Large-scale virtual-real combined network flow generating device and network testing method

Abstract

The invention provides a large-scale virtuality and reality combined network flow generating device and a network testing method. The large-scale virtuality and reality combined network flow generating device comprises a high-performance computing platform and a physical switching unit, wherein the high-performance computing platform comprises a service controller module, a flow generator cluster and a virtual switch set; the service controller module comprises a service modeling unit, a service control unit, a result evaluation unit, a network configuration unit and a container management unit; the flow generator cluster comprises N flow generation engines; the virtual switch unit comprises a virtual control switch and a plurality of virtual data switches; the physical switch group comprises a plurality of physical switches. According to the large-scale virtuality and reality combined network flow generation device, centralized control and data summarization analysis are realized through the service controller module, and performance indexes and comparative analysis of a plurality of tested links can be processed at the same time.

Description

Technical Field

[0001] The present invention relates to the field of network communication, and in particular to a large-scale virtual-real combined network traffic generating device and a network testing method. Background Art

[0002] A network traffic generator is a device that generates standard network traffic that conforms to the rules described by the user. The device also supports functions such as sending and receiving network traffic, recording sending and receiving statistics, data processing and test result evaluation. The network traffic generator is suitable for network testing and evaluation of various structures such as point-to-point, star network, mesh network, etc. During the test, the network traffic generator is connected to the network device under test, and the device generates network traffic and injects it into the network under test, transmits it through the network under test, records the statistical data such as packet timestamp, packet length, packet sequence number, sending and receiving address generated during the transmission process, and calculates and evaluates the performance indicators such as the bandwidth, delay and packet loss rate of the network under test. The current main technical solution is a network traffic generator based on a general computing platform and traffic generator software. Taking the flow generator based on iperf software as an example, the network traffic generator and the network device under test are physically connected one-to-one, and the network bandwidth of the device is fully matched with the network device under test. For large-scale network testing scenarios, multiple computing platforms are required to form a cluster, and multiple device data network ports are directly connected to multiple network devices under test. During the test, the user operates each network traffic generating device one by one; after the test, the user obtains the test evaluation results of each network traffic one by one.

[0003] The network traffic generating devices of the prior art have the following problems when applied in large-scale network testing scenarios. Large-scale network testing scenarios require multiple data streams to be transmitted in parallel in different test nodes, and the test data of all test nodes are summarized and analyzed after the test is completed. The prior art does not have the ability of centralized and unified control and data summary and analysis; in large-scale network testing scenarios, multiple computing platforms of the prior art form a cluster, and the time synchronization accuracy between computing platforms cannot meet the sub-millisecond one-way delay index test; in large-scale network testing scenarios, the management of multi-computing platform clusters is complex, and network deployment and implementation are complex; for large-scale multi-node low-speed networks, a single device is used as a dedicated traffic generating device for a single device under test, which can neither fully utilize the processing power of the computing platform nor the network bandwidth of the computing platform. This application proposal mainly solves the problems of centralized and unified control and data summary and analysis of devices in large-scale network scenarios, low cluster time synchronization accuracy, complex cluster management, complex real-time network deployment, and insufficient utilization of device processing power and network resources. Summary of the invention

[0004] In view of the defects in the prior art, the present invention provides a large-scale virtual-real combined network traffic generating device and a network testing method.

[0005] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0006] In one aspect, the present invention provides a large-scale virtual combined network traffic generation device, including a high-performance computing platform and a physical switch group;

[0007] The high-performance computing platform includes a business controller module, a traffic generator cluster, and a virtual switch group;

[0008] The service controller module, traffic generator cluster, virtual switch group and physical switch group are interconnected through the network to form a control network, providing control ports to the outside world;

[0009] The traffic generator cluster, virtual switch group and physical switch group are interconnected through the network to form a data network and provide data ports to the outside world;

[0010] The control port of the service controller module is connected to the corresponding port of the virtual switch group to form a control network between the service controller module and the virtual switch group, and the control network provides a control entry of the high-performance computing platform for the user control terminal;

[0011] The traffic generator cluster adopts containerized deployment mode, including N flow generator engine containers, corresponding to N devices to be tested, and the time between each container is completely synchronized with the host time; each flow generator engine container has two logical ports, one of which is a data port and the other is a control port; the control port and data port of each flow generator engine container are respectively connected to the corresponding ports of the virtual switch group;

[0012] The virtual switch group consists of a virtual control switch and multiple virtual data switches; the physical control network port of the high-performance computing platform is used to interconnect with the external user control terminal, the virtual control switch mounts the physical control network port of the high-performance computing platform, and provides N+1 logical ports, which correspond to the connection of N flow generation engine containers and 1 business controller module respectively; the physical data network port of the high-performance computing platform is interconnected with the network port of the physical switch group, the virtual data switch mounts the physical data network port of the high-performance computing platform, and provides N logical ports, which are interconnected with N flow generator engine containers respectively; the virtual data switch and the physical switch are divided into N virtual subnets to realize a one-to-one binding relationship between the flow generator engine container and the physical switch port, the subnet ID of each port of the physical switch is consistent with the subnet ID of each port of the virtual data switch, to realize a one-to-one mapping between the flow generator engine container and the tested device, and ensure that network traffic will not be transmitted inside the virtual data switch.

[0013] Furthermore, the flow generator engine container is integrated with network traffic generation software. The flow generator engine container supports parsing business flow scripts to generate services, supports the generated business flow rate to comply with specified statistical distribution rules, supports recording business sending and receiving logs, and provides a control management interface.

[0014] Further, the service controller module includes a service modeling unit, a service control unit, a result evaluation unit, a network configuration unit and a container management unit;

[0015] The container management unit implements the creation, destruction, start, stop, parameter configuration and status monitoring of the flow generation engine container by calling the container control interface, and provides users with a visual container management operation and monitoring page;

[0016] The network configuration unit creates networks, allocates IP addresses, and divides virtual subnets for the flow generation engine container, virtual switch group, and physical switch group, establishes data networks and control networks, and provides users with visual network configuration and network monitoring pages;

[0017] The business modeling unit includes business model management. The business model is a description file that describes the flow rate, packet size, duration, and distribution model parameters. The business modeling unit is responsible for parsing the description file and mapping it to the business script file supported by the flow generation engine container. The business modeling unit provides users with operations for creating, editing, and deleting business models through a visual page.

[0018] The business control unit is responsible for loading business flow scripts, starting, stopping, preprocessing results, and reporting results on the flow generator engine container, and provides users with visual business operations and monitoring pages;

[0019] The result evaluation unit summarizes the preprocessing results, calculates the rate, throughput, delay and packet loss rate indicators of each service flow in the tested network, and provides users with viewing and export operation pages in the form of visual icons and data.

[0020] Furthermore, the number of virtual data switches is equal to the number n of physical data network ports of the high performance computing platform.

[0021] Furthermore, each virtual data switch provides Pn=N / n internal logical ports, which are respectively connected to N flow generation engine containers;

[0022] The physical data network port bandwidth of the high-performance computing platform is Bw, and the bandwidth of a single flow generation engine container is (Bw×n) / N; the Pn internal logical ports of each virtual data switch are divided into Pn isolated virtual subnets to ensure that data will not be interconnected through the virtual data switch.

[0023] Furthermore, the number of physical switches is equal to the number of physical data network ports of the high-performance computing platform, and the physical network port rates Bw mounted on the physical switches and the virtual switch group should be consistent.

[0024] Furthermore, the total bandwidth of the device depends on the number n of physical data network ports and the bandwidth Bw of the high-performance computing platform. The total bandwidth of the device is n×Bw, and the bandwidth of a single flow generation engine container is (n×Bw) / N.

[0025] On the other hand, the present invention provides a network testing method implemented by using the above-mentioned large-scale virtual combined network traffic generating device, comprising the following steps:

[0026] S1, initialization configuration;

[0027] S1.1, instantiate a virtual switch group, bind physical ports, instantiate logical ports, and configure virtual subnets;

[0028] S1.2, instantiate the flow generation engine, connect the virtual switch group, configure the IP address, and start;

[0029] S1.3, configure the switch group virtual subnet;

[0030] S2, business modeling;

[0031] S2.1, business model creation, parameter configuration, and conversion into business script files;

[0032] S2.2, business script file distribution;

[0033] S3, business loading;

[0034] S3.1 The user terminal executes the service loading command;

[0035] S3.2 The business controller module controls the flow generation engine to parse the business script file and generate a business flow that conforms to the flow parameters in the business script;

[0036] S3.3 The flow generation engine records flow logs during the process of sending and receiving business flows;

[0037] S3.4 After the business flow ends, the flow generation engine pre-processes the flow log and reports it to the business controller module;

[0038] S3.5, the service controller module further calculates the rate, throughput, delay and packet loss rate statistics of the service flow based on the preprocessing results, and draws a visual chart;

[0039] S4. Result evaluation;

[0040] After the S4.1 service flow is completed, the user terminal executes the instruction to obtain the test result;

[0041] S4.2 The service controller module returns the test result to the user terminal.

[0042] Compared with the prior art, the technical effects that the present invention can produce are:

[0043] The large-scale virtual-real network traffic generating device described in the present invention realizes centralized control and data aggregation and analysis through a business controller module. Through this design, a control terminal can uniformly configure and control the traffic generator cluster, virtual switch group, and physical switch group of the entire device, and can simultaneously support parallel testing of N tested networking devices, and can simultaneously process performance indicators and comparative analysis of multiple tested links.

[0044] The present invention adopts virtualization container technology to support the deployment of N flow generation engines on a single high-performance computing platform, which can easily realize the large-scale flow generation engine deployment on a single machine, fully utilize the computing power resources of the high-performance computing platform, and significantly reduce the number of physical computing devices.

[0045] The virtual switch group of the present invention adopts virtualization switching technology and virtual subnet technology. The virtual switch group provides N virtual control ports and N virtual data ports for N flow generation engines, which significantly reduces the number of network physical switching devices for building a test system using existing technologies. By expanding the physical network ports of the high-performance computing platform, ultra-large bandwidth of a single device can be easily achieved. The virtual switch group can multiplex n physical network ports to N flow generation engines to fully utilize the bandwidth of the physical network ports. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0047] Figure 1 It is a schematic diagram of the equipment composition and connection relationship of a large-scale virtual-real combined network traffic generation device;

[0048] Figure 2 It is a module diagram of a large-scale virtual-real combined network traffic generation device;

[0049] Figure 3 A schematic diagram of the composition and connection relationship of a traffic generator cluster of a large-scale virtual-real combined network traffic generating device;

[0050] Figure 4A schematic diagram of the composition and connection relationship of the business controller module of a large-scale virtual-real combined network traffic generation device;

[0051] Figure 5 A schematic diagram of the composition and connection relationship of the virtual switch group of a large-scale virtual-real combined network traffic generation device;

[0052] Figure 6 A schematic diagram of the composition and connection relationship of the physical switch group of a large-scale virtual-real combined network traffic generation device;

[0053] Figure 7 A schematic diagram of the data network design for a large-scale virtual-real network traffic generation device;

[0054] Figure 8 A schematic diagram of the control network design for a large-scale virtual-real network traffic generation device;

[0055] Fig. 9 It is a flowchart of the network testing method;

[0056] Fig.10 A schematic diagram of the business data flow of a large-scale virtual-real combined network traffic generation device; DETAILED DESCRIPTION

[0057] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] One embodiment provides a large-scale virtual-real combined network traffic generation device, including a high-performance computing platform and a physical switch group;

[0059] The high-performance computing platform includes a service controller module, a traffic generator cluster, and a virtual switch group; the switch group includes a physical switch group;

[0060] The high-performance computing platform includes a business controller module, a traffic generator cluster, and a virtual switch group;

[0061] The service controller module, traffic generator cluster, virtual switch group and physical switch group are interconnected through the network to form a control network, providing control ports to the outside world;

[0062] The traffic generator cluster, virtual switch group and physical switch group are interconnected through the network to form a data network and provide data ports to the outside world;

[0063] The control port of the service controller module is connected to the corresponding port of the virtual switch group to form a control network between the service controller module and the virtual switch group, and the control network provides a control entry of the high-performance computing platform for the user control terminal;

[0064] The traffic generator cluster adopts containerized deployment mode, including N flow generator engine containers, corresponding to N devices to be tested, and the time between each container is completely synchronized with the host time; each flow generator engine container has two logical ports, one of which is a data port and the other is a control port; the control port and data port of each flow generator engine container are respectively connected to the corresponding ports of the virtual switch group;

[0065] The virtual switch group consists of a virtual control switch and multiple virtual data switches; the physical control network port of the high-performance computing platform is used to interconnect with the external user control terminal, the virtual control switch mounts the physical control network port of the high-performance computing platform, and provides N+1 logical ports, which correspond to the connection of N flow generation engine containers and 1 business controller module respectively; the physical data network port of the high-performance computing platform is interconnected with the network port of the physical switch group, the virtual data switch mounts the physical data network port of the high-performance computing platform, and provides N logical ports, which are interconnected with N flow generator engine containers respectively; the virtual data switch and the physical switch are divided into N virtual subnets to realize a one-to-one binding relationship between the flow generator engine container and the physical switch port, the subnet ID of each port of the physical switch is consistent with the subnet ID of each port of the virtual data switch, to realize a one-to-one mapping between the flow generator engine container and the tested device, and ensure that network traffic will not be transmitted inside the virtual data switch.

[0066] Reference Figure 1 An embodiment provides a large-scale virtual-real network traffic generating device, including a high-performance computing platform and a physical switch group; the device provides a control port and a data port to the outside, the control port is used to access the control terminal, and the data port is used to access the tested networking device; the control terminal manages the device in a unified manner, sends and receives network traffic through the data port of the device, analyzes the sending and receiving logs of the device, and evaluates the performance indicators of the wireless network group equipment such as network bandwidth, delay and packet loss rate.

[0067] Reference Figure 2 The large-scale virtual-real network traffic generating device includes a high-performance computing platform and a physical switch group; wherein the high-performance computing platform includes a business controller module, a traffic generator cluster, and a virtual switch group; the business controller module, the traffic generator cluster, the virtual switch group, and the physical switch group are interconnected to form a control network through a network, and a control port is provided to the outside; the traffic generator cluster, the virtual switch group, and the physical switch group are interconnected to form a data network through a network, and a data port is provided to the outside.

[0068] Reference Figure 3 , Figure 3 The diagram is a composition and connection relationship diagram of the flow generator cluster of a large-scale virtual-real combined network flow generating device. The core of the flow generator cluster is the flow generating engine; the flow generator cluster adopts a containerized deployment method, and the flow generator cluster contains N flow generating engines. The flow generating engine has two logical ports, a data port and a control port. The control port and data port of each flow generating engine are connected to the corresponding ports of the virtual switch group; the control port of the service controller module is also connected to the corresponding logical port of the virtual switch group, thereby forming a control network and a data network between the service controller module, the flow generator cluster and the virtual switch group. And the number N of the flow generating engine containers is the same as the number of devices to be tested. The flow generating engine is integrated with network traffic generation software, and the flow generator engine container supports parsing business flow scripts to generate services, supports the generated business flow rate to comply with the specified statistical distribution law (periodic, burst, Poisson, etc.), supports recording business sending and receiving logs, and provides a control management interface.

[0069] Reference Figure 4 , Figure 4The present invention is a schematic diagram of the composition and connection relationship of the business controller module of a large-scale virtual-real combined network traffic generating device. The business controller module is a background service software, including a business modeling unit, a business control unit, a result evaluation unit, a network configuration unit and a container management unit. The business controller module accesses the virtual switch group to form a control network through the control port, and provides the control port of the large-scale virtual-real combined network traffic generating device to the user control terminal through the control network. The container management unit realizes the creation, destruction, start, stop, parameter configuration and status monitoring of the flow generation engine by calling the container control interface, and provides the user with a visual container management operation and monitoring page; the network configuration unit is used to create the network, allocate IP addresses and divide virtual subnets for the flow generation engine, the virtual switch group and the physical switch group, establish the data network and control network of the device, and provide the user with a visual network configuration and network monitoring page; the core of the business modeling unit is business model management, and the business model is a description of the flow rate, packet size, duration, distribution model (random, burst, parked) and other aspects. The business modeling unit is used to parse and map the business model to the script file supported by the flow generation engine, and provide the user with the operations of creating, editing and deleting the business model through a visual page; the business control unit is used to load the business flow script, start, stop, preprocess the results, and report the results of the flow generation engine, and provide the user with a visual business operation and monitoring page; the result evaluation unit is used to summarize the preprocessing results, calculate the rate, throughput, delay and packet loss rate indicators of each business flow of the tested network group equipment, and provide the user with viewing and exporting operation pages in the form of visual icons and data.

[0070] Reference Figure 5 , Figure 5 The diagram is a diagram of the composition and connection relationship of the virtual switch group of a large-scale virtual-real network traffic generation device. The virtual switch group includes a virtual control switch and multiple virtual data switches; the virtual control switch is mounted with the physical control network port of the high-performance computing platform, providing N+1 internal logical ports. The virtual switch group is connected to N flow generation engines and 1 service controller module through the internal logical ports to form a control network. The number of virtual data switches is equal to the number of physical data network ports of the high-performance computing platform n. Each virtual data switch provides P n =N / n internal logical ports, corresponding to N flow generation engine containers; the physical data network port bandwidth of the high-performance computing platform is Bw, and the bandwidth of a single flow generation engine is (Bw×n) / N. The Pn internal logical ports of each virtual data switch are divided into Pn isolated virtual subnets (VLANs) to ensure that data will not be interconnected through the internal virtual data switch.

[0071] Reference Figure 6 , Figure 6 This is a schematic diagram of the composition and connection relationship of the physical switch group of a large-scale virtual-real network traffic generation device. The physical switch group includes multiple physical switches. The number of physical switches is equal to the number of physical data network ports n of the high-performance computing platform. The physical network port rates Bw mounted by the physical switch and the virtual switch group should be consistent, so as to ensure that the total bandwidth of the device reaches n*Bw. The network port of the physical switch is used to access the network group device under test. The virtual data switch and the physical switch are divided into N virtual subnets to achieve a one-to-one binding relationship between the flow generator engine container and the physical switch port. The subnet ID of each port of the physical switch is consistent with the subnet ID of each port of the virtual data switch, realizing a one-to-one mapping between the flow generator engine container and the device under test.

[0072] Reference Figure 7 , Figure 7 The data network design diagram of a large-scale virtual-real network traffic generating device. For the data network design of the large-scale virtual-real network traffic generating device, physically, the physical data network port of the high-performance computing platform is interconnected with the network port of the physical switch, and the network port rate specifications of the two need to be consistent; logically, the virtual data switch mounts the physical data network port of the high-performance computing platform, and provides N logical ports, which are interconnected with N flow generator engine containers respectively; the virtual data switch and the switch are divided into N virtual subnets to realize a one-to-one binding relationship between the flow generation engine and the switch port, and ensure that the network traffic will no longer be transmitted inside the virtual data switch. The total bandwidth of the entire device depends on the number of physical data network ports n and the bandwidth Bw of the high-performance computing platform, that is, n×Bw. The bandwidth of a single flow generation engine is (Bw×n) / N.

[0073] Reference Figure 8 , Figure 8 The schematic diagram of the control network design of a large-scale virtual-real network traffic generation device is shown in FIG. For the control network design of the large-scale virtual-real network traffic generation device, physically, the physical control network port of the high-performance computing platform is interconnected with the external user control terminal; logically, the virtual control switch mounts the physical control network port of the high-performance computing platform, providing N+1 logical ports, which respectively correspond to the connection of N flow generation engines and 1 business controller module, thereby realizing the physical isolation of the control network and the data network.

[0074] Reference Fig. 9 , one embodiment provides a network testing method implemented by using the above-mentioned large-scale virtual combined network traffic generation device, including the following steps:

[0075] S1, initialization configuration;

[0076] S1.1, instantiate a virtual switch group, bind physical ports, instantiate logical ports, and configure virtual subnets;

[0077] S1.2, instantiate the flow generation engine, connect the virtual switch group, configure the IP address, and start;

[0078] S1.3, configure the switch group virtual subnet;

[0079] S2, business modeling;

[0080] S2.1, business model creation, parameter configuration, and conversion into business script files;

[0081] S2.2, business script file distribution;

[0082] S3, business loading;

[0083] S3.1 The user terminal executes the service loading command;

[0084] S3.2 The business controller module controls the flow generation engine to parse the business script file and generate a business flow that conforms to the flow parameters in the business script;

[0085] S3.3 The flow generation engine records flow logs during the process of sending and receiving business flows;

[0086] S3.4 After the business flow ends, the flow generation engine pre-processes the flow log and reports it to the business controller module;

[0087] S3.5, the service controller module further calculates the rate, throughput, delay and packet loss rate statistics of the service flow based on the preprocessing results, and draws a visual chart;

[0088] S4. Result evaluation;

[0089] After the S4.1 service flow is completed, the user terminal executes the instruction to obtain the test result;

[0090] S4.2 The service controller module returns the test result to the user terminal.

[0091] Reference Fig.10 , Fig.10 The diagram is a service data flow diagram of a large-scale virtual-real network traffic generation device. The service data flow is generated by the flow generation engine, passes through the virtual data switch and the switch to the device under test, and the device under test receives the service data flow, passes through the switch and the virtual data switch to the flow generation engine. During the service transmission and reception process, the flow generation engine records the service flow log. By analyzing the timestamp, message length, packet sequence number and other information in the log file, the bandwidth, throughput, delay and packet loss rate of the link under test can be evaluated.

[0092] Matters not covered by the present invention are known technologies.

[0093] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0094] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

[0095] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A large-scale virtual-real network traffic generation device, characterized in that: Includes high-performance computing platform and physical switch group; The high-performance computing platform includes a business controller module, a traffic generator cluster, and a virtual switch group; The service controller module, traffic generator cluster, virtual switch group and physical switch group are interconnected through the network to form a control network, providing control ports to the outside world; The traffic generator cluster, virtual switch group and physical switch group are interconnected through the network to form a data network and provide data ports to the outside world; The control port of the service controller module is connected to the corresponding port of the virtual switch group to form a control network between the service controller module and the virtual switch group, and the control network provides a control entry of the high-performance computing platform for the user control terminal; The traffic generator cluster is deployed in a containerized manner, which includes N traffic generator engine containers, corresponding to N devices to be tested. The time between each container is completely synchronized with the host machine time. Each flow generation engine container has two logical ports, one of which is a data port and the other is a control port; the control port and data port of each flow generation engine container are respectively connected to the corresponding ports of the virtual switch group; The virtual switch group consists of a virtual control switch and multiple virtual data switches; the physical control network port of the high-performance computing platform is used to interconnect with the external user control terminal, the virtual control switch mounts the physical control network port of the high-performance computing platform, and provides N+1 logical ports, which correspond to the connection of N flow generation engine containers and 1 business controller module respectively; the physical data network port of the high-performance computing platform is interconnected with the network port of the physical switch group, the virtual data switch mounts the physical data network port of the high-performance computing platform, and provides N logical ports, which are interconnected with N flow generator engine containers respectively; the virtual data switch and the physical switch are divided into N virtual subnets to realize a one-to-one binding relationship between the flow generator engine container and the physical switch port, the subnet ID of each port of the physical switch is consistent with the subnet ID of each port of the virtual data switch, to realize a one-to-one mapping between the flow generator engine container and the tested device, and ensure that network traffic will not be transmitted inside the virtual data switch.

2. The large-scale virtual-real combined network traffic generating device as claimed in claim 1, characterized in that: The flow generator engine container integrates network traffic generation software. The flow generator engine container supports parsing business flow scripts to generate services, supports the generated business flow rate to comply with specified statistical distribution rules, supports recording business sending and receiving logs, and provides a control management interface.

3. The large-scale virtual-real combined network traffic generating device as claimed in claim 1, characterized in that: The service controller module includes a service modeling unit, a service control unit, a result evaluation unit, a network configuration unit and a container management unit; The container management unit implements the creation, destruction, start, stop, parameter configuration and status monitoring of the flow generation engine container by calling the container control interface, and provides users with a visual container management operation and monitoring page; The network configuration unit creates networks, allocates IP addresses, and divides virtual subnets for the flow generation engine container, virtual switch group, and physical switch group, establishes data networks and control networks, and provides users with visual network configuration and network monitoring pages; The business modeling unit includes business model management. The business model is a description file that describes the flow rate, packet size, duration, and distribution model parameters. The business modeling unit is responsible for parsing the description file and mapping it to the business script file supported by the flow generation engine container. The business modeling unit provides users with operations for creating, editing, and deleting business models through a visual page. The business control unit is responsible for loading business flow scripts, starting, stopping, preprocessing results, and reporting results on the flow generator engine container, and provides users with visual business operations and monitoring pages; The result evaluation unit summarizes the preprocessing results, calculates the rate, throughput, delay and packet loss rate indicators of each service flow in the tested network, and provides users with viewing and export operation pages in the form of visual icons and data.

4. The large-scale virtual-real combined network traffic generating device as claimed in claim 3, characterized in that: The number of virtual data switches is equal to the number n of physical data network ports of the high-performance computing platform.

5. The large-scale virtual-real combined network traffic generating device as claimed in claim 4, characterized in that: Each virtual data switch provides Pn=N / n internal logical ports, which are connected to N flow generation engine containers respectively; The physical data network port bandwidth of the high-performance computing platform is Bw, and the bandwidth of a single flow generation engine container is (Bw×n) / N; the Pn internal logical ports of each virtual data switch are divided into Pn isolated virtual subnets to ensure that data will not be interconnected through the virtual data switch.

6. The large-scale virtual-real combined network traffic generating device as claimed in claim 5, characterized in that: The number of physical switches is equal to the number of physical data network ports of the high-performance computing platform. The physical network port rates Bw of the physical switches and virtual switch groups mounted should be consistent.

7. The large-scale virtual-real combined network traffic generating device as claimed in claim 6, characterized in that: The total bandwidth of the device depends on the number n of physical data network ports and the bandwidth Bw of the high-performance computing platform. The total bandwidth of the device is n×Bw, and the bandwidth of a single flow generation engine container is (n×Bw) / N.

8. A network testing method based on the large-scale virtual-real combined network traffic generating device as claimed in claim 1, characterized in that: include: S1, initialization configuration; S1.1, instantiate a virtual switch group, bind physical ports, instantiate logical ports, and configure virtual subnets; S1.2, instantiate the flow generation engine, connect the virtual switch group, configure the IP address, and start; S1.3, configure the switch group virtual subnet; S2, business modeling; S2.1, business model creation, parameter configuration, and conversion into business script files; S2.2, business script file distribution; S3, business loading; S3.1 The user terminal executes the service loading command; S3.2 The business controller module controls the flow generation engine to parse the business script file and generate a business flow that conforms to the flow parameters in the business script; S3.3 The flow generation engine records flow logs during the process of sending and receiving business flows; S3.4 After the business flow ends, the flow generation engine pre-processes the flow log and reports it to the business controller module; S3.5, the service controller module further calculates the rate, throughput, delay and packet loss rate statistics of the service flow based on the preprocessing results, and draws a visual chart; S4. Result evaluation; After the S4.1 service flow is completed, the user terminal executes the instruction to obtain the test result; S4.2 The service controller module returns the test result to the user terminal.

Images