistics, the specific implementation is as follows:

The step S101, the dyeing the multiple data packets sent by the first virtual machine VM may specifically include:

The data packets sent by the first VM in multiple consecutive time segments are dyed, the dyeing identifiers of the data packets sent in the same time period are the same, and the dyeing identifiers of the data packets sent in the adjacent time segments are different.

The time period can be divided according to the requirements, which is not limited by the embodiment of the present invention.

Optionally, before the multiple data packets sent by the first virtual machine VM are dyed, a detection task for indicating that the multiple data packets sent by the first VM are to be dyed may be received, where the detection task may include time And segmenting, according to the time period of the received detection task, dividing the time when the first VM sends the data packet into multiple consecutive time segments, and dyeing the data packets in each time segment to make each node on the transmission path The number of dyed data packets in each time period is counted, and the difference in the number of dyed data packets between different nodes in the same time period is compared, and the network quality between different nodes in the time period is determined according to the difference.

For example, if 10 seconds is used as a time period, the TTL field of the IP packet header sent by VM1 within 10 seconds of the first time period may be filled with 249, and within 10 seconds of the second time period, E-mail packet header The TTL field is padded with 248. In the 10 seconds of the third time period, the TTL field of the outgoing IP packet header is filled with 249, and within 10 seconds of the fourth period, the TTL of the IP packet header will be sent. The field is filled with 248 and the subsequent time period is looped in this way.

In this way, the data packets sent by the VM can be divided into time intervals, and the data packets sent by the VM in each time period are dyed, so that each node on the transmission path counts the number of dyed data packets in each time period. And compare the difference in the number of dyed packets between different nodes in the same time period, and judge the quality problem of the network between nodes according to the difference of the number of dyed packets between different nodes in the time period.

As can be seen from the above, the embodiment of the present invention provides a data transmission quality detecting method, which dyes multiple data packets with the same transmission path sent by the first virtual machine VM, and detects and counts the dyed data packets at different nodes on the transmission path. The number of differences is compared with the number of dyed packets detected by different nodes, and the network quality between different nodes is judged according to the difference. In this way, the difference in the number of data packets sent between different nodes can be compared by dyeing, identifying, and counting the data packets, and then the packet loss problem between the nodes can be determined according to the difference, so that corresponding measures can be taken to restore the entire service.

The first embodiment only implements the detection of the packet loss problem during data transmission between VMs. The second embodiment describes the detection scheme of the delay problem during data transmission between VMs.

Embodiment 2

4 is a flowchart of a method for detecting data transmission quality according to an embodiment of the present invention. The method is applicable to an NFV system as shown in FIG. 1 for detecting data transmission quality between different VMs in an NFV system. As shown in FIG. 4, the method may include the following steps:

S201: Dye the first data packet of the plurality of data packets sent by the first virtual machine VM, where the multiple data packets have the same transmission path.

The first VM may be any VM in the VNF under the NFV system, and is an originating node that sends the first data packet.

The first data packet may be an IP data packet, or may be a private Layer 2 data packet, such as an Ethernet data packet, which is not limited in this embodiment of the present invention. The first data packet may be any data packet in a service flow, and the service flow is a service flow delay problem, a service flow to be quality-detected, or a service flow specified for the user, and may include a large amount of data. package.

For example, before the step S201, the quality of the multiple service flows sent by the first VM may be detected. If the quality of the first service flow is found, the first data in the first service flow sent by the first VM is detected. The package is subjected to dye detection;

Alternatively, before the step S201, the quintuple information sent by the user is received, and the first data packet in the first service flow sent by the first VM is dyed according to the received quintuple information, wherein the quintuple information is The quintuple information may include: a source IP address, a destination IP address, a protocol type, a source port number, and a destination port number of the service flow.

Usually, a service flow has a specific transmission path along which data packets in the service flow are transmitted. In the NFV system, the transmission path may be: a path from the first VM through the first virtual switch to the second VM, where the second VM, the first virtual switch, and the first VM may be in the same host;

The path may be a path from the first VM to the second virtual switch, and then from the second virtual switch to the second VM, where the first VM and the first virtual switch are located in the first host, and the second VM and the second The second virtual switch may be located in the second host, and the first host and the second host may be in the same VNF or in different VNFs;

The second VM may also be located in the VNF of the NFV system; the first virtual switch and the second virtual switch It can be located in a virtual network in the NFVI of the NFV system.

It can be understood that although the above describes only the transmission path between two VMs, the transmission path between two or more VMs may also be set by referring to the foregoing manner, that is, the VMs in the same host pass through the virtual in the host. The switch transmits data packets, and the VMs in different hosts transmit data packets through the virtual switches in the respective hosts, and details are not described herein.

Wherein, in the process of dyeing the first data packet, the dyeing bits are different for different types of data packets. Optionally, when the first data packet is an IP data packet, according to a special format of the IP data packet, the first coloring identifier may be filled in a field of an IP data packet header (Time To Live, TTL for short). The first coloring identifier is used to identify the first data packet.

When the first data packet is an Ethernet data packet, according to a special format of the Ethernet data packet, the first coloring identifier may be filled in a preset offset field of the Ethernet data packet header, and the dyeing identifier is used to identify the first data. package.

The coloring identifier may be a value that is set in advance, and the value may identify the first coloring data packet, which is not limited by the embodiment of the present invention.

Normally, for an IP packet, since the TTL field in the IP packet is a standard field defined by the protocol, the field is given a larger value (such as 255) when the IP packet is sent, and then the IP. Each time a packet passes through a virtual switch, the TTL value is decremented by 1. When the TTL value is reduced to 0, the IP packet will be discarded. Therefore, in the case of no dyeing, the TTL value of the originating node IP packet is 255, assuming After the virtual switch is passed N times in the middle of the transmission path, the destination node receives the IP packet with a TTL value of 255-N. That is, the TTL value ranges from 255 to 255-N on the transmission path, so when starting After the task is detected, the TTL value of the IP data packet can be set to less than 255-N at the originating node, that is, the value of the dye flag is set to (255-N), and the IP data packet can be distinguished as the dye data packet.

The preset offset field can be set as needed, which is not limited in this embodiment of the present invention. Optionally, the offset field may be pre-determined before step S201 to dye the Ethernet data packet according to the offset field in step S201.

It should be noted that, in the embodiment of the present invention, only the first data packet sent by the first VM may be dyed, and other data packets may be used in order to facilitate the subsequent node to distinguish the first data packet from the large number of data packets. The dyeing process may be performed on a plurality of data packets including the first data packet without performing the dyeing process, except that the dyeing identifier of the first data packet is different from the dyeing identifier of the other data packets, and the first data packet is not included. The dyeing identifier of the data packet can be identical.

S202: The time stamp of the first coloring data packet is detected and counted by different nodes on the transmission path, and the first coloring data packet is the first data packet that is stained.

The different nodes on the transmission path may include: a VM, a virtual switch, and the like on the transmission path.

Optionally, the dyeing bit of the padding identifier may be notified to each node on the transmission path in advance, and the dye bit may be a TTL field or a preset offset field as described in step S201;

Each node may identify the dyed bits of each received data packet according to the notification. If the dyeing bit of a certain data packet is filled with a dyed identifier for identifying the first data packet, the data packet is determined to be the first A stained packet and the time stamp of the first stained packet is counted.

S203: Compare differences in time stamps of the first dyed data packets detected by different nodes, and determine network quality between different nodes according to the differences.

Optionally, if the difference between the time stamps of the first dyed data packet detected between the first node and the second node is greater than the third section Determining, by the difference between the time stamp of the first dyed data packet and the fourth node, the transmission delay between the first node and the second node is greater than the third node and the fourth Transmission delay between nodes.

For example, in the transmission path shown in FIG. 3(b), the time stamp of the first coloring data packet detected by each node is as shown in Table 2, and the time stamp of the first coloring data packet detected by the point P1 is T1, P2. The time stamp of the first dyed data packet is detected as T2, the time stamp of the first dyed data packet is detected as T3 at the point P3, and the time stamp of the first dyed data packet is detected as T4 at the point P4, and at this time, if T3- If the time difference of T2 is greater than the time difference of T2-T1, it is determined that there is a transmission delay problem between P2 and P3. According to the judgment result, the maintenance personnel can quickly locate the problem and take necessary measures to prevent business damage.

Table 2

Statistical point	Time stamp of the first stained packet
P1	T1
P2	T2
P3	T3
P4	T4

In this way, the difference between the time stamps of the data packets between the nodes on the transmission path between the VMs can be determined by dyeing the data packets, and the delay time of the network between the nodes is determined, and the location of the fast delimitation problem is prevented to prevent business damage. .

In the actual application, the data packets sent by the VM are sequential, and it is difficult for each node to delimit the start time and the final occurrence time of the data packet, which is prone to the problem of statistical errors in the coloring data packet. Therefore, in the embodiment of the present invention, In order to avoid the statistical error of each node on the transmission path, in a feasible solution of the present invention, the coloring the first data packet of the plurality of data packets sent by the first virtual machine VM may include:

Data packets sent by the first VM in a plurality of consecutive time periods are dyed, wherein the first data packet sent in the same time period is dyed with the first coloring identifier, and the first data packet sent in the same time period except the first data packet All other data packets are dyed with the second staining identifier, and the first data packet sent in the adjacent time period has different dyeing identifiers.

The first data packet may be the first data packet sent in each time period. In addition, it should be noted that the foregoing time period may be divided according to requirements, which is not limited by the embodiment of the present invention; and the second dyeing identifiers used in the adjacent time periods may be the same or different, and the embodiment of the present invention Nor is it limited.

Optionally, before the first data packet of the plurality of data packets sent by the first virtual machine VM is dyed, a detection task for indicating that the multiple data packets sent by the first VM are to be dyed may be received. The detecting task may include a time period, and may further divide the time when the first VM sends the data packet into multiple consecutive time segments according to the time period in the received detecting task, and dye the data packet in each time segment. Having each node on the transmission path count the time stamp of the first coloring data packet in each time period, and compare the difference of the first coloring data packet time stamp between different nodes in the same time period, and determine different time periods according to the difference. Network quality between nodes.

For example, if 10 seconds is used as a time period, the TTL field of the first IP packet header sent by VM1 within 10 seconds of the first time period can be filled with 247, and 10 seconds in the second time period. Inside, the TTL field of the first IP packet header to be sent is filled with 246, and within 10 seconds of the third time period, the TTL field of the first IP packet header to be sent is filled with 247, in the fourth Within 10 seconds of the cycle, the TTL field of the first IP packet header to be sent is filled with 246, and the subsequent time period is cyclically set in this manner.

In this way, the data packets sent by the VM can be divided into time intervals, and the VM is sent in each time period. The first data packet is dyed, and each node on the transmission path counts the time stamp of the first coloring data packet in each time period, and compares the difference of the first coloring data packet time stamp between different nodes in the same time period. The delay time of the network between nodes is determined according to the difference of the first coloring packet time stamp between different nodes in any time period.

As can be seen from the above, the embodiment of the present invention provides a data transmission quality detecting method, which performs coloring on a first data packet of multiple data packets with the same transmission path sent by the first virtual machine VM, and detects different nodes on the transmission path. The number of the first dyed data packets is counted, and the difference of the time stamps of the first dyed data packets detected by different nodes is compared, and the network quality between different nodes is judged according to the difference. In this way, the difference between the time stamps of the data packets sent between different nodes can be compared by dyeing, identifying and counting the data packets, and then the delay between the nodes can be determined according to the difference, so that corresponding measures can be taken to restore the entire service.

The first embodiment and the second embodiment respectively implement the detection of the packet loss and the delay in the network. It can be understood that, in another feasible solution of the embodiment of the present invention, the first embodiment and the second embodiment can be used. The invention combines the detection of network packet loss and delay, and the present invention will not be described in detail herein.

For example, the first data packet of the first data packet sent by the first VM may be firstly dyed, and the other data packets may be secondly colored, and each node on the transmission path may detect and count the number of all the dyed data packets. And the timestamp of the first dyed packet, comparing the difference between the number of dyed packets between different nodes, and the difference of the first packetized timestamp, based on these two differences to determine the difference between the different nodes Loss and delay issues.

Or, in order to avoid the time-out of the data packets sent by the VM in the actual application, it is difficult for each node to delimit the start time and the final occurrence time of the data packet, and the problem of the statistics of the dyed data packet is easy to occur, and the first VM can be used. The sent data packet is divided according to a preset time period, and the data packet sent by the first VM in multiple consecutive time periods is dyed, wherein the first data packet sent in the same time period is dyed by the first dyeing identifier, and is sent in the same time period. All the data packets except the first data packet are dyed with the second dyeing identifier, and the first dyeing identifier used in the adjacent time period is different, and the second dyeing identifier used in the adjacent time period may be the same or different. . Wherein, the first dyeing identifier is different from the second dyeing identifier, so that the first coloring data packet in each time period and all the coloring data packets can be identified, according to the time stamp of the first coloring data packet, and the coloring data packet. The number simultaneously detects two problems of network delay and packet loss.

For example, if 10 seconds is used as a time period, the TTL field of the first IP packet header sent by VM1 within 10 seconds of the first time period may be filled with 247, and the TTL field of the remaining IP data packet header is filled. 249, in the 10 seconds of the second time period, the TTL field of the first IP packet header is filled with 246, and the TTL field of the remaining IP packet header is filled with 248, in the third time period. Within 10 seconds, the TTL field of the first IP packet header to be sent is filled with 247, the TTL field of the remaining IP packet header is filled with 249, and the first IP will be sent within 10 seconds of the fourth period. The TTL field of the packet header is filled with 246, and the TTL field of the remaining IP packet header is filled with 248, and the subsequent time period is cyclically set in this manner.

According to an embodiment of the present invention, the following embodiments further provide a data transmission quality detecting apparatus 30, which is preferably used to perform the data transmission quality detecting method according to the first embodiment and the second embodiment.

Embodiment 3

FIG. 5 is a structural diagram of a data transmission quality detecting apparatus 30 according to an embodiment of the present invention. As shown in FIG. 5, the apparatus 30 may include: a coloring module 301, a statistics module 302, and a comparison determining module 303.

The data transmission quality detecting apparatus 30 shown in FIG. 5 may be separately deployed in the NFV system, or each module of the data transmission quality detecting apparatus may be dispersed in each functional component of the NFV system, as shown in FIG. 5A. The schematic diagram of the NFV system architecture of the transmission quality detecting device 30 is as shown in FIG. 5:

The coloring module 301 is deployed in each VM in the NFV system VNF entity, and the statistics module 302 is deployed in each node on the data transmission path (eg, in each VM in the VNF entity, in the virtual switch of the virtual network) The comparison judging module 303 can be deployed in the NFV system as a separate virtual machine VM, and can be deployed as a functional component in the VNFM or EM of the NFV system. It should be noted that, for convenience of description, the comparison judging module 303 is deployed in the VNFM. The inside is explained as an example.

Typically, the staining module 301 performs a staining function within the VM from which the data originated. Specifically, the dyeing module 301 is configured to perform staining on multiple data packets sent by the first VM, where the multiple data packets have the same transmission path.

The statistics module 302 is configured to detect and count the number of dyed data packets on the node where the data is located, and the stained data packet is the data packet that is stained in the plurality of data packets.

The comparison judging module 303 is configured to compare differences in the number of dyed data packets detected by the statistical module 302 in different nodes, and determine network quality between different nodes according to the difference.

Optionally, when the data packet is an IP data packet, the coloring module 301 may be specifically configured to:

The coloring identifier is filled in the TTL field of the IP packet header, and the coloring identifier is used to identify the dyed IP packet.

When the data packet is an Ethernet data packet, the coloring module 301 is specifically configured to:

A dyed identifier is filled in the preset offset field of the Ethernet packet header, and the dyed identifier is used to identify the stained Ethernet packet.

Optionally, the comparison determining module 303 is specifically configured to:

Actively collect the number of dyed data packets counted by each node, or receive the number of dyed data packets reported by each node;

If the number of the dyed data packets detected by the first node is smaller than the number of the dyed data packets detected by the second node, determining that there is a packet loss problem between the first node and the second node; wherein the first node may be the first VM, the second node can be any other node on the transmission path.

It can be understood that when the node is a virtual switch, since the virtual switch is located in the virtual network of the NFV system, the comparison judgment module 303 cannot be directly communicated. Therefore, the virtual switch needs to collect the self-sampling dye data packet through the VIM in the NFV system. The quantity is forwarded to the comparison determination module 303.

Optionally, as shown in FIG. 5, the data transmission quality detecting apparatus may further include a data forwarding module 304. The data forwarding module 304 can be deployed in the VIM as shown in FIG. 5A.

The data forwarding module 304 is configured to receive the number of the dyed data packets counted by the virtual switch, and forward the received number of the dyed data packets to the comparison determining module 303.

In the actual application, the data packets sent by the VM are sequential, and it is difficult for each node to delimit the start time and the final occurrence time of the data packet, which is prone to the problem of statistical errors in the coloring data packet. Therefore, in the embodiment of the present invention, In a feasible solution of the present invention, the dyeing module 301 can be specifically used to:

The data packets sent by the first VM in a plurality of consecutive time segments are dyed, wherein the data packets sent in the same time period have the same dyeing identifier, and the data packets sent in the adjacent time segments have different dyeing identifiers.

In this way, the difference in the number of data packets sent between different nodes can be compared by dyeing, identifying, and counting the data packets, and then the packet loss problem between the nodes can be determined according to the difference, so that corresponding measures can be taken to restore the entire service.

In addition, in another feasible solution of the embodiment of the present invention, the apparatus provided in FIG. 5 can also be used to implement detection of a transmission delay problem between nodes, and the specific implementation is as follows:

The dyeing module 301 is further configured to perform staining on the first data packet of the plurality of data packets sent by the first virtual machine VM, where the plurality of data packets have the same transmission path.

The statistics module 302 is further configured to detect and count the time stamp of the first coloring data packet received by the node where the node is located, and the first coloring data packet is the first data packet that is dyed.

The comparison judging module 303 is further configured to compare the difference of the time stamps of the first dyed data packets detected by the statistics module 302 in different nodes, and determine the network quality between the different nodes according to the difference.

The dyeing module 301 can be used to: when the first data packet is an IP data packet, the

The first coloring identifier is filled in the time-to-live TTL field of the IP packet header, and the first coloring identifier is used to identify the first data packet.

When the first data packet is an Ethernet data packet, the coloring module 301 can be specifically used to:

Filling a first coloring identifier in a preset offset field of the first Ethernet packet header, the first coloring identifier is used to identify the first data packet.

Optionally, the comparison determining module 303 is specifically configured to:

If the difference between the timestamps of the first coloring data packet detected between the first node and the second node is greater than the difference between the timestamps of the first coloring data packet detected between the third node and the fourth node, The transmission delay between the node and the second node is greater than the transmission delay between the third node and the fourth node.

Optionally, the dyeing module 301 can be specifically configured to:

Data packets sent by the first VM in a plurality of consecutive time periods are dyed, wherein the first data packet sent in the same time period is dyed with the first coloring identifier, and the first data packet sent in the same time period except the first data packet All other data packets are dyed by the second coloring identifier, and the dyeing identifiers of the first data packets sent in the adjacent time segments are different, and the dyeing identifiers of all the data packets sent by the adjacent time segments except the first data packet are different. .

In this way, the difference between the time stamps of the data packets sent between different nodes can be compared by dyeing, identifying and counting the data packets, and then the delay between the nodes can be determined according to the difference, so that corresponding measures can be taken to restore the entire service.

The detection process shown in FIG. 5 is used to detect the transmission quality of the first service flow from the VM1 through the virtual switch 1 to the VM2, and the execution process of each module (shown in FIG. 6) is specifically described.

S1: The comparison judgment module 303 starts the detection task, and sends the detection task to the VM1, the virtual switch, and the VM2.

The detecting task is used to instruct each node device to perform data transmission quality detection, and the detecting task includes: quintuple information of the first service flow, and a time period.

Optionally, when the node is a VM, the comparison determination module 303 sends the detection task to the node. When the node is a virtual switch, as shown in step S1.1 in FIG. 6, the comparison determination module 303 sends the detection task. To the data forwarding module 304, the data forwarding module 304 forwards the detection task to the virtual switch.

S2: After receiving the detection task, the dyeing module 301 in the VM1 periodically dyes the data packet in the first service flow sent according to the time period.

For example, when the data packet is an IP data packet, the dyeing identifier is filled in its TTL field; when the data packet is an Ethernet data packet, the dyeing identifier is filled in its preset offset field.

Optionally, the first data packet in each period may be subjected to a first coloring, and the other data packets may be subjected to a second coloring, and the dyeing identifier of the first data packet is different from other dyeing identifiers in the time period. The dyeing identifier of the first data packet between adjacent time segments is different, and the dyeing periods of other data packets between adjacent time segments are also different.

S3: The statistics module 302 in VM1 detects and counts the number of dye packets sent by VM1 in each time period, and the time stamp of the first dyed data packet.

The first dyed data packet is the first data packet to be dyed in each cycle.

S4: After receiving the detection task, the statistics module 302 in the VM2 detects and counts the number of the dyed data packets sent by the virtual switch in each time period and the time stamp of the first coloring data packet.

S5: After receiving the detection task, the statistics module 302 in the virtual switch detects and counts the number of dye packets sent by VM1 in each time period and the time stamp of the first dyed data packet.

S6: The comparison judgment module 303 acquires a request message of the statistical data to each node.

The request message is used to request the number of the dyed data packets acquired by each node and the first coloring data packet time stamp.

Optionally, when the node is a virtual switch, as shown in step S6.1 in FIG. 6, the comparison determining module 303 sends a request message to the data forwarding module 304, and the data forwarding module 304 forwards the request message to the virtual switch.

S7: After receiving the request message, the statistic module 302 of the VM1 sends the number of the swatched data packets and the first smeared data packet timestamp to the comparison judgment module 303.

S8: After receiving the request message, the statistics module 302 in the virtual switch sends the number of the dyed data packets and the first coloring data packet time stamp of the statistics to the data forwarding module 304, and the data forwarding module 304 sends the coloring data packet to the comparison determining module 303. The number and time stamp of the first stained packet.

It should be noted that, because the virtual switch is in the virtual network in the NFV system, and the comparison judgment module 303 cannot directly communicate with each other, in the embodiment of the present invention, the statistics module 302 in the virtual switch needs to pass through the NFV system. The VIM communicates with the comparison determination module 303.

Optionally, the statistics module 302 in the virtual switch may send the statistics of the number of the dyed data packets and the first coloring data packet to the comparison judgment module 303 through the data forwarding module 304 in the VIM.

S9: After receiving the request message, the statistic module 302 in the VM2 sends the number of the swatched data packets and the first smeared data packet timestamp to the comparison judgment module 303.

S10: The comparison judging module 303 compares the difference between the number of dyed data packets counted between different nodes and the time stamp of the first dyed data packet, and determines the packet loss and delay of the inter-node network.

Optionally, the comparison determining module 303 can compare the difference in the number of the dyed data packets between different nodes, and determine, according to the difference, the packet loss problem occurs in the network between the nodes;

Comparing the difference of the time stamps of the first dyed data packets between different nodes, and determining the delay time of the network between the nodes according to the difference.

S11: The comparison judging module 303 transmits a notification message of stopping detection to each node. Optionally, when the node is a virtual switch, as shown in step S11.1 in FIG. 6, the comparison determining module 303 sends a notification message that stops detecting to the data forwarding module 304, and the data forwarding module 304 sends the notification message to the Virtual switch.

The notification message is used to notify each node to stop the detection of the data transmission quality.

It can be understood that the process shown in FIG. 6 only describes the detection of data transmitted between VM1 and VM2. In an actual application environment, the transmission path may exist through more than two VMs and multiple virtual switches. In this case, the specific processing of each module in the data transmission detecting device and the process shown in FIG. Similar, it will not be described in detail here.

As can be seen from the above, the embodiment of the present invention provides a data transmission quality detecting apparatus, which performs different coloring on a plurality of data packets and a first data packet having the same transmission path sent by the first virtual machine VM, and different in the transmission path. The node detects and counts the number of all dyed data packets and the time stamp of the first dyed data packet, compares the number of dyed data packets detected by different nodes, and the difference of the timestamp of the first colored data packet, and judges the loss between different nodes according to the difference. Package and latency issues. In this way, by comparing the data packet, the identification and the statistics, the difference between the number of data packets sent by different nodes and the time stamp can be compared, and then the packet loss delay between the nodes can be determined according to the difference, so that corresponding measures can be taken to recover. The entire business.

A person skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working process of the above-mentioned unit and device can refer to the corresponding process in the foregoing method embodiment, and details are not described herein again.

In the several embodiments provided in the present application, it should be understood that the disclosed apparatus, apparatus, and method may be implemented in other manners. For example, the device embodiments described above are merely illustrative. For example, the division of the unit is only a logical function division. In actual implementation, there may be another division manner, for example, multiple units or components may be combined or It can be integrated into another device, or some features can be ignored or not executed.

The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may be physically separate, or two or more units may be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of hardware plus software functional units.

The above-described integrated unit implemented in the form of a software functional unit can be stored in a computer readable storage medium. The software functional units described above are stored in a storage medium and include instructions for causing a computer device (which may be a personal computer, server, or network device, etc.) to perform portions of the steps of various embodiments of the present invention. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and the like, and the program code can be stored. Medium.

A person of ordinary skill in the art may understand that all or part of the steps of the foregoing embodiments may be completed by a program to instruct related hardware (for example, a processor), and the program may be stored in a computer readable storage medium. The storage medium may include a read only memory, a random access memory, a magnetic disk or an optical disk, or the like.

It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and are not limited thereto; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that The technical solutions described in the foregoing embodiments are modified, or the equivalents of the technical features are replaced. The modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (24)

1. A data transmission quality detection method is applied to a network function virtualization NFV system, characterized in that the method comprises:

   Sampling a plurality of data packets sent by the first virtual machine VM, the multiple data packets having the same transmission path;

   Detecting and counting the number of dyed data packets at different nodes on the transmission path, the dyed data packets being data packets dyed in the plurality of data packets;

   Comparing the differences in the number of dye packets detected by different nodes, and determining the network quality between the different nodes according to the difference.
2. The method according to claim 1, wherein the data packet is an IP data packet, and the plurality of data packets sent by the first virtual machine VM are dyed, including:

   A dyeing identifier is filled in the time-to-live TTL field of the IP packet header, and the dye identifier is used to identify the dyed IP packet.
3. The method according to claim 1, wherein the data packet is an Ethernet data packet, and the plurality of data packets sent by the first virtual machine VM are dyed, including:

   A dyeing identifier is filled in a preset offset field of the Ethernet packet header for identifying the stained Ethernet packet.
4. The method according to any one of claims 1-3, wherein the coloring the plurality of data packets sent by the first virtual machine VM comprises:

   The first VM sends a data packet in a plurality of consecutive time periods, wherein the data packets sent by the same time period have the same dyeing identifier, and the data packets sent by the adjacent time segments have different dyeing identifiers.
5. A method according to any one of claims 1 to 4, characterized in that

   The transmission path is: a path from the first VM to the second VM through the first virtual switch, the second VM, the first virtual switch, and the first VM are in the same host;

   Or the transmission path is: a path from the first VM to the second virtual switch, the second virtual switch to the second VM, the first VM and the first virtual The switch is located at the first host, and the second VM and the second virtual switch are located at the second host;

   The first VM and the second VM are located in a virtual network function VNF entity in the NFV system;

   The first virtual switch and the second virtual switch are located in a virtual network of the network function virtualization infrastructure NFVI in the NFV system.
6. The method according to any one of claims 1 to 5, wherein the comparing the difference in the number of dyed data packets detected between different nodes, and determining the network quality between the different nodes according to the difference, including :

   If the number of the dyed data packets detected by the first node is greater than the number of the dyed data packets detected by the second node, determining that there is a packet loss problem between the first node and the second node;

   The first node is the first VM, and the second node is any node except the first VM of the transmission path.
7. A data transmission quality detection method is applied to a network function virtualization NFV system, characterized in that the method comprises:

   Sampling a first one of the plurality of data packets sent by the first virtual machine VM, the plurality of data packets having phases Same transmission path;

   Detecting and counting time stamps of the first dyed data packet at different nodes on the transmission path, the first dyed data packet being the first data packet being dyed;

   Comparing the difference of time stamps of the first dyed data packets detected by different nodes, and determining the network quality between the different nodes according to the difference.
8. The method according to claim 7, wherein the first data packet is an IP data packet, and the first data packet of the plurality of data packets sent by the first virtual machine VM is dyed, including:

   Filling a first coloring identifier in a time-to-live TTL field of the IP packet header, the first coloring identifier is used to identify the first data packet.
9. The method according to claim 7, wherein the first data packet is an Ethernet data packet, and the first data packet of the plurality of data packets sent by the first virtual machine VM is dyed, including:

   Filling a first coloring identifier in a preset offset field of the first Ethernet packet header, the first coloring identifier is used to identify the first data packet.
10. The method according to any one of claims 7-9, wherein the coloring the first data packet of the plurality of data packets sent by the first virtual machine VM comprises:

    The first VM sends a data packet in a plurality of consecutive time periods, wherein the first data packet sent in the same time period is dyed with the first coloring identifier, and all but the first data packet sent in the same time period The data packet is dyed with the second staining identifier, and the dyeing identifier of the first data packet sent in the adjacent time period is different.
11. A method according to any one of claims 7 to 10, wherein

    The transmission path is: a path from the first VM to the second VM through the first virtual switch, the second VM, the first virtual switch, and the first VM are in the same host;

    Or the transmission path is: a path from the first VM to the second virtual switch, the second virtual switch to the second VM, the first VM and the first virtual The switch is located at the first host, and the second VM and the second virtual switch are located at the second host;

    The first VM and the second VM are located in a virtual network function VNF entity in the NFV system;

    The first virtual switch and the second virtual switch are located in a virtual network of the network function virtualization infrastructure NFVI in the NFV system.
12. The method according to any one of claims 7 to 11, wherein the comparing the difference of time stamps of the first dyed data packets detected by different nodes, determining the difference between the different nodes according to the difference Network quality, including:

    If the difference between the time stamps of the first coloring data packets detected between the first node and the second node is greater than the difference between the time stamps of the first coloring data packets detected between the third node and the fourth node And determining, that the transmission delay between the first node and the second node is greater than a transmission delay between the third node and the fourth node.
13. A data transmission quality detecting device, comprising: a dyeing module, a statistics module, and a comparison judging module;

    The coloring module is located in a first virtual machine VM in the network function virtualization NFV system, and the coloring module is configured to perform coloring on multiple data packets sent by the first VM, where the multiple data packets have the same Transmission path

    The statistic module is located in different nodes on the transmission path, and the statistic module is configured to detect and count the number of dye data packets received by the node where the node is located, where the dye data packet is in the multiple data packets. Stained data packet;

    The comparison judging module is configured to compare differences in the number of dye packets detected by the statistical module in different nodes, Determining the network quality between the different nodes according to the difference.
14. The device according to claim 13, wherein the data packet is an IP data packet, and the coloring module is specifically configured to:

    A dyeing identifier is filled in the time-to-live TTL field of the IP packet header, and the dye identifier is used to identify the dyed IP packet.
15. The device according to claim 13, wherein the data packet is an Ethernet data packet, and the coloring module is specifically configured to:

    A dyeing identifier is filled in a preset offset field of the Ethernet packet header for identifying the stained Ethernet packet.
16. The device according to any one of claims 13 to 15, wherein the dyeing module is specifically configured to:

    The data packets sent by the first VM in a plurality of consecutive time periods are dyed, wherein the data packets sent by the same time period have the same dyeing identifier, and the data packets sent by the adjacent time segments have different dyeing identifiers.
17. Apparatus according to any of claims 13-16, wherein

    The transmission path is: a path from the first VM to the second VM through the first virtual switch, the second VM, the first virtual switch, and the first VM are in the same host;

    Or the transmission path is: a path from the first VM to the second virtual switch, the second virtual switch to the second VM, the first VM and the first virtual The switch is located at the first host, and the second VM and the second virtual switch are located at the second host;

    The first VM and the second VM are located in a virtual network function VNF entity in the NFV system;

    The first virtual switch and the second virtual switch are located in a virtual network of the network function virtualization infrastructure NFVI in the NFV system.
18. The device according to any one of claims 13-17, wherein the comparison determining module is specifically configured to:

    If the number of the dyed data packets detected by the first node is greater than the number of the dyed data packets detected by the second node, determining that there is a packet loss problem between the first node and the second node;

    The first node is the first VM, and the second node is any node except the first VM of the transmission path.
19. A data transmission quality detecting device, comprising: a dyeing module, a statistics module, and a comparison judging module;

    The coloring module is located in a first virtual machine VM in the network function virtualization NFV system, and the coloring module is configured to perform coloring on the first data packet in the plurality of data packets sent by the first virtual machine VM, where the multiple Packets have the same transmission path;

    The statistic module is located in a different node on the transmission path, and the statistic module is configured to detect and count a time stamp of the first dyed data packet received by the node where the node is located, where the first coloring data packet is Dyeing the first data packet;

    The comparison judging module is configured to compare differences in time stamps of the first dyed data packets detected by the statistics modules in different nodes, and determine network quality between the different nodes according to the differences.
20. The device according to claim 19, wherein the first data packet is an IP data packet, and the coloring module is specifically configured to:

    Filling a first dye identifier in a time-to-live TTL field of the IP packet header, where the first dye identifier is used The first data packet is identified.
21. The device according to claim 19, wherein the first data packet is an Ethernet data packet, and the coloring module is specifically configured to:

    Filling a first coloring identifier in a preset offset field of the first Ethernet packet header, the first coloring identifier is used to identify the first data packet.
22. The device according to any one of claims 19 to 21, wherein the dyeing module is specifically configured to:

    And filtering the data packet sent by the first VM in a plurality of consecutive time periods, wherein the first data packet sent in the same time period is colored by the first coloring identifier, and the first data packet sent in the same time period is All other data packets are dyed with the second staining identifier, and the dyeing identifier of the first data packet sent in the adjacent time period is different.
23. A device according to any one of claims 19-22, wherein

    The transmission path is: a path from the first VM to the second VM through the first virtual switch, the second VM, the first virtual switch, and the first VM are in the same host;

    Or the transmission path is: a path from the first VM to the second virtual switch, the second virtual switch to the second VM, the first VM and the first virtual The switch is located at the first host, and the second VM and the second virtual switch are located at the second host;

    The first VM and the second VM are located in a virtual network function VNF entity in the NFV system;

    The first virtual switch and the second virtual switch are located in a virtual network of the network function virtualization infrastructure NFVI in the NFV system.
24. The device according to any one of claims 19 to 23, wherein the comparison determining module is specifically configured to:

    If the difference between the time stamps of the first coloring data packets detected between the first node and the second node is greater than the difference between the time stamps of the first coloring data packets detected between the third node and the fourth node And determining, that the transmission delay between the first node and the second node is greater than a transmission delay between the third node and the fourth node.

Images