CN115766549B - Antithetical couplet test method, device, electronic equipment and storage medium - Google Patents

Abstract

The embodiment of the invention provides a antithetical couplet test method, a device, electronic equipment and a storage medium, wherein the method comprises the following steps: when the PCI-E equipment to be tested is accessed into the switch group, a second bandwidth value of the PCI-E equipment to be tested is obtained, the target PCI-E equipment to be connected in parallel of the same type is determined, the second bandwidth value is compared with the first bandwidth value of each target PCI-E equipment to be connected in parallel, and if the first bandwidth value is not smaller than the second bandwidth value, the corresponding target PCI-E equipment to be connected in parallel is connected in parallel with the PCI-E equipment to be tested. By integrating a plurality of first servers into the server side and taking the switch group as an intermediate medium, redundant servers are not required to be built in performance pressure testing, and only PCI-E equipment to be tested is required to be accessed into the switch group, the server side automatically acquires a second bandwidth value of the PCI-E equipment to be tested, and the target PCI-E equipment to be connected, which is of the same type and can meet the second bandwidth value, is automatically matched, so that the matched two PCI-E equipment are subjected to the connection test, the workload is greatly reduced, and the testing efficiency is greatly improved.

Description

Couplet testing method, device, electronic device and storage medium

Technical Field

The present invention relates to the technical field of data transmission, and in particular to a couplet testing method, a couplet testing device, an electronic device and a computer-readable storage medium.

Background technique

Server riser cards (adapter cards/function expansion cards) are often equipped with network cards, HCA (host channel adapter) cards and other PCI-E (Peripheral Component Interconnect-Express, a high-speed serial computer expansion bus standard) devices to achieve communication and information exchange. Based on the needs of server design, PCI-E devices need to be stress-tested before they can be added to the server's parts list. The performance stress test criteria for PCI-E devices mainly test the bandwidth rate of PCI-E devices.

During each test, the test engineer needs to install the PCI-E device to be tested into a server. However, to cooperate with the test of this PCI-E device, the test engineer also needs to build another server and install the same PCI-E device to test it in conjunction with it. Such a testing principle undoubtedly increases the workload. The singleness of the test situation does not conform to the usage principles in actual applications and is not conducive to the discovery of problems.

Summary of the invention

In view of the above problems, embodiments of the present invention are proposed to provide a couplet testing method that overcomes the above problems or at least partially solves the above problems.

The embodiment of the present invention also provides a couplet testing device, an electronic device and a storage medium to ensure the implementation of the above method.

In order to solve the above problems, an embodiment of the present invention discloses a pairing test method, which is applied to a server, the server is connected to a switch group, the server is integrated with a plurality of first servers, the plurality of first servers are respectively installed with a plurality of types of PCI-E devices to be paired, the plurality of PCI-E devices to be paired respectively have one or more first data ports, and the method comprises:

When the second server accesses the switch group as a client, obtaining a second bandwidth value of a second data port of the PCI-E device to be tested on the second server;

Determine a plurality of target PCI-E devices to be coupled that match the type of the PCI-E device to be tested;

Determine the first data ports of the plurality of target PCI-E devices to be coupled as candidate first data ports, and determine the first bandwidth value of each candidate first data port;

Compare the second bandwidth value of the second data port with the first bandwidth values of each candidate first data port in sequence;

If the first bandwidth value is not less than the second bandwidth value, using the candidate first data port corresponding to the first bandwidth value as the data exchange port of the second data port;

Allocating an IP address to the second data port;

The second data port and the data exchange port are connected by using the IP address, so that the PCI-E device to be coupled corresponding to the data exchange port is subjected to a coupling test with the PCI-E device to be tested.

Optionally, the multiple first servers are respectively deployed in different first system environments; if the first bandwidth value is not less than the second bandwidth value, using the candidate first data port corresponding to the first bandwidth value as the data interaction port of the second data port includes:

If the first bandwidth value is not less than the second bandwidth value, determining a second system environment in which the second server is deployed, and determining a target first system environment in which the candidate first data port corresponding to the first bandwidth value is located;

Determining whether the target first system environment is consistent with the second system environment;

If the target first system environment is consistent with the second system environment, the candidate first data port corresponding to the first bandwidth value is used as the data exchange port of the second data port.

Optionally, the determining of a plurality of target PCI-E devices to be coupled that match the type of the PCI-E device to be tested includes:

receiving, through the switch group, a data packet sent by the second server;

Sending the data packets to the multiple first servers respectively;

In each first server, determining a PCI-E device to be coupled that can receive the data packet;

The PCI-E device to be paired that can receive the data packet is determined as a target PCI-E device to be paired that matches the type of the PCI-E device to be tested.

Optionally, the data packet includes bandwidth data; and determining the first bandwidth value of each candidate first data port includes:

Determine an upper bandwidth limit value of each candidate first data port that can receive the bandwidth data;

Determine the state of each candidate first data port; the state includes an idle state and a used state;

If the candidate first data port is in an idle state, using the bandwidth upper limit value as the first bandwidth value of the candidate first data port;

If the candidate first data port is in use, the difference between the bandwidth upper limit value and the used bandwidth value is calculated, and the difference is used as the first bandwidth value of the candidate first data port.

Optionally, the method further comprises:

Based on the second data ports and the data exchange ports connected in pairs, transmitting the test tool to the second server, so that the second server deploys the test environment using the test tool;

After the second server completes the deployment of the test environment, a performance stress test for the PCI-E device to be tested is started.

Optionally, the method further comprises:

If the first bandwidth value is smaller than the second bandwidth value, the candidate first data port corresponding to the first bandwidth value is not used as the data exchange port of the second data port.

Optionally, the method further comprises:

If all the first bandwidth values are smaller than the second bandwidth value, it is determined to feed back the remaining usage time of the first data port that satisfies the second bandwidth value.

The embodiment of the present invention further discloses a pairing test device, which is applied to a server end, the server end is connected to a switch group, the server end is integrated with a plurality of first servers, the plurality of first servers are respectively installed with a plurality of types of PCI-E devices to be paired, the plurality of PCI-E devices to be paired respectively have one or more first data ports, and the device comprises:

A second bandwidth value acquisition module, configured to acquire a second bandwidth value of a second data port of a PCI-E device to be tested on the second server when the second server accesses the switch group as a client;

A target PCI-E device to be coupled determination module is used to determine a plurality of target PCI-E devices to be coupled that match the type of the PCI-E device to be tested;

A first bandwidth value determination module, configured to determine the first data ports of the plurality of target PCI-E devices to be coupled as candidate first data ports, and determine a first bandwidth value of each candidate first data port;

a bandwidth value comparison module, configured to compare the second bandwidth value of the second data port with the first bandwidth values of the candidate first data ports in sequence;

A first data exchange port determination module, configured to use the candidate first data port corresponding to the first bandwidth value as the data exchange port of the second data port if the first bandwidth value is not less than the second bandwidth value;

An IP address allocation module, used for allocating an IP address to the second data port;

The connection module is used to connect the second data port and the data exchange port by using the IP address, so that the PCI-E device to be coupled corresponding to the data exchange port can be coupled with the PCI-E device to be tested.

Optionally, the multiple first servers are respectively deployed in different first system environments; and the first data interaction port determination module includes:

a system environment determination submodule, configured to determine, if the first bandwidth value is not less than the second bandwidth value, a second system environment in which the second server is deployed, and determine a target first system environment in which the candidate first data port corresponding to the first bandwidth value is located;

A determination submodule, used to determine whether the target first system environment is consistent with the second system environment;

The first data exchange port determination submodule is configured to use the candidate first data port corresponding to the first bandwidth value as the data exchange port of the second data port if the target first system environment is consistent with the second system environment.

Optionally, the target to-be-coupled PCI-E device determination module includes:

A data packet receiving submodule, configured to receive a data packet sent by the second server through the switch group;

A data packet sending submodule, used for sending the data packets to the multiple first servers respectively;

A submodule for determining a PCI-E device to be paired that can receive a data packet, used to determine, in each first server, a PCI-E device to be paired that can receive the data packet;

The target PCI-E device to be coupled determination submodule is used to determine the PCI-E device to be coupled that can receive the data packet as the target PCI-E device to be coupled that matches the type of the PCI-E device to be tested.

Optionally, the data packet includes bandwidth data; and the first bandwidth value determining module includes:

A bandwidth upper limit determination submodule, used to determine a bandwidth upper limit value that each candidate first data port can receive the bandwidth data;

A state determination submodule, used to determine the state of each candidate first data port; the state includes an idle state and a used state;

a first bandwidth value determination submodule in an idle state, configured to use the bandwidth upper limit value as the first bandwidth value of the candidate first data port if the candidate first data port is in an idle state;

The first bandwidth value determination submodule in use is used to calculate the difference between the bandwidth upper limit value and the used bandwidth value if the candidate first data port is in use, and use the difference as the first bandwidth value of the candidate first data port.

Optionally, the device further comprises:

A test script transmission module, used for transmitting a test tool to the second server based on the second data port and the data interaction port connected in pairs, so that the second server deploys a test environment using the test tool;

The performance stress test starting module is used to start the performance stress test for the PCI-E device to be tested after the second server completes the deployment of the test environment.

Optionally, the device further comprises:

The second data exchange port determination module is configured to not use the candidate first data port corresponding to the first bandwidth value as the data exchange port of the second data port if the first bandwidth value is smaller than the second bandwidth value.

Optionally, the device further comprises:

The remaining used time determination module is used to determine and feedback the remaining used time of the first data port that meets the second bandwidth value if all the first bandwidth values are smaller than the second bandwidth value.

The embodiment of the present invention further discloses an electronic device, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;

The memory is used to store computer programs;

The processor is used to implement the couplet testing method described in the embodiment of the present invention when executing the program stored in the memory.

The embodiment of the present invention further discloses one or more computer-readable media on which instructions are stored. When executed by one or more processors, the processors execute the couplet testing method as described in the embodiment of the present invention.

Compared with the prior art, the embodiments of the present invention have the following advantages:

In an embodiment of the present invention, when the second server accesses the switch group as a client, the server obtains the second bandwidth value of the second data port of the PCI-E device to be tested on the second server, and then determines a plurality of target PCI-E devices to be coupled that match the type of the PCI-E device to be tested, determines the first data ports of the plurality of target PCI-E devices to be coupled as candidate first data ports, and determines the first bandwidth value of each candidate first data port, and then compares the second bandwidth value of the second data port with the first bandwidth value of each candidate first data port in sequence, if the first bandwidth value is not less than the second bandwidth value, then uses the candidate first data port corresponding to the first bandwidth value as the data exchange port of the second data port, and then allocates an IP address to the second data port, so as to connect the second data port and the data exchange port by using the IP address, so that the PCI-E device to be coupled corresponding to the data exchange port performs a coupling test with the PCI-E device to be tested. The embodiment of the present invention integrates multiple first servers into the server end, and uses the switch group as an intermediate medium, so that there is no need to build extra servers in the performance stress test. It only needs to connect the PCI-E device to be tested to the switch group, and the server end automatically obtains the second bandwidth value of the PCI-E device to be tested, and automatically matches the target PCI-E device to be coupled that is of the same type and can meet the second bandwidth value, so as to perform a coupled test on the two matched PCI-E devices, thereby greatly reducing the workload and greatly improving the test efficiency.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments 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 these drawings without creative work.

FIG1 is a flow chart of steps of a couplet testing method provided by an embodiment of the present invention;

FIG2 is a flowchart of another method for testing couplets provided by an embodiment of the present invention;

3 is a schematic diagram of the interaction between a first server, a switch group, and a second server provided in an embodiment of the present invention;

FIG. 4 is a structural block diagram of a couplet testing device provided by an embodiment of the present invention.

Detailed ways

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 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.

PCI-E is a high-speed serial point-to-point dual-channel high-bandwidth transmission. The connected devices are allocated exclusive channel bandwidth and do not share bus bandwidth. It mainly supports active power management, error reporting, end-to-end reliable transmission, hot plugging, and quality of service functions.

The server realizes the function of communication and information exchange by installing PCI-E devices. Based on the requirements of server design, PCI-E devices need to be subjected to performance stress testing before they can be added to the server's parts list. The performance stress test criteria of PCI-E devices mainly test the bandwidth rate of PCI-E devices.

The existing testing method requires the test engineer to pair two servers for testing, one server as the server and the other as the client, and connect the two servers. However, in each test, the test engineer needs to prepare two identical PCI-E devices and install them on the server and client respectively. Each time the server is built, a lot of debugging time is wasted and multiple identical PCI-E devices need to be prepared. Such testing principles undoubtedly increase the workload. The singleness of the test situation does not conform to the use principles in actual applications and is not conducive to the discovery of problems.

In view of the above problems, an embodiment of the present invention provides a pair test method. The embodiment of the present invention integrates multiple first servers into a server, and uses a switch group as an intermediate medium, so that there is no need to build extra servers in the performance stress test. It is only necessary to connect the PCI-E device to be tested to the switch group. The server automatically obtains the second bandwidth value of the PCI-E device to be tested, and automatically matches the target PCI-E device to be paired that is of the same type and can meet the second bandwidth value, so as to perform pair test on the two matched PCI-E devices, thereby greatly reducing the workload and greatly improving the test efficiency.

1, a flowchart of a pairing test method provided by an embodiment of the present invention is shown, which is applied to a server, the server is connected to a switch group, the server is integrated with a plurality of first servers, the plurality of first servers are respectively installed with a plurality of types of PCI-E devices to be paired, the plurality of PCI-E devices to be paired respectively have one or more first data ports, and the method may specifically include the following steps:

Step 101: When a second server is connected to the switch group as a client, a second bandwidth value of a second data port of a PCI-E device to be tested on the second server is obtained.

In an embodiment of the present invention, it is applied to the server side, and the server side is integrated with multiple first servers, and high-performance, high-specification servers can be used as first servers. Each first server is installed with multiple types of PCI-E devices to be coupled, and high-bandwidth PCI-E devices can be used as PCI-E devices to be coupled. For example, PCI-E devices with a bandwidth of 200G or above can be used as PCI-E devices to be coupled. In this way, according to the test strategy of high bandwidth compatibility with low bandwidth, each high-bandwidth PCI-E device to be coupled can simultaneously receive bandwidth data of multiple low-bandwidth PCI-E devices to be tested. Among them, the PCI-E devices to be coupled can include network cards and HCA cards, that is, each first server can be installed with multiple network cards and multiple HCA cards.

A network card is also called a network adapter (NIC) or a network interface card. The core of a network adapter is a link layer controller, which is usually a single chip with a specific purpose to implement many link layer services, including framing, link access, flow control, error detection, etc. A network adapter is a device that connects a computer to a network. A network card is usually a network adapter that connects a PC (Personal Computer) to a LAN. A network card is inserted into a slot on a computer motherboard and is responsible for converting the data that the user wants to transmit into a format that can be recognized by other devices on the network and transmitted through the network medium.

HCA cards can be used in enterprise data centers, high-performance computing, and embedded environments, providing high-bandwidth, low-latency solutions for server/storage cluster applications. In applications such as cluster databases, parallel applications, processing services, and high-performance embedded I/O, HCA cards can maximize performance, shorten operation cycles, and reduce resource loss. In a virtual server environment, ConnectX IB technology meets the needs of clustering, communication, storage, I/O management, and enhanced performance, simplifying the entire network configuration. These chips can be fully used in blade servers or motherboard designs due to their compact packaging.

It should be noted that the performance index evaluation criteria of the network card may include bandwidth, throughput and latency. Among them, bandwidth is used to evaluate the maximum transmission rate in the network link; throughput is used to evaluate the amount of data successfully transmitted per unit time; latency is used to indicate the delay time from the request sent from the network to the remote response. In different scenarios, latency has different meanings. For example, when requesting a connection, such as TCP (Transmission Control Protocol) handshake, it represents the delay of the connection, and in the transmission of ordinary data packets, it represents the time delay of the data packet to and from. In the consideration criteria of the performance stress test of PCI-E devices, the bandwidth performance of PCI-E devices is mainly tested.

In an embodiment of the present invention, the server is connected to a switch group, and the switch group can be obtained by cascading multiple switches in the laboratory together. The switch group serves as an intermediate medium, and its function is to provide more connection ports in the sub-network so as to connect more servers. A switch can be connected to multiple first servers, but in order to avoid multiple first servers connecting to the same switch competing for bandwidth resources, a switch can be connected to one first server. In addition, a connection quantity threshold can be set. For example, if the connection quantity threshold is set to 3, then a switch can be connected to no more than 3 first servers. The embodiment of the present invention does not limit the number of switch cascades.

In the embodiment of the present invention, the switch group can not only connect to the server, but also connect to the client, and the client is the second server to be tested. In other words, if the network card needs to be tested, the network card can be used as the PCI-E device to be tested, and the PCI-E device to be tested can be installed on the second server, and then the second server can be connected to the switch group as a client; if the HCA card needs to be tested, the HCA card can be used as the PCI-E device to be tested, and the PCI-E device to be tested can be installed on the second server, and then the second server can be connected to the switch group as a client.

The switch group can notify the server, and the server will carry out a pair test process for the PCI-E device to be tested. Each PCI-E device to be tested has one or more first data ports, and each PCI-E device to be tested has one or more second data ports. The server can first obtain the information of the PCI-E device to be tested on the second server. The information of the PCI-E device to be tested can include the second bandwidth value of the second data port. The second bandwidth value is the required bandwidth value of each second data port of the PCI-E device to be tested, so that the PCI-E device to be tested that can meet the second bandwidth value can be found in multiple first data ports based on the required bandwidth value, so that the matched PCI-E device to be tested and the PCI-E device to be tested can be tested in pair.

Step 102: determine a plurality of target PCI-E devices to be coupled that match the type of the PCI-E device to be tested.

In an embodiment of the present invention, there are various types of PCI-E devices to be coupled on the first server, and multiple first servers are integrated into the server, so it is necessary to find multiple target PCI-E devices to be coupled that match the type of the PCI-E device to be tested on the server. For example, if the PCI-E device to be tested is a network card, then it is necessary to find multiple target PCI-E devices to be coupled that belong to the network card type on the server; if the PCI-E device to be tested is an HCA card, then it is necessary to find multiple target PCI-E devices to be coupled that belong to the HCA card type on the server.

Step 103: determine the first data ports of the plurality of target PCI-E devices to be coupled as candidate first data ports, and determine the first bandwidth value of each candidate first data port.

In the embodiment of the present invention, after finding multiple target PCI-E devices of the same type to be coupled, each target PCI-E device to be coupled has one or more first data ports, so the first data port of each target PCI-E device to be coupled can be determined as a candidate first data port. The second data port and the candidate first data port are data ports of the same type, for example, if the PCI-E device to be tested is a network card, then the second data port and the candidate first data port are both network card data ports; if the PCI-E device to be tested is an HCA card, then the second data port and the candidate first data port are both HCA card data ports.

Just as the second data port has a corresponding second bandwidth value, each candidate first data port also has a corresponding first bandwidth value. Since the candidate first data port may be in use, that is, the candidate first data port may be occupied by the second data port of other PCI-E devices to be tested, that is, the first bandwidth value changes in real time, the server needs to determine the first bandwidth value of each candidate first data port at this time.

Step 104: Compare the second bandwidth value of the second data port with the first bandwidth values of the candidate first data ports in sequence.

In the embodiment of the present invention, each candidate first data port may be traversed in sequence to compare the second bandwidth value of the second data port with the first bandwidth value of each candidate first data port in sequence, thereby finding a candidate first data port that can meet the second bandwidth value.

Step 105: If the first bandwidth value is not less than the second bandwidth value, the candidate first data port corresponding to the first bandwidth value is used as the data exchange port of the second data port.

In an embodiment of the present invention, if the first bandwidth value is not less than the second bandwidth value, that is, the first bandwidth value is greater than or equal to the second bandwidth value, then the candidate first data port corresponding to the first bandwidth value can be used as the data interaction port of the second data port; if the first bandwidth value is less than the second bandwidth value, then there is no need to use the candidate first data port corresponding to the first bandwidth value as the data interaction port of the second data port.

Step 106: assign an IP address to the second data port.

In the embodiment of the present invention, after finding the data exchange port of the second data port, the server can assign an IP address to the second data port, and each second data port has an independent IP address. This is just like a computer. When you plug in a network cable, you will get an IP address. This IP address is automatically assigned to the client host by the DHCP server. DHCP (Dynamic Host Configuration Protocol) is a network protocol for a local area network that uses the UDP protocol. In addition to automatically assigning IP addresses to internal networks or network service providers, DHCP servers can also be used as a means of central management of all computers for users or internal network administrators.

The server can connect two data ports only after it has assigned an IP address. For example, in the network card test, the two data ports of a network card A on the first server have their own IP addresses, which are respectively connected to the two data ports of a network card B on the second server. In the performance stress test, the IP addresses of the two data ports will have data interaction, so there must be IP addresses that are connected to each other.

Step 107: Use the IP address to connect the second data port and the data exchange port, so that the PCI-E device to be coupled corresponding to the data exchange port performs a coupled test with the PCI-E device to be tested.

In an embodiment of the present invention, after the server allocates an IP address to the second data port, the server can use the IP address to connect the second data port and the data interaction port, so that the PCI-E device to be coupled corresponding to the data interaction port and the PCI-E device to be tested can be coupled to each other.

In summary, in an embodiment of the present invention, when the second server accesses the switch group as a client, the server obtains the second bandwidth value of the second data port of the PCI-E device to be tested on the second server, and then determines a plurality of target PCI-E devices to be coupled that match the type of the PCI-E device to be tested, determines the first data ports of the plurality of target PCI-E devices to be coupled as candidate first data ports, and determines the first bandwidth value of each candidate first data port, and then compares the second bandwidth value of the second data port with the first bandwidth value of each candidate first data port in turn, if the first bandwidth value is not less than the second bandwidth value, then uses the candidate first data port corresponding to the first bandwidth value as the data exchange port of the second data port, and then allocates an IP address to the second data port, so as to connect the second data port and the data exchange port by using the IP address, so that the PCI-E device to be coupled corresponding to the data exchange port performs a coupling test with the PCI-E device to be tested. The embodiment of the present invention integrates multiple first servers into the server end, and uses the switch group as an intermediate medium, so that there is no need to build extra servers in the performance stress test. It only needs to connect the PCI-E device to be tested to the switch group, and the server end automatically obtains the second bandwidth value of the PCI-E device to be tested, and automatically matches the target PCI-E device to be coupled that is of the same type and can meet the second bandwidth value, so as to perform a coupled test on the two matched PCI-E devices, thereby greatly reducing the workload and greatly improving the test efficiency.

2, a flowchart of another pairing test method provided by an embodiment of the present invention is shown, which is applied to a server, the server is connected to a switch group, the server is integrated with a plurality of first servers, the plurality of first servers are respectively installed with a plurality of types of PCI-E devices to be paired, the plurality of PCI-E devices to be paired respectively have one or more first data ports, and the method may specifically include the following steps:

Step 201: When a second server accesses the switch group as a client, a second bandwidth value of a second data port of a PCI-E device to be tested on the second server is obtained.

In an embodiment of the present invention, it is applied to the server side, and the server side is integrated with multiple first servers, and each first server is a high-performance, high-specification server. Each first server is installed with multiple types of PCI-E devices to be coupled, and each PCI-E device to be coupled is a high-bandwidth PCI-E device, such as 200G or above bandwidth. According to the test strategy of high bandwidth compatibility with low bandwidth, each high-bandwidth PCI-E device to be coupled can simultaneously receive bandwidth data of multiple low-bandwidth PCI-E devices to be tested. Among them, the PCI-E devices to be coupled can include network cards and HCA cards, that is, each first server can be installed with multiple network cards and multiple HCA cards.

Taking the network card test as an example, when testing the network card with a 10G bandwidth data port, the network card on the server can simultaneously receive data sources from at least 18 10G bandwidth data ports (considering data loss issues); when testing the network card with a 25G bandwidth data port, the network card on the server can simultaneously receive data sources from at least 6 25G ports (considering data loss issues).

The server is connected to a switch group, which can be obtained by cascading multiple switches. The switch group, as an intermediate medium, can provide more connection ports in the sub-network to connect more servers. Among them, one switch can be connected to one or more first servers, so the switch group can be connected to multiple first servers.

The switch group can not only connect to the server, but also to the client, which is the second server to be tested. In other words, if you need to test the network card, you can use the network card as the PCI-E device to be tested, install the PCI-E device to be tested on the second server, and then connect the second server to the switch group as a client; if you need to test the HCA card, you can use the HCA card as the PCI-E device to be tested, install the PCI-E device to be tested on the second server, and then connect the second server to the switch group as a client.

Wherein, each PCI-E device to be coupled has one or more first data ports, and each PCI-E device to be tested has one or more second data ports. Referring to FIG3, an interactive schematic diagram of a first server, a switch group, and a second server provided in an embodiment of the present invention is shown, and the server integrates three first servers, namely, a first server 1, a first server 2, and a first server 3, wherein the first server 1 is installed with an HCA card 1, a network card 2, and a network card 3, the first server 2 is installed with a network card 4 and an HCA card 2, and the first server 3 is installed with an HCA card 3, an HCA card 4, and a network card 5; further, the HCA card 1 has first data ports a to b, the network card 2 has a first data port c, the network card 3 has a first data port d to f, the network card 4 has a first data port g to h, the HCA card 2 has a first data port i to j, the HCA card 3 has a first data port k to l, the HCA card 4 has a first data port m to n, and the network card 5 has a first data port o to q. When the network card 1 needs to be tested, the network card 1 can be used as a PCI-E device to be tested, and then the PCI-E device to be tested is installed on the second server, and the second server is connected to the switch group as a client, wherein the network card 1 has second data ports X to Y.

The server may first obtain information about the PCI-E device to be tested on the second server. The information about the PCI-E device to be tested may include a second bandwidth value of the second data port. The second bandwidth value is a required bandwidth value of each second data port of the PCI-E device to be tested, so that a PCI-E device to be coupled that can meet the second bandwidth value can be found in multiple first data ports based on the required bandwidth value, thereby performing a coupled test on the matched PCI-E device to be tested and the PCI-E device to be coupled.

Step 202: determine a plurality of target PCI-E devices to be coupled that match the type of the PCI-E device to be tested.

In an embodiment of the present invention, there are various types of PCI-E devices to be coupled on the first server, and multiple first servers are integrated into the server, so it is necessary to find multiple target PCI-E devices to be coupled that match the type of the PCI-E device to be tested on the server. For example, if the PCI-E device to be tested is a network card, then it is necessary to find multiple target PCI-E devices to be coupled that belong to the network card type on the server; if the PCI-E device to be tested is an HCA card, then it is necessary to find multiple target PCI-E devices to be coupled that belong to the HCA card type on the server.

In an optional embodiment of the present invention, step 202 may include the following sub-steps:

Sub-step S11, receiving a data packet sent by the second server through the switch group;

Sub-step S12, sending the data packets to the multiple first servers respectively;

Sub-step S13, determining, in each first server, a PCI-E device to be coupled that can receive the data packet;

Sub-step S14, determining the PCI-E device to be paired that can receive the data packet as the target PCI-E device to be paired that matches the type of the PCI-E device to be tested.

In the embodiment of the present invention, the client only sends data packets to PCI-E devices of the same type. For example, if the client wants to test the network card, the client only sends data packets to the network card data port on the server; if the client wants to test the HCA card, the client only sends data packets to the HCA card data port on the server. Therefore, the client can send data packets to different first servers to find multiple target PCI-E devices to be paired that match the type of the PCI-E device to be tested.

In a specific implementation, the second server acts as a client and sends data packets to the server through a switch group. The server sends these data packets to multiple first servers respectively, and then in each first server, determines the PCI-E device to be coupled that can receive these data packets, thereby determining the PCI-E device to be coupled that can receive these data packets as the target PCI-E device to be coupled, and the target PCI-E device to be coupled is a PCI-E device to be coupled that is of the same type as the PCI-E device to be tested.

In one example, as shown in FIG3 , when the client starts testing, the stress test script can be run, and the second server sends data packets to the first server 1 for one hour, then sends data packets to the first server 2 for one hour, and then sends data packets to the first server 3 for one hour… Based on this test principle, the stress test script is run on the client for more than 12 hours, and data packets are sent to different first servers with a time limit of one hour for testing.

As can be seen from Figure 3, the PCI-E device to be tested on the second server is a network card type, so network card 2 and network card 3 on the first server 1 can receive data packets, and network card 4 on the first server 2 can receive data packets, and network card 5 on the first server 3 can receive data packets, so network cards 2 to 5 are all target PCI-E devices to be coupled that match the type of the PCI-E device to be tested.

Step 203: determine the first data ports of the plurality of target PCI-E devices to be coupled as candidate first data ports, and determine the first bandwidth value of each candidate first data port.

In the embodiment of the present invention, after finding multiple target PCI-E devices of the same type to be coupled, each target PCI-E device to be coupled has one or more first data ports, so the first data port of each target PCI-E device to be coupled can be determined as a candidate first data port. The second data port and the candidate first data port are data ports of the same type, for example, if the PCI-E device to be tested is a network card, then the second data port and the candidate first data port are both network card data ports; for example, if the PCI-E device to be tested is an HCA card, then the second data port and the candidate first data port are both HCA card data ports.

In an example, as shown in Figure 3, network cards 2 to 5 are all target PCI-E devices to be paired that match the type of the PCI-E device to be tested, wherein network card 2 has a first data port c, network card 3 has first data ports d to f, network card 4 has first data ports g to h, and network card 5 has first data ports o to q, so the first data ports c, d, e, f, g, h, o, p, and q are all candidate first data ports.

Just as the second data port has a corresponding second bandwidth value, each candidate first data port also has a corresponding first bandwidth value. Since the candidate first data port may be in use, that is, the candidate first data port may be occupied by the second data port of other PCI-E devices to be tested, that is, the first bandwidth value changes in real time, the server needs to determine the first bandwidth value of each candidate first data port at this time.

In an optional embodiment of the present invention, the data packet includes bandwidth data; step 203 may include the following sub-steps:

Sub-step S21, determining the upper limit of the bandwidth that each candidate first data port can receive the bandwidth data;

Sub-step S22, determining the state of each candidate first data port; the state includes an idle state and a used state;

Sub-step S23, if the candidate first data port is in an idle state, using the bandwidth upper limit value as the first bandwidth value of the candidate first data port;

Sub-step S24: If the candidate first data port is in use, then the difference between the bandwidth upper limit value and the used bandwidth value is calculated, and the difference is used as the first bandwidth value of the candidate first data port.

In an embodiment of the present invention, a data packet may include bandwidth data. Each first server may collect bandwidth data each time, analyze the bandwidth data, and determine the bandwidth upper limit value of the bandwidth data that each candidate first data port can receive. The bandwidth upper limit value is the maximum bandwidth data that the candidate first data port can receive, which is equivalent to the maximum received data value. Therefore, the maximum received data value can be used as the bandwidth upper limit value of the bandwidth data that the candidate first data port can receive.

Since the candidate first data port may be occupied by the second data port of other PCI-E devices to be tested, the server needs to determine the state of each candidate first data port, where the state may include an idle state and a used state.

If the candidate first data port is in an idle state, it can be explained that the candidate first data port is not occupied by the second data port of other PCI-E devices to be tested, so the bandwidth upper limit value can be directly used as the first bandwidth value of the candidate first data port; if the candidate first data port is in use, it can be explained that the candidate first data port is occupied by the second data port of other PCI-E devices to be tested, so it is necessary to calculate the remaining bandwidth space of the candidate first data port, and the remaining bandwidth space can be obtained by subtracting the used bandwidth value from the bandwidth upper limit value of the candidate first data port, and the difference obtained is used as the first bandwidth value of the candidate first data port.

Step 204: Compare the second bandwidth value of the second data port with the first bandwidth values of the candidate first data ports in sequence.

In the embodiment of the present invention, each candidate first data port may be traversed in sequence to compare the second bandwidth value of the second data port with the first bandwidth value of each candidate first data port in sequence, thereby finding a candidate first data port that can meet the second bandwidth value.

In addition, in order to facilitate the subsequent automatic allocation of candidate first data ports that meet the bandwidth requirements, and to prevent the phenomenon that the total bandwidth of the second server is greater than the total bandwidth of the first server, for example, if the first server is 200G and the second server with a bandwidth of 100G is tested in parallel, it can be performed at full speed. On the contrary, if the first server is 100G and a second server with a bandwidth of 200G needs to be tested in parallel, the limit performance of the PCI-E device to be tested cannot be tested, resulting in inaccurate testing. Therefore, before comparing the second bandwidth value of the second data port with the first bandwidth value of the candidate first data port, the server can first determine the bandwidth upper limit value of the second server and the bandwidth upper limit value of each first server, and then compare the bandwidth upper limit value of the second server with the bandwidth upper limit value of each first server in turn. If the bandwidth upper limit value of a first server is greater than or equal to the bandwidth upper limit value of the second server, then the step of comparing the second bandwidth value of the second data port with the first bandwidth value of the candidate first data port can be performed for the first server; if the bandwidth upper limit value of a first server is less than the bandwidth upper limit value of the second server, then the first server can be skipped, that is, there is no need to perform the step of comparing the second bandwidth value of the second data port with the first bandwidth value of the candidate first data port for the first server.

In a specific implementation, each first server can collect bandwidth data each time, analyze the bandwidth data, calculate the bandwidth upper limit value of the bandwidth data that can be received by all the PCI-E devices to be coupled on the first server, and use the bandwidth upper limit value of the bandwidth data that can be received by all the PCI-E devices to be coupled as the bandwidth upper limit value of the bandwidth data that can be received by the first server. Specifically, the bandwidth upper limit values of the first data ports of the PCI-E devices to be coupled can be added to obtain the bandwidth upper limit value of the bandwidth data that can be received by the PCI-E devices to be coupled, and then the bandwidth upper limit values of the PCI-E devices to be coupled on the first server can be added to obtain the bandwidth upper limit value of the bandwidth data that can be received by the first server. Similarly, the bandwidth upper limit value of the second server can be obtained by adding the second bandwidth values of the second data ports of the PCI-E devices to be coupled on the second server.

Step 205: If the first bandwidth value is not less than the second bandwidth value, the candidate first data port corresponding to the first bandwidth value is used as the data exchange port of the second data port.

In the embodiment of the present invention, if the first bandwidth value is not less than the second bandwidth value, that is, the first bandwidth value is greater than or equal to the second bandwidth value, then the candidate first data port corresponding to the first bandwidth value can be used as the data exchange port of the second data port.

In an optional embodiment of the present invention, the method may further include:

If the first bandwidth value is smaller than the second bandwidth value, the candidate first data port corresponding to the first bandwidth value is not used as the data exchange port of the second data port.

In an embodiment of the present invention, if the first bandwidth value is smaller than the second bandwidth value, then the candidate first data port corresponding to the first bandwidth value may not be used as the data interaction port of the second data port, and other candidate first data ports may be traversed until a data interaction port that meets the bandwidth requirements is found.

In an example, as shown in FIG3 , the candidate first data ports include first data ports c, d, e, f, g, h, o, p, and q, and the PCI-E device to be tested (network card 1) has second data ports X and Y. At this time, the second bandwidth value X of the second data port X can be compared with the first bandwidth value c of the candidate first data port c. If the first bandwidth value c is greater than or equal to the second bandwidth value X, then the candidate first data port c can be used as the data exchange port of the second data port X; if the first bandwidth value c is less than the second bandwidth value X, then the candidate first data port c cannot be used as the data exchange port of the second data port X, and other candidate first data ports can be traversed at this time. Specifically, if the first bandwidth value c is less than the second bandwidth value X, then the second bandwidth value X of the second data port X can be compared with the first bandwidth value d of the candidate first data port d, ..., if the first bandwidth values of the candidate first data ports c, d, e, and f on the first server 1 are all less than the second bandwidth value of the second data port X, then the second bandwidth value of the second data port X can be compared with the first bandwidth values of the candidate first data ports g and h on the first server 2 in sequence, ..., if the first bandwidth values of the candidate first data ports g and h on the first server 2 are all less than the second bandwidth value of the second data port X, then the second bandwidth value of the second data port X can be compared with the first bandwidth values of the candidate first data ports o, p, and q on the first server 3 in sequence, until a data interaction port that meets the bandwidth requirements is found.

It should be noted that the comparison step of the second data port Y is the same as the comparison step of the second data port X, which will not be described in detail herein.

The above examples are only used to help those skilled in the art better understand the embodiments of the present invention, and the present invention is not limited thereto.

In an optional embodiment of the present invention, the multiple first servers respectively deploy different first system environments; step 205 may include the following sub-steps:

Sub-step S31, if the first bandwidth value is not less than the second bandwidth value, determining a second system environment in which the second server is deployed, and determining a target first system environment in which the candidate first data port corresponding to the first bandwidth value is located;

Sub-step S32, determining whether the target first system environment is consistent with the second system environment;

Sub-step S33: if the target first system environment is consistent with the second system environment, using the candidate first data port corresponding to the first bandwidth value as the data exchange port of the second data port.

In an embodiment of the present invention, the server side always maintains a data receiving state, and different first servers on the server side can deploy different first system environments. For example, the first server 1 deploys a centos8.2 system, the first server 2 deploys a Redhat8.2 system, the first server 3 deploys a Windows2022 system, the first server 4 deploys an Ubuntu12.06 system... and so on.

The second server is deployed with a second system environment. In order to make the two PCI-E devices in the joint test in the same system environment, the server can further determine the second system environment deployed by the second server when the first bandwidth value is not less than the second bandwidth value, and determine the target first system environment where the candidate first data port corresponding to the first bandwidth value is located, and then determine whether the target first system environment is consistent with the second system environment. If the second system environment is a Redhat8.2 system and the target first system environment is also a Redhat8.2 system, then it can be determined that the target first system environment is consistent with the second system environment, so the candidate first data port corresponding to the first bandwidth value can be used as the data interaction port of the second data port; if the second system environment is a Redhat8.2 system and the target first system environment is a Windows2022 system, then it can be determined that the target first system environment is inconsistent with the second system environment, so there is no need to use the candidate first data port corresponding to the first bandwidth value as the data interaction port of the second data port.

Step 206: assign an IP address to the second data port.

In the embodiment of the present invention, after finding the data exchange port of the second data port, the server can allocate an IP address to the second data port, and each second data port has an independent IP address. The IP address can be used to connect the second data port and the data exchange port, so that the two matched PCI-E devices can be tested in parallel.

Step 207: Use the IP address to connect the second data port and the data exchange port, so that the PCI-E device to be coupled corresponding to the data exchange port performs a coupled test with the PCI-E device to be tested.

In an embodiment of the present invention, after the server allocates an IP address to the second data port, the server can use the IP address to connect the second data port and the data interaction port, so that the PCI-E device to be coupled corresponding to the data interaction port and the PCI-E device to be tested can be coupled to each other.

In an optional embodiment of the present invention, the method may further include:

Based on the second data port and the data interaction port connected in pairs, the test tool is transmitted to the second server so that the second server deploys the test environment using the test tool; after the second server completes the deployment of the test environment, the performance stress test of the PCI-E device to be tested is started.

In the embodiment of the present invention, after the second data port and the data exchange port are connected by an IP address, the server can transmit the test tool to the second server based on the second data port and the data exchange port connected in pairs. The test tool is used to deploy the test environment. After the second server receives the test tool through the switch group, the test tool can be automatically installed, thereby automatically deploying the test environment. After the second server completes the deployment of the test environment, the server can start the performance stress test for the PCI-E device to be tested.

In an optional embodiment of the present invention, the method may further include:

If all the first bandwidth values are smaller than the second bandwidth value, it is determined to feed back the remaining usage time of the first data port that satisfies the second bandwidth value.

In an embodiment of the present invention, if after traversing all candidate first data ports, no data interaction port that meets the bandwidth space requirements of the PCI-E device to be tested is found, that is, all first bandwidth values are less than the second bandwidth value, then the remaining usage time of the feedback test engineer can be determined, and the remaining usage time refers to the remaining time of the first data port that meets the second bandwidth value. In other words, all first data ports on the server are in use, and there are no other redundant first data ports available for testing. At this time, there will be a test sequence, for example, the test is generally 24 hours, then the remaining time can be calculated according to the test sequence, and when a first data port test is completed, an instruction can be sent to notify the test engineer to test the second server, so as to achieve reasonable allocation.

Compared with the existing test method, the test engineer needs to match two servers, one server as the server and the other server as the client, and connect the two servers. Specifically, in the current actual test work, testing an HCA card or network card requires another server to be matched with it for testing. For example, when testing a 100G HCA card, the HCA card to be tested is installed on a server, but to cooperate with the test of this HCA card, it is necessary to build another server to install a 100G or greater bandwidth HCA card for matching testing. When deploying two servers for testing, it is also necessary to deploy the same system environment in the two servers. For example, if the client installs the Redhat7.8 system, the server also needs to install the Redhat7.8 system to cooperate with the test. Sometimes, due to compatibility issues, the same HCA card needs to be installed on the client and the server for testing. It can be seen that such a test principle undoubtedly increases the workload, and the singleness of the test situation does not conform to the use principle in actual applications, which is not conducive to the discovery of problems.

The embodiment of the present invention obtains a switch group by cascading multiple switches, integrates multiple high-performance, high-specification first servers into the server, uniformly deploys the server as a receiving end, and configures a high-bandwidth PCI-E device to be coupled. According to the test strategy of high bandwidth compatibility with low bandwidth, each high-bandwidth PCI-E device to be coupled can simultaneously receive bandwidth data of multiple low-bandwidth PCI-E devices to be tested. Different first system environments are deployed in different first servers, and the second server is connected to the switch group as a client. The second server implements the matching principle of multiple situations by sending data packets to different first servers to achieve the effect of real simulation application scenarios, which is convenient for discovering more problems. The server collects and analyzes the maximum bandwidth requirements that can be met when interacting with the client, and each first server defines a bandwidth upper limit value, and collects and analyzes the maximum received data value of each first data port of each PCI-E device to be coupled, and defines the maximum received data value as the bandwidth upper limit value of the first data port. The server calculates the remaining bandwidth space of the first data port according to the bandwidth upper limit value and the real-time interaction data, compares the remaining bandwidth space of the first data port with the second bandwidth value of the second data port, and selects the candidate first data port that meets the bandwidth requirement as the data interaction port of the second data port of the PCI-E device to be tested. The server can automatically allocate an IP address to the second data port, thereby connecting the second data port and the data interaction port using the IP address, and then transmits the test tool to the second server, so that the second server uses the test tool to automatically deploy the test environment, thereby achieving the purpose of automated and intelligent testing.

In summary, in an embodiment of the present invention, when the second server accesses the switch group as a client, the server obtains the second bandwidth value of the second data port of the PCI-E device to be tested on the second server, and then determines a plurality of target PCI-E devices to be coupled that match the type of the PCI-E device to be tested, determines the first data ports of the plurality of target PCI-E devices to be coupled as candidate first data ports, and determines the first bandwidth value of each candidate first data port, and then compares the second bandwidth value of the second data port with the first bandwidth value of each candidate first data port in turn, if the first bandwidth value is not less than the second bandwidth value, then uses the candidate first data port corresponding to the first bandwidth value as the data exchange port of the second data port, and then allocates an IP address to the second data port, so as to connect the second data port and the data exchange port by using the IP address, so that the PCI-E device to be coupled corresponding to the data exchange port performs a coupling test with the PCI-E device to be tested. The embodiment of the present invention integrates multiple first servers into the server end, and uses the switch group as an intermediate medium, so that there is no need to build extra servers in the performance stress test. It only needs to connect the PCI-E device to be tested to the switch group, and the server end automatically obtains the second bandwidth value of the PCI-E device to be tested, and automatically matches the target PCI-E device to be coupled that is of the same type and can meet the second bandwidth value, so as to perform a coupled test on the two matched PCI-E devices, thereby greatly reducing the workload and greatly improving the test efficiency.

Referring to FIG4 , a structural block diagram of a pairing test device provided by an embodiment of the present invention is shown, which is applied to a server end, the server end is connected to a switch group, the server end is integrated with a plurality of first servers, the plurality of first servers are respectively installed with a plurality of types of PCI-E devices to be paired, the plurality of PCI-E devices to be paired respectively have one or more first data ports, and specifically may include the following modules:

A second bandwidth value acquisition module 401 is used to acquire a second bandwidth value of a second data port of a PCI-E device to be tested on the second server when the second server accesses the switch group as a client;

A target PCI-E device to be coupled determination module 402 is used to determine a plurality of target PCI-E devices to be coupled that match the type of the PCI-E device to be tested;

A first bandwidth value determination module 403, configured to determine the first data ports of the plurality of target PCI-E devices to be coupled as candidate first data ports, and determine a first bandwidth value of each candidate first data port;

A bandwidth value comparison module 404, configured to compare the second bandwidth value of the second data port with the first bandwidth values of the candidate first data ports in sequence;

A first data exchange port determination module 405, configured to use the candidate first data port corresponding to the first bandwidth value as the data exchange port of the second data port if the first bandwidth value is not less than the second bandwidth value;

An IP address allocation module 406, configured to allocate an IP address to the second data port;

The connection module 407 is used to connect the second data port and the data exchange port by using the IP address, so that the PCI-E device to be coupled corresponding to the data exchange port can be coupled with the PCI-E device to be tested.

In an optional embodiment of the present invention, the multiple first servers are respectively deployed in different first system environments; the first data interaction port determination module 405 may include:

a system environment determination submodule, configured to determine, if the first bandwidth value is not less than the second bandwidth value, a second system environment in which the second server is deployed, and determine a target first system environment in which the candidate first data port corresponding to the first bandwidth value is located;

A determination submodule, used to determine whether the target first system environment is consistent with the second system environment;

The first data exchange port determination submodule is configured to use the candidate first data port corresponding to the first bandwidth value as the data exchange port of the second data port if the target first system environment is consistent with the second system environment.

In an optional embodiment of the present invention, the target to-be-associated PCI-E device determination module 402 may include:

A data packet receiving submodule, configured to receive a data packet sent by the second server through the switch group;

A data packet sending submodule, used for sending the data packets to the multiple first servers respectively;

A submodule for determining a PCI-E device to be paired that can receive a data packet, used to determine, in each first server, a PCI-E device to be paired that can receive the data packet;

The target PCI-E device to be coupled determination submodule is used to determine the PCI-E device to be coupled that can receive the data packet as the target PCI-E device to be coupled that matches the type of the PCI-E device to be tested.

In an optional embodiment of the present invention, the data packet includes bandwidth data; the first bandwidth value determination module 403 may include:

A bandwidth upper limit determination submodule, used to determine a bandwidth upper limit value that each candidate first data port can receive the bandwidth data;

A state determination submodule, used to determine the state of each candidate first data port; the state includes an idle state and a used state;

a first bandwidth value determination submodule in an idle state, configured to use the bandwidth upper limit value as the first bandwidth value of the candidate first data port if the candidate first data port is in an idle state;

The first bandwidth value determination submodule in use is used to calculate the difference between the bandwidth upper limit value and the used bandwidth value if the candidate first data port is in use, and use the difference as the first bandwidth value of the candidate first data port.

In an optional embodiment of the present invention, the device may further include:

A test script transmission module, used for transmitting a test tool to the second server based on the second data port and the data interaction port connected in pairs, so that the second server deploys a test environment using the test tool;

The performance stress test starting module is used to start the performance stress test for the PCI-E device to be tested after the second server completes the deployment of the test environment.

In an optional embodiment of the present invention, the device may further include:

The second data exchange port determination module is configured to not use the candidate first data port corresponding to the first bandwidth value as the data exchange port of the second data port if the first bandwidth value is smaller than the second bandwidth value.

In an optional embodiment of the present invention, the device may further include:

The remaining used time determination module is used to determine and feedback the remaining used time of the first data port that meets the second bandwidth value if all the first bandwidth values are smaller than the second bandwidth value.

In summary, in an embodiment of the present invention, when the second server accesses the switch group as a client, the server obtains the second bandwidth value of the second data port of the PCI-E device to be tested on the second server, and then determines a plurality of target PCI-E devices to be coupled that match the type of the PCI-E device to be tested, determines the first data ports of the plurality of target PCI-E devices to be coupled as candidate first data ports, and determines the first bandwidth value of each candidate first data port, and then compares the second bandwidth value of the second data port with the first bandwidth value of each candidate first data port in turn, if the first bandwidth value is not less than the second bandwidth value, then uses the candidate first data port corresponding to the first bandwidth value as the data exchange port of the second data port, and then allocates an IP address to the second data port, so as to connect the second data port and the data exchange port using the IP address, so that the PCI-E device to be coupled corresponding to the data exchange port performs a coupling test with the PCI-E device to be tested. The embodiment of the present invention integrates multiple first servers into the server end, and uses the switch group as an intermediate medium, so that there is no need to build extra servers in the performance stress test. It only needs to connect the PCI-E device to be tested to the switch group, and the server end automatically obtains the second bandwidth value of the PCI-E device to be tested, and automatically matches the target PCI-E device to be coupled that is of the same type and can meet the second bandwidth value, so as to perform a coupled test on the two matched PCI-E devices, thereby greatly reducing the workload and greatly improving the test efficiency.

As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

An embodiment of the present invention also provides an electronic device, including: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, each process of the above-mentioned couplet test method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the various processes of the above-mentioned couplet test method embodiment are implemented and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

Those skilled in the art will appreciate that the embodiments of the embodiments of the present invention may be provided as methods, devices, or computer program products. Therefore, the embodiments of the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the embodiments of the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

The embodiments of the present invention are described with reference to the flowcharts and/or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and/or box in the flowchart and/or block diagram, as well as the combination of the processes and/or boxes in the flowchart and/or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate a device for implementing the functions specified in one process or multiple processes in the flowchart and/or one box or multiple boxes in the block diagram.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device that implements the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce computer-implemented processing, so that the instructions executed on the computer or other programmable terminal device provide steps for implementing the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.

Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

The couplet testing method, device, electronic device and computer-readable storage medium provided by the present invention are introduced in detail above. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims (10)

1. A pairing test method, characterized in that it is applied to a server, the server is connected to a switch group, the server is integrated with a plurality of first servers, the plurality of first servers are respectively installed with a plurality of types of high-speed serial computer expansion bus standard PCI-E devices to be paired, the plurality of PCI-E devices to be paired respectively have one or more first data ports, the method comprising: When the second server accesses the switch group as a client, obtaining a second bandwidth value of a second data port of the PCI-E device to be tested on the second server; Determine a plurality of target PCI-E devices to be coupled that match the type of the PCI-E device to be tested; Determine the first data ports of the plurality of target PCI-E devices to be coupled as candidate first data ports, and determine the first bandwidth value of each candidate first data port; Compare the second bandwidth value of the second data port with the first bandwidth values of each candidate first data port in sequence; If the first bandwidth value is not less than the second bandwidth value, taking the candidate first data port corresponding to the first bandwidth value as the data exchange port of the second data port; Allocating an Internet Protocol address (IP) address to the second data port; The second data port and the data exchange port are connected by using the IP address, so that the PCI-E device to be coupled corresponding to the data exchange port is subjected to a coupling test with the PCI-E device to be tested. 2. The method according to claim 1, wherein the plurality of first servers are respectively deployed in different first system environments; and if the first bandwidth value is not less than the second bandwidth value, using the candidate first data port corresponding to the first bandwidth value as the data interaction port of the second data port comprises: If the first bandwidth value is not less than the second bandwidth value, determining a second system environment in which the second server is deployed, and determining a target first system environment in which the candidate first data port corresponding to the first bandwidth value is located; Determining whether the target first system environment is consistent with the second system environment; If the target first system environment is consistent with the second system environment, the candidate first data port corresponding to the first bandwidth value is used as the data exchange port of the second data port. 3. The method according to claim 2, wherein determining a plurality of target PCI-E devices to be coupled that match the type of the PCI-E device to be tested comprises: receiving, through the switch group, a data packet sent by the second server; Sending the data packets to the multiple first servers respectively; In each first server, determining a PCI-E device to be coupled that can receive the data packet; The PCI-E device to be paired that can receive the data packet is determined as a target PCI-E device to be paired that matches the type of the PCI-E device to be tested. 4. The method according to claim 3, wherein the data packet includes bandwidth data; and the determining the first bandwidth value of each candidate first data port comprises: Determine an upper bandwidth limit value of each candidate first data port that can receive the bandwidth data; Determine the state of each candidate first data port; the state includes an idle state and a used state; If the candidate first data port is in an idle state, using the bandwidth upper limit value as the first bandwidth value of the candidate first data port; If the candidate first data port is in use, the difference between the bandwidth upper limit value and the used bandwidth value is calculated, and the difference is used as the first bandwidth value of the candidate first data port. 5. The method according to claim 1, characterized in that the method further comprises: Based on the second data ports and the data exchange ports connected in pairs, transmitting the test tool to the second server, so that the second server deploys the test environment using the test tool; After the second server completes the deployment of the test environment, a performance stress test for the PCI-E device to be tested is started. 6. The method according to claim 1, characterized in that the method further comprises: If the first bandwidth value is smaller than the second bandwidth value, the candidate first data port corresponding to the first bandwidth value is not used as the data exchange port of the second data port. 7. The method according to claim 1, characterized in that the method further comprises: If all the first bandwidth values are smaller than the second bandwidth value, it is determined to feed back the remaining usage time of the first data port that satisfies the second bandwidth value. 8. A pairing test device, characterized in that it is applied to a server, the server is connected to a switch group, the server is integrated with a plurality of first servers, the plurality of first servers are respectively installed with a plurality of types of PCI-E devices to be paired, the plurality of PCI-E devices to be paired respectively have one or more first data ports, and the device comprises: A second bandwidth value acquisition module, configured to acquire a second bandwidth value of a second data port of a PCI-E device to be tested on the second server when the second server accesses the switch group as a client; A target PCI-E device to be coupled determination module is used to determine a plurality of target PCI-E devices to be coupled that match the type of the PCI-E device to be tested; A first bandwidth value determination module, configured to determine the first data ports of the plurality of target PCI-E devices to be coupled as candidate first data ports, and determine a first bandwidth value of each candidate first data port; a bandwidth value comparison module, configured to compare the second bandwidth value of the second data port with the first bandwidth values of the candidate first data ports in sequence; A first data exchange port determination module, configured to use the candidate first data port corresponding to the first bandwidth value as the data exchange port of the second data port if the first bandwidth value is not less than the second bandwidth value; An IP address allocation module, used for allocating an IP address to the second data port; The connection module is used to connect the second data port and the data exchange port by using the IP address, so that the PCI-E device to be coupled corresponding to the data exchange port can be coupled with the PCI-E device to be tested. 9. An electronic device, characterized in that it comprises: a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the computer program is executed by the processor, the steps of the couplet testing method according to any one of claims 1 to 7 are implemented. 10. A computer-readable storage medium, characterized in that a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps of the couplet testing method according to any one of claims 1 to 7 are implemented.