800G optical module InfiniBand networking test method

Abstract

The invention relates to an 800G optical module InfiniBand networking test method, which comprises the following steps of: respectively connecting two ports of a switch with network cards of two servers by using an 800G optical module, and configuring the network cards into an InfiniBand mode; the two servers are used as a server side and a client side respectively, the server side and the client side are configured through the network card, an actual flow report of the 800G optical module is read, and an 800G optical module InfiniBand networking test environment is constructed; and according to the 800G optical module InfiniBand networking test environment, respectively carrying out a stability test, a function availability test and a robustness test on the 800G optical module. According to the method disclosed by the invention, the current network application can be simulated, the performance of the optical module in the server intercommunication service can be verified, and the function availability of the module can be evaluated; the possible operation of the current network on the server and the switch can be simulated, and the robustness of the module is evaluated; different use environments of the existing network can be simulated, and the use stability of the module is evaluated.

Description

Technical Field

[0001] The present invention relates to the technical field of 800G optical module testing, and in particular to an 800G optical module InfiniBand networking testing method. Background Art

[0002] In recent years, with the increasing popularity and expansion of high-capacity bandwidth-demanding network services such as live streaming, short videos, online dramas, and AI, the capacity of switches within data centers has increased, and the speed of modules on these switches has correspondingly increased. This has also led to the emergence of InfiniBand networks, which are more suitable for AI. Existing technologies generally use Ethernet networking for testing. However, Ethernet networking cannot verify InfiniBand modules, and only individual tests can be performed. Therefore, a method for testing 800G optical modules in InfiniBand networking is urgently needed to address the existing technical issues. Summary of the Invention

[0003] In view of the above problems, the present invention is proposed to provide an 800G optical module InfiniBand networking testing method that overcomes the above problems or at least partially solves the above problems.

[0004] In order to solve the above technical problems, the embodiments of the present application disclose the following technical solutions:

[0005] In a first aspect, an embodiment of the present invention discloses a method for testing an 800G optical module InfiniBand network, comprising:

[0006] S100. Use 800G optical modules to connect the two ports of the switch to the network cards of the two servers, and configure the network cards to InfiniBand mode.

[0007] S200. The two servers are used as the server and client, respectively, to issue commands to simulate traffic, read the actual traffic results on the server, and build an 800G optical module InfiniBand networking test environment;

[0008] S300. According to the 800G optical module InfiniBand networking test environment, the 800G optical module is subjected to stability test, functional availability test and robustness test respectively.

[0009] Furthermore, in S100, the switch adopts NVIDIA's MQM9700-NS2F model InfiniBand switch, and the network card adopts BlueField3 or ConnectX7 network card.

[0010] Furthermore, in S200, a command is issued to simulate traffic, and the specific method includes: using the InfiniBand mode network card write bandwidth client command to start traffic request on the client, using the InfiniBand mode network card write bandwidth server command to start traffic service on the server, and using the InfiniBand mode network card write bandwidth client command to start traffic request on the client; after the server sends the traffic for a specified time, the InfiniBand network card read bandwidth server and client commands are used respectively to read the actual traffic results on the server.

[0011] Furthermore, in S200, the actual traffic report includes at least the throughput, bit error rate and number of packet losses of the traffic.

[0012] Furthermore, in S300, the stability test of the 800G optical module is performed, and the specific method includes: placing the switch in a temperature cycle box, and subjecting the 800G optical module to a flow test for a preset time in the temperature cycle box, wherein the temperature cycle box is configured to maintain high and low temperatures for 3 hours respectively, and the heating and cooling speed is 1°C / min; when the 800G optical module is subjected to a flow test in the temperature cycle box for the preset time, if the 800G optical module has no packet loss, no connection disconnection, and no flash disconnection, the stability test of the 800G optical module passes.

[0013] Furthermore, in S300, a functional availability test is performed on the 800G optical module, and the specific method includes: reading the basic information, DDM information, register information and electric eye diagram of the optical module according to the constructed 800G optical module InfiniBand networking test environment; performing a bit error rate test, a symbol error test and a pseudo-random binary sequence bit error rate test according to the basic information, DDM information, register information and electric eye diagram of the optical module respectively; when the bit error rate test, symbol error test and pseudo-random binary sequence bit error rate test of the optical module pass, the functional availability test of the 800G optical module passes.

[0014] Furthermore, in S300, a bit error rate test is performed on the optical module, and the specific method includes:

[0015] Generate a test pattern using a pseudo-random binary sequence as a test signal that can simulate the statistical characteristics of a real data stream and is suitable for bit error rate testing of high-speed communication systems;

[0016] Send the test signal, and send the generated PRBS signal to the device under test through the signal generator;

[0017] Receive and compare signals. After the device under test processes the signal and returns the result, it is compared with the reference data.

[0018] Calculate the bit error rate based on the number of error bits and the total number of bits transmitted.

[0019] Furthermore, in S300, a symbol error test is performed on the optical module. The specific method includes:

[0020] Generate a test symbol sequence, using a specific symbol sequence as a test signal, wherein the symbol sequence is generated based on a specific encoding method;

[0021] Sending a test symbol sequence, sending the generated symbol sequence to the device under test through a signal generator;

[0022] Receive and decode symbols. The device under test receives the symbol sequence and decodes it. After receiving the decoding result of the device under test, it compares it with the reference symbol sequence;

[0023] Calculate the symbol error rate based on the number of erroneous symbols and the total number of transmitted symbols: symbol error rate = number of erroneous symbols / total number of transmitted symbols.

[0024] Furthermore, in S300, the robustness test of the 800G optical module is performed, and the specific method includes: performing plug-in test, optical fiber plug-in test, server restart test, network card fwreset test, switch port repeated shutdown test, network card toggle test, switch soft restart test and switch hard restart test on the optical module respectively; when after the plug-in test, optical fiber plug-in test, server restart test, network card fwreset test, switch port repeated shutdown test, network card toggle test, switch soft restart test and switch hard restart test, the basic information query of the optical module is normal and the traffic returns to normal, the robustness test of the 800G optical module passes.

[0025] In a second aspect, an embodiment of the present invention discloses an electronic device, including:

[0026] one or more processors;

[0027] a memory for storing one or more programs;

[0028] When the one or more programs are executed by the one or more processors, the one or more processors implement the optical module InfiniBand networking testing method.

[0029] The beneficial effects of the above technical solutions provided by the embodiments of the present invention include at least:

[0030] The embodiment of the present invention discloses an 800G optical module InfiniBand networking test method, comprising: using an 800G optical module to connect the two ports of a switch to the network cards of two servers respectively, and configuring the network cards to InfiniBand mode; using the two servers as the server and client respectively, issuing commands to simulate traffic, reading the actual traffic results on the server, and building an 800G optical module InfiniBand networking test environment; according to the 800G optical module InfiniBand networking test environment, performing stability testing, functional availability testing, and robustness testing on the 800G optical module respectively. The method disclosed in the present invention can simulate existing network applications, verify the performance of optical modules in server intercommunication services, and evaluate the functional availability of modules; can simulate the operations that may be performed on servers and switches in the existing network to evaluate the robustness of the module; can simulate different usage environments of the existing network to evaluate the stability of module use.

[0031] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0033] Figure 1 This is a flowchart of a method for testing InfiniBand networking of an 800G optical module in Example 1 of the present invention;

[0034] Figure 2 This is a test network diagram of an 800G optical module InfiniBand network in Example 1 of the present invention;

[0035] Figure 3 This is a structural diagram of an electronic device in Example 2 of the present invention. DETAILED DESCRIPTION

[0036] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0037] In order to solve the problems existing in the prior art, an embodiment of the present invention provides an 800G optical module InfiniBand networking testing method.

[0038] Example 1

[0039] The present invention discloses a method for testing 800G optical module InfiniBand network. Figure 1 ,include:

[0040] S100. Figure 2 , use 800G optical modules to connect the two ports of the switch to the network cards of the two servers respectively, and configure the network cards to InfiniBand mode;

[0041] In S100 of this embodiment, the switch uses an NVIDIA MQM9700-NS2F InfiniBand switch, and the network card uses a BlueField3 or ConnectX7 network card. Specifically, the main function of the 800G optical module is to perform photoelectric conversion, that is, converting electrical signals into optical signals for transmission, and then converting the received optical signals back into electrical signals. This conversion process enables data to be transmitted at extremely high speeds in optical fibers, meeting the needs of large-scale data transmission and processing. MQM9700-NS2F is a high-performance InfiniBand switch launched by NVIDIA that supports multiple transmission rates, including 400G NDR (Non-Data Rate), 200G HDR (High Data Rate), 100GEDR (Enhanced Data Rate), 56G FDR (Four Data Rate) and 40G ODR (Old Data Rate). BlueField-3 is a data processing unit (DPU) designed to offload and isolate various data center infrastructure services to improve the performance, efficiency and security of data centers. ConnectX-7 is a SmartNIC that focuses on providing high-speed network connectivity and data transmission capabilities.

[0042] S200. The two servers are used as the server and client, respectively, to issue commands to simulate traffic, read the actual traffic results on the server, and build an 800G optical module InfiniBand networking test environment;

[0043] In S200 of this embodiment, a command is issued to simulate traffic flow. The specific method includes: using the InfiniBand mode network card write bandwidth client command to initiate a traffic request on the client; using the InfiniBand mode network card write bandwidth server command to initiate a traffic service on the server; and using the InfiniBand mode network card write bandwidth client command to initiate a traffic request on the client; after the server sends traffic flow for a specified period of time, using the InfiniBand mode network card read bandwidth server and client commands respectively to read the actual traffic flow results on the server. The actual traffic flow report includes at least the traffic throughput, bit error rate, and packet loss number.

[0044] Specifically, during the test, the two servers were used as the server and client respectively. On the server side, the InfiniBand network card write bandwidth server command was used to start the traffic service, and on the client side, the InfiniBand network card write bandwidth client command was used to start the traffic request. After sending the traffic for the specified time, the InfiniBand network card read bandwidth server and client commands were used respectively to read the actual traffic report and obtain statistical information such as throughput, bit error rate, and number of packet losses.

[0045] S300. According to the 800G optical module InfiniBand networking test environment, the 800G optical module is subjected to stability test, functional availability test and robustness test respectively.

[0046] In S300 of this embodiment, the stability test of the 800G optical module is performed, and the specific method includes: placing the switch in a temperature cycle box, and subjecting the 800G optical module to a flow test for a preset time in the temperature cycle box, wherein the temperature cycle box is configured to maintain high and low temperatures for 3 hours respectively, and a heating and cooling rate of 1°C / min; when the 800G optical module is subjected to a flow test in the temperature cycle box for a preset time, if the 800G optical module has no packet loss, no connection disconnection, and no flash disconnection, the stability test of the 800G optical module passes.

[0047] In S300 of this embodiment, a functional availability test is performed on the 800G optical module. The specific method includes: reading the basic information, DDM information, register information and electric eye diagram of the optical module according to the constructed 800G optical module InfiniBand networking test environment; performing a bit error rate test (BER test), a symbol error test (Symbol Errors test) and a pseudo-random binary sequence bit error rate test (PRBS BER test) according to the basic information, DDM information, register information and electric eye diagram of the optical module; wherein, BER (bit error rate) and Symbol Errors (symbol error) tests are crucial test links in digital communication systems, which are used to evaluate the reliability and accuracy of data transmission. When the optical module bit error rate test, symbol error test and pseudo-random binary sequence bit error rate test pass, the functional availability test of the 800G optical module passes. The basic information of the optical module includes at least the basic information of the optical module, wavelength, and transmission distance; the DDM information includes at least the voltage, bias current, transmit and receive optical power, and module case temperature; the register information includes at least the register value of the optical module; and the electrical eye diagram is the electrical eye diagram of the server or switch.

[0048] In some preferred embodiments, a bit error rate (BER) test is performed on the optical module, and the specific method includes:

[0049] Generate a test pattern using a pseudo-random binary sequence as a test signal that can simulate the statistical characteristics of a real data stream and is suitable for bit error rate testing of high-speed communication systems;

[0050] Send the test signal, and send the generated PRBS signal to the device under test through the signal generator;

[0051] Receive and compare signals. After the device under test processes the signal and returns the result, it is compared with the reference data.

[0052] Calculate the bit error rate based on the number of error bits and the total number of bits transmitted.

[0053] In some preferred embodiments, a symbol error test is performed on the optical module. The specific method includes:

[0054] Generate a test symbol sequence, using a specific symbol sequence as a test signal, wherein the symbol sequence is generated based on a specific encoding method;

[0055] Sending a test symbol sequence, sending the generated symbol sequence to the device under test through a signal generator;

[0056] Receive and decode symbols. The device under test receives the symbol sequence and decodes it. After receiving the decoding result of the device under test, it compares it with the reference symbol sequence;

[0057] Calculate the symbol error rate based on the number of erroneous symbols and the total number of transmitted symbols: symbol error rate = number of erroneous symbols / total number of transmitted symbols.

[0058] BER and Symbol Error testing are essential components of digital communication systems. By comprehensively evaluating a system's bit error rate and symbol error rate, we can gain a comprehensive understanding of the system's transmission performance and take effective measures to improve system reliability and accuracy.

[0059] In this embodiment, the PRBS (pseudo-random binary sequence) BER (bit error rate) test is a commonly used test method in modern communications and data transmission to evaluate the performance of communication systems. RBS testing is a method based on generating a specific binary sequence that simulates the characteristics of real data but in a predictable manner, and is used to test communication links. PRBS sequences have a balanced number of 0s and 1s and frequent bit transitions, making them ideal test sequences. PRBS sequences can be generated using a linear feedback shift register (LFSR), which consists of a series of storage cells and feedback logic. The initial state of the LFSR defines the specific value at the beginning of the sequence, and specific feedback logic determines the next value generated by the sequence.

[0060] In S300 of some preferred embodiments, the 800G optical module is subjected to a robustness test, and the specific method includes: performing plug-in test, fiber plug-in test, server restart test, network card fwreset test, switch port repeated shutdown test, network card toggle test, switch soft restart test and switch hard restart test on the optical module respectively; when after the plug-in test, fiber plug-in test, server restart test, network card fwreset test, switch port repeated shutdown test, network card toggle test, switch soft restart test and switch hard restart test, the basic information query of the optical module is normal and the traffic returns to normal, the 800G optical module robustness test passes.

[0061] This embodiment discloses a 800G optical module InfiniBand networking test method, including: using an 800G optical module to connect the two ports of a switch to the network cards of two servers respectively, and configuring the network cards to InfiniBand mode; using the two servers as the server and client respectively, issuing commands to simulate traffic, reading the actual traffic results on the server, and building an 800G optical module InfiniBand networking test environment; according to the 800G optical module InfiniBand networking test environment, performing stability testing, functional availability testing and robustness testing on the 800G optical module respectively. The method disclosed in the present invention can simulate existing network applications, verify the performance of optical modules in server intercommunication services, and evaluate the functional availability of modules; can simulate the operations that may be performed on servers and switches in the existing network to evaluate the robustness of modules; can simulate different usage environments of the existing network to evaluate the stability of module use.

[0062] Example 2

[0063] Based on the same inventive concept, an embodiment of the present disclosure further provides an electronic device. Figure 2 FIG. 1 is a schematic diagram of the structure of an electronic device according to an embodiment of the present disclosure. Figure 2 As shown, an embodiment of the present disclosure provides an electronic device comprising: one or more processors 101, a memory 102, and one or more I / O interfaces 103. The memory 102 stores one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors implement any of the optimization methods described in the above embodiments; the one or more I / O interfaces 103 are connected between the processor and the memory and are configured to implement information exchange between the processor and the memory.

[0064] Among them, the processor 101 is a device with data processing capabilities, including but not limited to a central processing unit (CPU); the memory 102 is a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically such as SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), and flash memory (FLASH); the I / O interface (read-write interface) 103 is connected between the processor 101 and the memory 102, and can realize information interaction between the processor 101 and the memory 102, including but not limited to a data bus (Bus), etc.

[0065] In some embodiments, the processor 101 , the memory 102 , and the I / O interface 103 are connected to each other via a bus 104 , and further connected to other components of the computing device.

[0066] In some embodiments, the one or more processors 101 include a field programmable gate array.

[0067] According to an embodiment of the present disclosure, a computer-readable medium is further provided, wherein a computer program is stored on the computer-readable medium, wherein when the program is executed by a processor, the steps of any optimization method in the above-mentioned embodiment are implemented.

[0068] It should be understood that the specific order or hierarchy of steps in the disclosed processes is an example of an exemplary method. Based on design preferences, it should be understood that the specific order or hierarchy of steps in the process can be rearranged without departing from the scope of the present disclosure. The accompanying method claims present elements of the various steps in an exemplary order and are not intended to be limited to the specific order or hierarchy described.

[0069] In the foregoing detailed description, various features are grouped together in a single embodiment to simplify the disclosure. This method of disclosure should not be interpreted as reflecting an intention that embodiments of the claimed subject matter require more features than are expressly recited in each claim. On the contrary, as reflected in the appended claims, the invention comprises less than all the features of any individual disclosed embodiment. The appended claims are therefore hereby expressly incorporated into the detailed description, with each claim standing on its own as a separate preferred embodiment of the invention.

[0070] Those skilled in the art will also appreciate that the various illustrative logic blocks, modules, circuits, and algorithmic steps described in conjunction with the embodiments herein may be implemented as electronic hardware, computer software, or a combination thereof. In order to clearly illustrate the interchangeability between hardware and software, the various illustrative components, blocks, modules, circuits, and steps described above are generally described around their functions. Whether such functions are implemented as hardware or software depends on the specific application and the design constraints imposed on the entire system. A skilled person may implement the described functions in an adaptable manner for each specific application, but such implementation decisions should not be interpreted as departing from the scope of protection of this disclosure.

[0071] The steps of the methods or algorithms described in conjunction with the embodiments herein may be directly embodied as hardware, software modules executed by a processor, or a combination thereof. The software module may be located in a RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, register, hard disk, removable disk, CD-ROM, or any other form of storage medium well known in the art. An exemplary storage medium is connected to the processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium may also be an integral part of the processor. The processor and storage medium may be located in an ASIC. The ASIC may be located in a user terminal. Of course, the processor and storage medium may also be present in a user terminal as discrete components.

[0072] For software implementation, the techniques described in this application can be implemented using modules (e.g., procedures, functions, etc.) that perform the functions described in this application. These software codes can be stored in a memory unit and executed by a processor. The memory unit can be implemented within the processor or external to the processor. In the latter case, it is communicatively coupled to the processor via various means, which are well known in the art.

[0073] The foregoing description includes examples of one or more embodiments. Of course, it is not possible to describe all possible combinations of components or methods for the purposes of describing the above embodiments, but one of ordinary skill in the art will recognize that the various embodiments may be further combined and arranged. Therefore, the embodiments described herein are intended to encompass all such changes, modifications and variations that fall within the scope of the appended claims. Furthermore, to the extent the term "comprising" is used in the specification or claims, the term is intended to be encompassed in a manner similar to the term "including," as explained in terms of "including," used as a transitional word in the claims. Furthermore, any use of the term "or" in the specification of the claims is intended to mean a "non-exclusive or."

Claims

1. A 800G optical module InfiniBand networking test method, characterized in that: include: S100. Use 800G optical modules to connect the two ports of the switch to the network cards of the two servers, and configure the network cards to InfiniBand mode. S200. The two servers are used as the server and client, respectively, and the server and client are configured through the network card, and the actual flow report of the 800G optical module is read to build an 800G optical module InfiniBand networking test environment; S300. According to the 800G optical module InfiniBand networking test environment, the 800G optical module is subjected to stability test, functional availability test and robustness test respectively.

2. The 800G optical module InfiniBand networking test method according to claim 1, wherein: In S100, the switch uses NVIDIA's MQM9700-NS2F model InfiniBand switch, and the network card uses BlueField3 or ConnectX7 network card.

3. The 800G optical module InfiniBand networking test method according to claim 1, wherein: In S200, the server and client are configured through the network card. The specific method includes: using the InfiniBand mode network card to write bandwidth client command to start traffic request on the client, using the InfiniBand mode network card to write bandwidth server command to start traffic service on the server, and using the InfiniBand mode network card to write bandwidth client command to start traffic request on the client; after the server sends the traffic for the specified time, it uses the InfiniBand network card to read bandwidth server and client commands respectively to read the actual traffic report of the 800G optical module.

4. The 800G optical module InfiniBand networking test method according to claim 1, wherein: In S200, the actual traffic report includes at least the throughput, bit error rate and number of packet losses of the traffic.

5. The 800G optical module InfiniBand networking test method according to claim 1, wherein: In S300, the stability test of the 800G optical module is performed. The specific method includes: placing the switch in a temperature cycle box, and subjecting the 800G optical module to a flow test for a preset time in the temperature cycle box, wherein the temperature cycle box is configured to maintain high and low temperatures for 3 hours respectively, and a temperature rise and fall rate of 1°C / min; when the 800G optical module is subjected to a flow test in the temperature cycle box for a preset time, if the 800G optical module has no packet loss, no connection disconnection, and no flash disconnection, the stability test of the 800G optical module is passed.

6. The 800G optical module InfiniBand networking test method according to claim 1, wherein: In S300, a functional availability test is performed on the 800G optical module. The specific method includes: reading the basic information, DDM information, register information and electrical eye diagram of the optical module according to the constructed 800G optical module InfiniBand networking test environment; performing a bit error rate test, a symbol error test and a pseudo-random binary sequence bit error rate test according to the basic information, DDM information, register information and electrical eye diagram of the optical module; when the bit error rate test, symbol error test and pseudo-random binary sequence bit error rate test of the optical module pass, the functional availability test of the 800G optical module passes.

7. The 800G optical module InfiniBand networking test method according to claim 6, characterized in that: Perform bit error rate testing on optical modules. Specific methods include: Generate a test pattern using a pseudo-random binary sequence as a test signal that can simulate the statistical characteristics of a real data stream and is suitable for bit error rate testing of high-speed communication systems; Send the test signal, and send the generated PRBS signal to the device under test through the signal generator; Receive and compare signals. The device under test processes the signal and returns the result. The error detector receives the output data of the device under test and compares it with the reference data. Calculate the bit error rate based on the number of error bits and the total number of bits transmitted.

8. The 800G optical module InfiniBand networking test method according to claim 6, wherein: Perform symbol error testing on optical modules. Specific methods include: Generate a test symbol sequence, using a specific symbol sequence as a test signal, wherein the symbol sequence is generated based on a specific encoding method; Sending a test symbol sequence, sending the generated symbol sequence to the device under test through a signal generator; Receive and decode symbols. The device under test receives the symbol sequence and decodes it. The error detector receives the decoding result of the device under test and compares it with the reference symbol sequence. Calculate the symbol error rate based on the number of erroneous symbols and the total number of transmitted symbols: symbol error rate = number of erroneous symbols / total number of transmitted symbols.

9. The 800G optical module InfiniBand networking test method according to claim 1, wherein: In S300, the robustness test of the 800G optical module is performed, and the specific method includes: performing plug-in test, fiber plug-in test, server restart test, network card fwreset test, switch port repeated shutdown test, network card toggle test, switch soft restart test and switch hard restart test on the optical module respectively; when the plug-in test, fiber plug-in test, server restart test, network card fwreset test, switch port repeated shutdown test, network card toggle test, switch soft restart test and switch hard restart test are completed, if the basic information query of the optical module is normal and the traffic returns to normal, the robustness test of the 800G optical module is passed.

10. An electronic device comprising: one or more processors; a memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the 800G optical module InfiniBand networking testing method described in any one of claims 1-9.

Images