A server system and a data processing method
By uninstalling the card and connecting to heterogeneous cards, the problem of server resource occupation and management complexity in cloud scenarios is solved, the server is lightweight and security is improved, and diverse needs are supported.
Patent Information
- Application Number
- CN202011193205.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-10-30
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2040-10-30
AI Technical Summary
In cloud scenarios, the expansion of servers directly connecting to heterogeneous cards increases the complexity of the management surface and occupies a large amount of resources, resulting in the server not complying with the trend of lightweighting, and there are problems of security and lack of flexibility.
By uninstalling the card and connecting to the heterogeneous card, the uninstalling card is responsible for handling the control surface of the second virtualization device of the heterogeneous card, realizing communication between the server and the heterogeneous card, saving server resources, and isolating the abnormal queue through the management surface of the uninstalling card , solve security problems.
It realizes resource savings on servers, reduces management complexity, improves system flexibility and security, and supports diversified cloud customer needs.
Smart Images

Figure CN114528032B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of computer technology, and in particular to a server system and a data processing method. Background Art
[0002] In a cloud scenario, cloud instances are deployed on servers, such as virtual machines (VMs), etc. In order to improve the processing speed of input / output (I / O) services, some I / O services in the server can be offloaded to low-cost devices for execution, thereby freeing up the server's central processing unit (CPU) resources. These devices for offloading I / O services are usually called offload cards, such as application specific integrated circuits (ASICs) or field programmable gate arrays (FPGAs), etc.
[0003] Under the special requirements of some cloud scenarios, it is necessary to introduce additional heterogeneous cards, such as graphics processing unit (GPU), network interface controller (NIC), etc. Generally, these heterogeneous cards can be directly connected to the server, so that the offload card and the heterogeneous card occupy different high-speed serial computer expansion bus standard (Peripheral Component Interconnect Express, PCIE) interfaces of the server respectively.
[0004] Heterogeneous cards usually support single-root I / O virtualization (SRIOV) function, so heterogeneous cards can create virtualized devices. Since these virtualized devices are directly connected to the cloud instance of the server, the heterogeneous card can communicate with the cloud instance of the server based on these virtualized devices.
[0005] In the above solution, the server directly connects to the expansion of heterogeneous cards, which increases the complexity of the management plane. Therefore, the front-end server will occupy a lot of resources, which are used to process the management and control plane of the virtualized devices of the heterogeneous cards, which does not conform to the lightweight trend of the front-end server. Summary of the invention
[0006] The embodiment of the present application provides a server system and a method for data processing. By connecting an offload card to an expansion of a heterogeneous card, the offload card can be responsible for processing the control plane of the second virtualization device of the heterogeneous card, which can save server resources and help promote the lightweight of the server.
[0007] The first aspect of the embodiment of the present application provides a server system, which includes: a server, an offload card and a heterogeneous card, wherein the offload card is connected to the server, and the heterogeneous card is connected to the offload card. The offload card is used to record the mapping relationship between the first virtualization device of the offload card and the second virtualization device of the heterogeneous card, and the communication between the cloud instance of the server and the second virtualization device of the heterogeneous card is realized according to the mapping relationship, wherein the first virtualization device of the offload card is directly connected to the cloud instance of the server, and the second virtual device of the heterogeneous card is directly connected to the offload card.
[0008] In the above-mentioned server system, the offload card is connected between the server and the heterogeneous card. Since the first virtualization device of the offload card is directly connected to the cloud instance of the server, the offload card can realize communication with the cloud instance of the server based on the first virtualization device. Furthermore, since the second virtual device of the heterogeneous card is directly connected to the offload card, the heterogeneous card can realize communication with the offload card based on the second virtual device. Therefore, the offload card can serve as a bridge between the heterogeneous card and the cloud instance of the server to realize communication between the heterogeneous card and the cloud instance of the server. Since the offload card is directly connected to the extension of the heterogeneous card, the offload card can be responsible for processing the management and control surface of the second virtualization device of the heterogeneous card, which can save server resources and help promote the lightweight of the server.
[0009] In a possible implementation, the server includes a first interface, the offload card includes a second interface and a third interface, and the heterogeneous card includes a fourth interface; the first interface is connected to the second interface, and the third interface is connected to the fourth interface. In the aforementioned implementation, the offload card itself has two interfaces, the offload card is connected to the server through one of the interfaces, and is connected to the heterogeneous card through the other interface, so that the offload card is connected between the offload card and the server.
[0010] In a possible implementation, the server system also includes an expansion card, the server includes a first interface, the expansion card includes a second interface, a third interface, a fourth interface and a fifth interface, the uninstall card includes a sixth interface and a seventh interface, and the heterogeneous card includes an eighth interface; the first interface is connected to the second interface, the second interface is connected to the third interface, the fourth interface is connected to the fifth interface, the sixth interface is connected to the third interface, the seventh interface is connected to the fourth interface, and the eighth interface is connected to the fifth interface. In the aforementioned implementation, the server, the uninstall card and the heterogeneous card are respectively connected to different interfaces of the same expansion card, and the connection relationship between the interfaces of the expansion card is set so that the uninstall card is connected between the uninstall card and the server.
[0011] In a possible implementation, the heterogeneous card supports a first protocol, the offload card is loaded with a first driver corresponding to the first protocol, the server supports a second protocol, and the cloud instance of the server is loaded with a second driver corresponding to the second protocol; the offload card is specifically used to receive a first operation request from a second virtualization device through the first driver, and the first operation request complies with the format requirements of the first protocol; the offload card is specifically used to convert the first operation request into a second operation request, and the second operation request complies with the format requirements of the second protocol; the offload card is specifically used to send the second operation request to the second driver through the first virtualization device, so that the cloud instance of the server reads the target data from the memory of the cloud instance of the server or writes the target data to the memory of the cloud instance of the server based on the second operation request. In the above implementation, after the offload card receives the first operation request from the second virtualization device of the heterogeneous card through the first driver, the first operation request can be converted into the second operation request, and the first operation request complies with the format requirements of the first protocol, and the second operation request complies with the format requirements of the second protocol. Then, the offload card sends the first operation request to the second driver of the cloud instance of the server through the first virtualization device, so that the cloud instance of the server completes the corresponding operation based on the first operation request. Therefore, in the server system provided in the present application, even if the cloud instance of the server only loads the second driver, communication between the cloud instance of the server and the heterogeneous card can be achieved.
[0012] In a possible implementation, the offload card includes an engine module supporting the second protocol; the offload card is specifically used to load the first driver after determining that the heterogeneous card is started; the offload card is specifically used to obtain the device information of the heterogeneous card through the first driver; the offload card is specifically used to configure the engine module as a first virtualized device based on configuration information, and the configuration information is preset information; the offload card is specifically used to receive the enumeration result from the second driver through the first virtualized device, and the enumeration result is used to indicate that the cloud instance of the server has detected the first virtualized device; the offload card is specifically used to construct a mapping relationship between the second virtualized device of the heterogeneous card and the first virtualized device of the offload card based on the device information and the enumeration result. In the above implementation, the offload card can configure the engine module as the first virtualized device based on the configuration information, and can construct a mapping relationship between the first virtualized device of the offload card and the second virtualized device of the heterogeneous card based on the enumeration result of the cloud instance of the server and the device information of the heterogeneous card. For the server side, the engine module can be presented as the first virtualized device, so the cloud instance of the server can send information to the first virtualized device according to its own needs. After the offload card receives the information through the first virtualization device, it can send the information to the second virtualization device of the heterogeneous card based on the mapping relationship between the first virtualization device of the offload card and the second virtualization device of the heterogeneous card, thereby realizing communication between the cloud instance of the server and the heterogeneous card.
[0013] In a possible implementation, if the first operation request is a first read request, the offload card is specifically used to receive a second send request from a second driver through a first virtualization device, the second send request conforms to the format requirements of the second protocol, and the second send request is used to indicate that the cloud instance of the server is about to send the target data; the offload card is specifically used to convert the second send request into a first send request, the first send request conforms to the format requirements of the first protocol; the offload card is specifically used to send the first send request to the second virtualization device through the first driver, so that the heterogeneous card sends the first read request to the offload card through the second virtualization device based on the first send request. In the above implementation, after the offload card receives the second send request from the second driver of the cloud instance of the server through the first virtualization device, the second send request can be converted into a first send request, the second send request conforms to the format requirements of the second protocol, and the first send request conforms to the format requirements of the first protocol. Then, the offload card sends the first send request to the second virtualization device of the heterogeneous card through the first driver, so that the heterogeneous card can determine that the cloud instance of the server is about to send the target data based on the first send request, thereby initiating the first read request.
[0014] In a possible implementation, the offload card is specifically used to receive the target data from the second driver through the first virtualization device; the offload card is specifically used to send the target data to the second virtualization device through the first driver. In the above implementation, after reading the target data, the cloud instance of the server can send the target data to the first virtualization device of the offload card through its own second driver. Then, the offload card sends the target data to the second virtualization device of the heterogeneous card through the first driver, so that the heterogeneous card successfully reads the target data.
[0015] In a possible implementation, if the first operation request is a first read request and the second operation request is a second read request, the offload card is also used to send the second read request to the second driver through the first virtualization device, so that the cloud instance of the server reads the target data from the memory of the cloud instance of the server based on the second read request and transmits it to the first virtualization device; the offload card is also used to store the target data in the memory of the offload card; the offload card is also used to receive the first read request from the second virtualization device through the first driver; the offload card is also used to read the target data from the memory of the offload card based on the first read request; the offload card is also used to send the target data to the second virtualization device through the first driver. In the above implementation, the offload card can send the second read request to the second driver of the cloud instance of the server through the first virtualization device to obtain the target data from the cloud instance of the server and store it in the memory of the offload card. When the offload card receives the first read request from the second virtualization device of the heterogeneous card through the first driver, the target data can be read from the memory and sent to the second virtualization device of the heterogeneous card, so that the heterogeneous card successfully reads the target data.
[0016] In a possible implementation, if the first operation request is a first write request, the offload card is specifically used to receive a first receive request from a second virtualization device through a first driver, the first receive request conforms to the format requirements of a first protocol, and the first receive request is used to indicate that the heterogeneous card requests to obtain the available memory space information of the cloud instance of the server; the offload card is specifically used to convert the first receive request into a second receive request, the second receive request conforms to the format requirements of a second protocol; the offload card is specifically used to send the second receive request to the second driver through the first virtualization device; the offload card is specifically used to receive the available memory space information from the second driver through the first virtualization device; the offload card is specifically used to send the available memory space information to the second virtualization device through the first driver, so that the heterogeneous card sends a first write request to the offload card through the second virtualization device based on the available memory space information. In the aforementioned implementation, after the offload card receives the first receive request from the second virtualization device of the heterogeneous card through the first driver, the first receive request can be converted into a second receive request, the first receive request conforms to the format requirements of the first protocol, and the second receive request conforms to the format requirements of the second protocol. Then, the offload card sends the first receiving request to the second driver of the cloud instance of the server through the first virtualization device, so that the cloud instance of the server sends the available memory space information to the offload card through the second driver. Next, the offload card sends the available memory space information to the second virtualization device of the heterogeneous card through the first driver, so that the heterogeneous card determines which memory location of the cloud instance to write the target data to based on the available memory space information, thereby initiating a first write request.
[0017] A second aspect of an embodiment of the present application provides a method for data processing, characterized in that the method is applied to a server system, the server system comprising: a server, an offload card and a heterogeneous card, the offload card is connected to the server, and the heterogeneous card is connected to the offload card; the offload card records a mapping relationship between a first virtualized device of the offload card and a second virtualized device of the heterogeneous card; the offload card implements communication between a cloud instance of the server and the second virtualized device of the heterogeneous card according to the mapping relationship, wherein the first virtualized device of the offload card is directly connected to the cloud instance of the server, and the second virtual device of the heterogeneous card is directly connected to the offload card.
[0018] In the above method, the offload card is connected between the server and the heterogeneous card. Since the first virtualized device of the offload card is directly connected to the cloud instance of the server, the offload card can realize communication with the cloud instance of the server based on the first virtualized device. Furthermore, since the second virtual device of the heterogeneous card is directly connected to the offload card, the heterogeneous card can realize communication with the offload card based on the second virtual device. Therefore, the offload card can serve as a bridge between the heterogeneous card and the cloud instance of the server to realize communication between the heterogeneous card and the cloud instance of the server. Since the offload card is directly connected to the extension of the heterogeneous card, the offload card can be responsible for processing the management and control surface of the second virtualized device of the heterogeneous card, which can save server resources and help promote the lightweight of the server.
[0019] In one possible implementation, the heterogeneous card supports a first protocol, the offload card is loaded with a first driver corresponding to the first protocol, the server supports a second protocol, and the cloud instance of the server is loaded with a second driver corresponding to the second protocol, and the offload card implements communication between the cloud instance of the server and the second virtualization device of the heterogeneous card according to a mapping relationship, including: the offload card receives a first operation request from the second virtualization device through the first driver, and the first operation request complies with the format requirements of the first protocol; the offload card converts the first operation request into a second operation request, and the second operation request complies with the format requirements of the second protocol; the offload card sends the second operation request to the second driver through the first virtualization device, so that the cloud instance of the server reads target data from the memory of the cloud instance of the server or writes target data to the memory of the cloud instance of the server based on the second operation request.
[0020] In a possible implementation, the offload card includes an engine module that supports a second protocol, and the offload card records a mapping relationship between a first virtualized device of the offload card and a second virtualized device of the heterogeneous card, including: after determining that the heterogeneous card is started, the offload card loads a first driver; the offload card obtains device information of the heterogeneous card through the first driver; the offload card configures the engine module as the first virtualized device based on configuration information, and the configuration information is preset information; the offload card receives an enumeration result from the second driver through the first virtualized device, and the enumeration result is used to indicate that the cloud instance of the server has detected the first virtualized device; the offload card constructs a mapping relationship between the second virtualized device of the heterogeneous card and the first virtualized device of the offload card based on the device information and the enumeration result.
[0021] In one possible implementation, if the first operation request is a first read request, before the offload card receives the first operation request from the second virtualization device through the first driver, the method also includes: the offload card receives a second send request from the second driver through the first virtualization device, the second send request complies with the format requirements of the second protocol, and the second send request is used to indicate that the cloud instance of the server is about to send target data; the offload card converts the second send request into a first send request, and the first send request complies with the format requirements of the first protocol; the offload card sends the first send request to the second virtualization device through the first driver, so that the heterogeneous card sends a first read request to the offload card through the second virtualization device based on the first send request.
[0022] In a possible implementation, the method further includes: the offload card receiving the target data from the second driver through the first virtualization device; and the offload card sending the target data to the second virtualization device through the first driver.
[0023] In one possible implementation, if the first operation request is a first read request and the second operation request is a second read request, after the offload card sends the first send request to the second virtualization device through the first driver, the method also includes: the offload card sends the second read request to the second driver through the first virtualization device, so that the cloud instance of the server reads the target data from the memory of the cloud instance of the server based on the second read request and transmits it to the first virtualization device; the offload card stores the target data in the memory of the offload card; the offload card receives the first read request from the second virtualization device through the first driver; the offload card reads the target data from the memory of the offload card based on the first read request; the offload card sends the target data to the second virtualization device through the first driver.
[0024] In one possible implementation, if the first operation request is a first write request, before the offload card receives the first operation request from the second virtualization device through the first driver, the method also includes: the offload card receives a first receive request from the second virtualization device through the first driver, the first receive request complies with the format requirements of the first protocol, and the first receive request is used to indicate that the heterogeneous card requests to obtain available memory space information of the cloud instance of the server; the offload card converts the first receive request into a second receive request, and the second receive request complies with the format requirements of the second protocol; the offload card sends the second receive request to the second driver through the first virtualization device; the offload card receives the available memory space information from the second driver through the first virtualization device; the offload card sends the available memory space information to the second virtualization device through the first driver, so that the heterogeneous card sends the first write request to the offload card through the second virtualization device based on the available memory space information.
[0025] A third aspect of an embodiment of the present application provides a server system, which includes: a server, an offload card and a heterogeneous card, the heterogeneous card supports a first protocol, the server and the offload card are both loaded with a first driver corresponding to the first protocol, one end of the offload card is connected to the server, and the other end of the offload card is connected to the heterogeneous card; the offload card is used to receive an operation request from the heterogeneous card through the first driver of the offload card, and the operation request complies with the format requirements of the first protocol; the offload card is also used to send the operation request to the first driver of the cloud instance of the server through a direct channel, so that the cloud instance of the server reads target data from the memory of the cloud instance of the server or writes target data to the memory of the cloud instance of the server based on the operation request.
[0026] In a possible implementation, the server includes a first interface, the offload card includes a second interface and a third interface, and the heterogeneous card includes a fourth interface; the first interface is connected to the second interface, and the third interface is connected to the fourth interface.
[0027] In one possible implementation, the server system also includes an expansion card, the server includes a first interface, the expansion card includes a second interface, a third interface, a fourth interface and a fifth interface, the uninstall card includes a sixth interface and a seventh interface, and the heterogeneous card includes an eighth interface; the first interface is connected to the second interface, the second interface is connected to the third interface, the fourth interface is connected to the fifth interface, the sixth interface is connected to the third interface, the seventh interface is connected to the fourth interface, and the eighth interface is connected to the fifth interface.
[0028] In one possible implementation, the offload card is further used to receive an initialization request from a first driver of a cloud instance of a server through a direct channel; the offload card is further used to send the initialization request to the heterogeneous card through the first driver of the offload card, so that the heterogeneous card sends device information of the heterogeneous card to the first driver of the offload card based on the initialization request; the offload card is further used to send the device information of the heterogeneous card to the first driver of the cloud instance of the server through the direct channel after receiving the device information of the heterogeneous card through the first driver of the offload card.
[0029] In one possible implementation, if the operation request is a read request, the offload card is also used to receive a send request from a first driver of a cloud instance of the server through a direct channel, the send request conforms to the format requirements of the first protocol, and the send request is used to indicate that the cloud instance of the server is about to send target data; the offload card is also used to send the send request to the heterogeneous card through the first driver of the offload card, so that the heterogeneous card sends a read request to the first driver of the offload card based on the send request.
[0030] In a possible implementation, the offload card is further used to receive target data from a first driver of a cloud instance of the server through a pass-through channel; the offload card is further used to send the target data to the heterogeneous card through the first driver of the offload card.
[0031] In one possible implementation, the offload card is also used to send a read request to the first driver of the server's cloud instance through a direct channel, so that the server's cloud instance reads target data from the memory of the server's cloud instance based on the read request of the offload card, and sends the target data to the direct channel through the first driver of the server's cloud instance; the offload card is also used to store the target data in the memory of the offload card; the offload card is also used to receive a read request from a heterogeneous card through the first driver of the offload card; the offload card is also used to read the target data from the memory of the offload card based on the read request of the heterogeneous card; the offload card is also used to send the target data to the heterogeneous card through the first driver of the offload card.
[0032] In one possible implementation, if the operation request is a write request, the offload card is also used to receive a receive request from the heterogeneous card through the first driver of the offload card, the receive request conforms to the format requirements of the first protocol, and the receive request is used to indicate that the heterogeneous card requests to obtain available memory space information of the cloud instance of the server; the offload card is also used to send the receive request to the first driver of the cloud instance of the server through a direct channel; the offload card is also used to receive the available memory space information from the first driver of the cloud instance of the server through the direct channel; the offload card is also used to send the available memory space information to the heterogeneous card through the first driver of the offload card, so that the heterogeneous card sends a write request to the offload card based on the available memory space information.
[0033] A fourth aspect of an embodiment of the present application provides a method for data processing, which is applied to a server system, wherein the server system includes a server, an offload card and a heterogeneous card, the heterogeneous card supports a first protocol, the server and the offload card are both loaded with a first driver corresponding to the first protocol, one end of the offload card is connected to the server, and the other end of the offload card is connected to the heterogeneous card; the method includes: the offload card receives an operation request from the heterogeneous card through the first driver of the offload card, and the operation request complies with the format requirements of the first protocol; the offload card sends the operation request to the first driver of the cloud instance of the server, so that the cloud instance of the server reads target data from the memory of the cloud instance of the server or writes target data to the memory of the cloud instance of the server based on the operation request.
[0034] In a possible implementation, the method also includes: the offload card receives an initialization request from a first driver of a cloud instance of a server through a direct channel; the offload card sends the initialization request to the heterogeneous card through the first driver of the offload card, so that the heterogeneous card sends device information of the heterogeneous card to the first driver of the offload card based on the initialization request; after receiving the device information of the heterogeneous card through the first driver of the offload card, the offload card sends the device information of the heterogeneous card to the first driver of the cloud instance of the server through the direct channel.
[0035] In one possible implementation, if the operation request is a read request, before the offload card receives the operation request from the heterogeneous card through the first driver of the offload card, the method also includes: the offload card receives a send request from the first driver of the cloud instance of the server through a direct channel, the send request complies with the format requirements of the first protocol, and the send request is used to indicate that the cloud instance of the server is about to send target data; the offload card sends the send request to the heterogeneous card through the first driver of the offload card, so that the heterogeneous card sends a read request to the first driver of the offload card based on the send request.
[0036] In a possible implementation, the method further includes: the offload card receiving target data of a first driver from a cloud instance of the server through a through channel; and the offload card sending the target data to the heterogeneous card through the first driver of the offload card.
[0037] In one possible implementation, after the offload card sends a sending request to the heterogeneous card through the first driver of the offload card, the method also includes: the offload card sends a read request to the first driver of the cloud instance of the server through a direct channel, so that the cloud instance of the server reads the target data from the memory of the cloud instance of the server based on the read request of the offload card, and sends the target data to the direct channel through the first driver of the cloud instance of the server; the offload card stores the target data in the memory of the offload card; the offload card receives the read request from the heterogeneous card through the first driver of the offload card; the offload card reads the target data from the memory of the offload card based on the read request of the heterogeneous card; the offload card sends the target data to the heterogeneous card through the first driver of the offload card.
[0038] In one possible implementation, if the operation request is a write request, before the offload card receives the operation request from the heterogeneous card through the first driver of the offload card, the method also includes: the offload card receives a receive request from the heterogeneous card through the first driver of the offload card, the receive request complies with the format requirements of the first protocol, and the receive request is used to indicate that the heterogeneous card requests to obtain available memory space information of the cloud instance of the server; the offload card sends the receive request to the first driver of the cloud instance of the server through a direct channel; the offload card receives the available memory space information from the first driver of the cloud instance of the server through the direct channel; the offload card sends the available memory space information to the heterogeneous card through the first driver of the offload card, so that the heterogeneous card sends a write request to the offload card based on the available memory space information.
[0039] The fifth aspect of an embodiment of the present application provides an offload card, one end of the offload card is connected to a server, and the other end of the offload card is connected to a heterogeneous card, the offload card comprising: a recording module for recording a mapping relationship between a first virtualization device of the offload card and a second virtualization device of the heterogeneous card, and a communication module for realizing communication between a cloud instance of the server and the second virtualization device of the heterogeneous card according to the mapping relationship, wherein the first virtualization device of the offload card is directly connected to the cloud instance of the server, and the second virtual device of the heterogeneous card is directly connected to the offload card.
[0040] In a possible implementation, the server includes a first interface, the offload card includes a second interface and a third interface, and the heterogeneous card includes a fourth interface; the first interface is connected to the second interface, and the third interface is connected to the fourth interface.
[0041] In one possible implementation, the server system also includes an expansion card, the server includes a first interface, the expansion card includes a second interface, a third interface, a fourth interface and a fifth interface, the uninstall card includes a sixth interface and a seventh interface, and the heterogeneous card includes an eighth interface; the first interface is connected to the second interface, the second interface is connected to the third interface, the fourth interface is connected to the fifth interface, the sixth interface is connected to the third interface, the seventh interface is connected to the fourth interface, and the eighth interface is connected to the fifth interface.
[0042] In one possible implementation, the heterogeneous card supports a first protocol, the offload card is loaded with a first driver corresponding to the first protocol, the server supports a second protocol, and the cloud instance of the server is loaded with a second driver corresponding to the second protocol; the communication module is specifically used to: receive a first operation request from a second virtualization device through the first driver, the first operation request conforming to the format requirements of the first protocol; convert the first operation request into a second operation request, the second operation request conforming to the format requirements of the second protocol; send the second operation request to the second driver through the first virtualization device, so that the cloud instance of the server reads target data from the memory of the cloud instance of the server or writes target data to the memory of the cloud instance of the server based on the second operation request.
[0043] In a possible implementation, the offload card includes an engine module that supports the second protocol; the recording module is specifically used to: load the first driver after determining that the heterogeneous card is started; obtain device information of the heterogeneous card through the first driver; configure the engine module as a first virtualized device based on configuration information, and the configuration information is preset information; receive an enumeration result from the second driver through the first virtualized device, and the enumeration result is used to indicate that the cloud instance of the server has detected the first virtualized device; and construct a mapping relationship between the second virtualized device of the heterogeneous card and the first virtualized device of the offload card based on the device information and the enumeration result.
[0044] In one possible implementation, if the first operation request is a first read request, the communication module is specifically used to: receive a second send request from a second driver through a first virtualization device, the second send request complies with the format requirements of a second protocol, and the second send request is used to indicate that the cloud instance of the server is about to send target data; convert the second send request into a first send request, the first send request complies with the format requirements of the first protocol; send the first send request to the second virtualization device through the first driver, so that the heterogeneous card sends a first read request to the offload card through the second virtualization device based on the first send request.
[0045] In a possible implementation manner, the communication module is specifically configured to: receive target data from the second driver through the first virtualization device; and send the target data to the second virtualization device through the first driver.
[0046] In one possible implementation, if the first operation request is a first read request and the second operation request is a second read request, the communication module is also used to: send the second read request to the second driver through the first virtualization device, so that the cloud instance of the server reads the target data from the memory of the cloud instance of the server based on the second read request and transmits it to the first virtualization device; stores the target data in the memory of the offload card; receives the first read request from the second virtualization device through the first driver; reads the target data from the memory of the offload card based on the first read request; and sends the target data to the second virtualization device through the first driver.
[0047] In one possible implementation, if the first operation request is a first write request, the communication module is specifically used to: receive a first receive request from a second virtualization device through a first driver, the first receive request conforming to the format requirements of a first protocol, and the first receive request is used to indicate that the heterogeneous card requests to obtain available memory space information of a cloud instance of a server; convert the first receive request into a second receive request, the second receive request conforming to the format requirements of a second protocol; send the second receive request to the second driver through the first virtualization device; receive available memory space information from the second driver through the first virtualization device; send the available memory space information to the second virtualization device through the first driver, so that the heterogeneous card sends a first write request to the offload card through the second virtualization device based on the available memory space information.
[0048] The sixth aspect of an embodiment of the present application provides an offload card, one end of the offload card is connected to a server, and the other end of the offload card is connected to a heterogeneous card, the heterogeneous card supports a first protocol, and the server and the offload card are both loaded with a first driver corresponding to the first protocol; the offload card includes: a communication module, used to receive an operation request from the heterogeneous card through the first driver of the offload card, the operation request conforming to the format requirements of the first protocol; the communication module is also used to send the operation request to the first driver of the cloud instance of the server through a direct channel, so that the cloud instance of the server reads target data from the memory of the cloud instance of the server or writes target data to the memory of the cloud instance of the server based on the operation request.
[0049] In a possible implementation, the server includes a first interface, the offload card includes a second interface and a third interface, and the heterogeneous card includes a fourth interface; the first interface is connected to the second interface, and the third interface is connected to the fourth interface.
[0050] In one possible implementation, the server system also includes an expansion card, the server includes a first interface, the expansion card includes a second interface, a third interface, a fourth interface and a fifth interface, the uninstall card includes a sixth interface and a seventh interface, and the heterogeneous card includes an eighth interface; the first interface is connected to the second interface, the second interface is connected to the third interface, the fourth interface is connected to the fifth interface, the sixth interface is connected to the third interface, the seventh interface is connected to the fourth interface, and the eighth interface is connected to the fifth interface.
[0051] In a possible implementation, the communication module is also used to: receive an initialization request from a first driver of a cloud instance of a server through a direct channel; send the initialization request to the heterogeneous card through the first driver of the unloaded card, so that the heterogeneous card sends device information of the heterogeneous card to the first driver of the unloaded card based on the initialization request; after receiving the device information of the heterogeneous card through the first driver of the unloaded card, send the device information of the heterogeneous card to the first driver of the cloud instance of the server through the direct channel.
[0052] In one possible implementation, if the operation request is a read request, the communication module is also used to: receive a send request from a first driver of a cloud instance of a server through a direct channel, the send request conforms to the format requirements of a first protocol, and the send request is used to indicate that the cloud instance of the server is about to send target data; send the send request to the heterogeneous card through the first driver of the unload card, so that the heterogeneous card sends a read request to the first driver of the unload card based on the send request.
[0053] In a possible implementation, the communication module is further used to: receive target data of a first driver from a cloud instance of a server through a through channel; and send the target data to a heterogeneous card through the first driver of the offload card.
[0054] In one possible implementation, the communication module is also used to: send a read request to the first driver of the cloud instance of the server through a direct channel, so that the cloud instance of the server reads the target data from the memory of the cloud instance of the server based on the read request of the offload card, and sends the target data to the direct channel through the first driver of the cloud instance of the server; stores the target data in the memory of the offload card; receives the read request from the heterogeneous card through the first driver of the offload card; reads the target data from the memory of the offload card based on the read request of the heterogeneous card; and sends the target data to the heterogeneous card through the first driver of the offload card.
[0055] In one possible implementation, if the operation request is a write request, the communication module is also used to: receive a receive request from a heterogeneous card through the first driver of the unload card, the receive request complies with the format requirements of the first protocol, and the receive request is used to indicate that the heterogeneous card requests to obtain available memory space information of the cloud instance of the server; send the receive request to the first driver of the cloud instance of the server through a direct channel; receive available memory space information from the first driver of the cloud instance of the server through the direct channel; send the available memory space information to the heterogeneous card through the first driver of the unload card, so that the heterogeneous card sends a write request to the unload card based on the available memory space information.
[0056] The seventh aspect of an embodiment of the present application provides an unloading card, which includes: a processor and a memory; the memory is used to store computer execution instructions; the processor is used to execute the computer execution instructions stored in the memory, so that the unloading card implements the method described in the second aspect, any possible implementation method of the second aspect, the fourth aspect, or any possible implementation method of the fourth aspect.
[0057] An eighth aspect of an embodiment of the present application provides a computer storage medium, comprising computer-readable instructions, which, when executed, implement the method described in the second aspect, any possible implementation of the second aspect, the fourth aspect, or any possible implementation of the fourth aspect.
[0058] A ninth aspect of an embodiment of the present application provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute a method as described in the second aspect, any possible implementation of the second aspect, the fourth aspect, or any possible implementation of the fourth aspect.
[0059] In the server system provided in the embodiment of the present application, the offload card is connected between the server and the heterogeneous card. Since the first virtualization device of the offload card is directly connected to the cloud instance of the server, the offload card can realize communication with the cloud instance of the server based on the first virtualization device. Furthermore, since the second virtual device of the heterogeneous card is directly connected to the offload card, the heterogeneous card can realize communication with the offload card based on the second virtual device. Therefore, the offload card can serve as a bridge between the heterogeneous card and the cloud instance of the server to realize communication between the heterogeneous card and the cloud instance of the server. Since the offload card is directly connected to the extension of the heterogeneous card, the offload card can be responsible for processing the management and control surface of the second virtualization device of the heterogeneous card, which can save the resources of the server and is conducive to promoting the lightweight of the server. BRIEF DESCRIPTION OF THE DRAWINGS
[0060] Figure 1 A schematic diagram of the structure of a server system provided in an embodiment of the present application;
[0061] Figure 2 Another structural diagram of the server system provided in the embodiment of the present application;
[0062] Figure 3 Another structural diagram of the server system provided in the embodiment of the present application;
[0063] Figure 4 A structural schematic diagram of an uninstall card provided in an embodiment of the present application;
[0064] Figure 5 Another structural schematic diagram of the uninstall card provided in the embodiment of the present application;
[0065] Figure 6 A flowchart of a data processing method provided in an embodiment of the present application;
[0066] Figure 7 Another schematic diagram of a data processing method provided in an embodiment of the present application;
[0067] Figure 8 Another schematic diagram of a data processing method provided in an embodiment of the present application;
[0068] Fig. 9 Another schematic diagram of a data processing method provided in an embodiment of the present application;
[0069] Fig.10 Another schematic diagram of a data processing method provided in an embodiment of the present application;
[0070] Fig.11 Another schematic diagram of a data processing method provided in an embodiment of the present application;
[0071] Fig.12 Another schematic diagram of a data processing method provided in an embodiment of the present application;
[0072] Fig.13 Another schematic diagram of a data processing method provided in an embodiment of the present application;
[0073] Fig.14 Another structural schematic diagram of an uninstall card provided in an embodiment of the present application;
[0074] Fig.15 Another structural schematic diagram of an uninstall card provided in an embodiment of the present application;
[0075] Fig.16 Another structural schematic diagram of the unloading card provided in an embodiment of the present application. DETAILED DESCRIPTION
[0076] The embodiment of the present application provides a server system and a method for data processing. By connecting an offload card to an expansion of a heterogeneous card, the offload card can be responsible for processing the control plane of the second virtualization device of the heterogeneous card, which can save server resources and help promote the lightweight of the server.
[0077] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and need not be used to describe a specific order or sequential order. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, which is only to describe the distinction mode adopted by the objects of the same attributes when describing in the embodiments of the present application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment comprising a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment.
[0078] The embodiments of the present application are applied in cloud scenarios. In such scenarios, at least one cloud instance is usually deployed on the server (host), such as a VM, a bare metal server (BMS), an application container engine (docker), etc. In order to meet various requirements of cloud scenarios, heterogeneous cards such as GPUs, NICs, and other devices are generally connected to the server.
[0079] Servers usually support standard protocols, while heterogeneous cards usually support non-standard protocols. In order to communicate with heterogeneous cards, the cloud instance of the server needs to load the dedicated driver of the heterogeneous card (that is, the driver corresponding to the non-standard protocol supported by the heterogeneous card) to achieve communication between the cloud instance of the server and the heterogeneous card. Therefore, if the cloud instance of the server only loads the universal driver (that is, the driver corresponding to the standard protocol supported by the server), the communication between the cloud instance of the server and the heterogeneous card cannot be achieved, and there are certain ecological problems.
[0080] Furthermore, the expansion of the server directly connecting to heterogeneous cards will increase the complexity of the management plane. Therefore, the front-end server will occupy a lot of resources, which are used to process the management and control plane of the virtualized devices of the heterogeneous cards, which does not conform to the lightweight trend of the front-end server.
[0081] Furthermore, kernel lightweight technology is being explored and evolved, and is also a future development trend. The continuous optimization and slimming of the kernel will also bring compatibility issues with certain forms of heterogeneous card drivers, and may not be able to adapt to certain hardware devices.
[0082] Furthermore, the current technology has security issues. Since the heterogeneous cards are directly inserted into the server, all VF devices are exposed to the front-end virtual machines. Each VM will affect each other, and there is no particularly good way to recover and isolate them.
[0083] Furthermore, the design is not flexible enough and overly dependent on heterogeneous card manufacturers. When encountering changes in usage scenarios, it is impossible to meet the diverse needs of cloud customers by simply modifying the design.
[0084] In order to solve the above problems, an embodiment of the present application provides a server system. Figure 1 A schematic diagram of the structure of the server system provided in the embodiment of the present application. Figure 1 As shown, the server system includes: a server, an offload card and a heterogeneous card, wherein one end of the offload card is connected to the server, and the other end of the offload card is connected to the heterogeneous card. Specifically, the physical connection relationship between the server, the offload card and the heterogeneous card can be realized in a variety of ways, which will be introduced below.
[0085] In one possible implementation, Figure 2 As shown ( Figure 2Another structural schematic diagram of the server system provided for an embodiment of the present application), the server system includes: a server, an uninstall card, a heterogeneous card and an expansion card (for example, a riser card, i.e., a functional expansion card or adapter card of a PCIe interface). The server has a first interface, the expansion card has a second interface, a third interface, a fourth interface and a fifth interface, the uninstall card has a sixth interface and a seventh interface, and the heterogeneous card has an eighth interface. Inside the expansion card, the second interface is connected to the third interface, and the fourth interface is connected to the fifth interface. Outside the expansion card, the first interface of the server is connected to the second interface of the expansion card. The sixth interface of the uninstall card is connected to the third interface of the expansion card, the seventh interface of the uninstall card is connected to the fourth interface of the expansion card, and the eighth interface of the heterogeneous card is connected to the fifth interface of the expansion card, thereby realizing the above-mentioned physical connection relationship, that is, the heterogeneous card is connected to the server through the uninstall card.
[0086] In another possible implementation, Figure 3 As shown ( Figure 3 Another structural diagram of the server system provided in the embodiment of the present application) The server system includes: a server, an offload card and a heterogeneous card. The server has a first interface, the offload card has a second interface and a third interface, and the heterogeneous card has a fourth interface. The first interface of the server is connected to the second interface of the offload card, and the third interface of the offload card is connected to the fourth interface of the heterogeneous card, thereby realizing the above-mentioned physical connection relationship, that is, the heterogeneous card is connected to the server through the offload card.
[0087] It should be understood that the aforementioned interfaces may be PCIe interfaces or other types of interfaces, which are not limited here.
[0088] In the above server system, the offload card can serve as a bridge between the server and the heterogeneous card. When the server and the heterogeneous card exchange information, the offload card can centrally manage the information transmitted between the two. The following will introduce the server, offload card and heterogeneous card respectively:
[0089] At least one cloud instance can be deployed on the server, and the cloud instance can be VM-BMS, docker, etc., and the server supports standard virtualization I / O protocols (hereinafter referred to as standard protocols), such as virtio protocol, non-volatile memory express (NVMe) protocol, etc. Therefore, the server and the cloud instance on the server are usually loaded with drivers corresponding to the standard protocol, that is, standard drivers (also called universal drivers), and then interact with devices that support standard protocols through standard drivers.
[0090] Heterogeneous cards can be GPUs, NICs and other devices developed by different manufacturers. Since the protocols customized by each manufacturer are different from the standard protocols, heterogeneous cards usually support non-standard protocols (i.e., non-standard virtualized I / O protocols). Since the offload card needs to communicate with the heterogeneous card, the offload card needs to load the driver corresponding to the non-standard protocol supported by the heterogeneous card, that is, the non-standard driver (also called a dedicated driver). Furthermore, the heterogeneous card can support the SRIOV function, that is, the heterogeneous card can create a virtualized device, including at least one physical function (physical function, PF) and multiple virtual functions (virtual function, VF). The virtualized device of the heterogeneous card is directly connected to the offload card, that is, the heterogeneous card can communicate with the offload card through its virtualized device.
[0091] Figure 4 A schematic diagram of the structure of the uninstall card provided in the embodiment of the present application. Figure 4 As shown, the unloading card is a device that supports the standard protocol, including a hardware system and a processor. The hardware system and the processor are electrically connected, so the processor can realize information exchange with the hardware system. It is worth noting that the unloading card can be implemented in various forms. For example, the unloading card can be an ASIC chip, that is, the ASIC chip includes two parts: a hardware system and a processor. For another example, the unloading card can be a combination of an FPGA chip + an ARM chip, wherein the FPGA chip serves as a hardware system, the ARM chip serves as a processor, and so on, which is not limited here.
[0092] Specifically, Figure 5 As shown ( Figure 5 Another structural diagram of the offload card provided in the embodiment of the present application), after the offload card is started, when the processor of the offload card runs the corresponding code, the processor can be presented as a software system inside the offload card (that is, the steps implemented by the software system can be regarded as being executed by the processor from the perspective of physical hardware). At this time, the offload card can be used as a device that supports software and hardware collaboration. The two systems cooperate with each other, so that the offload card works between the server and the heterogeneous card, serving as a bridge for information transmission between the two, and realizing functions such as data format conversion and transmission of information. The following will be combined with Figure 5 A brief introduction to the hardware system and software system:
[0093] The hardware system includes modules such as a protocol processing module, a protocol conversion module, and a storage module (i.e., the memory of the unloading card). Among them, the protocol processing module includes multiple engine modules that support different standard protocols, for example, at least one engine module (virtio engine) that supports the virtio protocol, at least one engine module (NVMe engine) that supports the NVMe protocol, and the like. In addition, the protocol processing module also includes a pass-through channel (PF bypass) and a PCIe switch (PCIe switch), and the engine module and the pass-through channel are mounted on the PCIe switch. It should be noted that the engine module can be configured by the processing module of the software system as a virtualized device that is directly connected to the cloud instance of the server. The virtualized device and the pass-through channel inside the unloading card can be used to receive information from the cloud instance of the server through the PCIe switch or send information to the cloud instance of the server. It is worth noting that the virtualized device inside the hardware system works in a non-pass-through scenario (in this scenario, the unloading card needs to convert the data format of the transmitted information), while the pass-through channel works in a pass-through scenario (in this scenario, the unloading card does not need to convert the data format of the transmitted information). The protocol conversion module can be used to convert the data format of information in non-pass-through scenarios. For example, information from a heterogeneous card (information sent from a heterogeneous card to the cloud instance of a server) is converted into information that complies with the format requirements of a standard protocol, thereby sending it to the server through the engine module. Another example is converting information from a virtualized device (i.e., information sent from the cloud instance of the server to a heterogeneous card) into information that complies with the format requirements of a non-standard protocol, thereby sending it to a heterogeneous card, and so on. In addition, the protocol conversion module does not convert the data format of information in a pass-through scenario. In non-pass-through scenarios and pass-through scenarios, the storage module is used to store the mapping relationship between data and virtualized devices, etc., which will not be elaborated here.
[0094] The software system includes a processing module and an initialization module. In a non-pass-through scenario, the processing module can be used to configure the engine module so that the engine module is presented as a virtualized device to the outside, so that it can be discovered and used by the server. Furthermore, after the processing module completes the configuration of the engine module to obtain the virtualized device, it can bind the virtualized device inside the offload card with the virtualized device of the heterogeneous card (i.e., build a mapping relationship between the virtualized device of the offload card and the virtualized device of the heterogeneous card), so that the virtualized device inside the offload card serves as a dedicated channel for the heterogeneous card. For example, after the processing module configures the two engine modules respectively, it obtains virtualized device A and virtualized device B. Then, the processing module can bind virtualized device A with virtualized devices D and E of heterogeneous card C, and bind virtualized device B with virtualized devices G, H and I of heterogeneous card F, so that virtualized device A serves as a dedicated channel for heterogeneous card C, virtualized device B serves as a dedicated channel for heterogeneous card F, and so on. In both the non-pass-through scenario and the pass-through scenario, the initialization module is used to load the non-standard driver corresponding to the heterogeneous card and initialize the heterogeneous card based on the non-standard driver, so that information can be exchanged between the heterogeneous card and the offload card.
[0095] In the embodiment of the present application, the cloud computing system can work in non-pass-through scenarios and pass-through scenarios, and the two scenarios will be introduced separately below. It should be noted that the cloud computing system can be set up in advance according to actual needs in which mode the cloud computing system works. For the convenience of explanation, the non-standard protocol supported by the heterogeneous card is referred to as the first protocol, the standard protocol supported by the server is referred to as the second protocol, the driver corresponding to the non-standard protocol is referred to as the first driver, the driver corresponding to the standard protocol is referred to as the second driver, the virtualized device of the uninstalled card is referred to as the first virtualized device, and the virtualized device of the heterogeneous card is referred to as the second virtualized device.
[0096] First, the non-pass-through scenario is introduced. In the non-pass-through scenario, the offload card can record the mapping relationship between the first virtualized device of the offload card and the second virtualized device of the heterogeneous card (i.e., the initialization process), and then realize the communication between the cloud instance of the server and the second virtualized device of the heterogeneous card according to the mapping relationship (i.e., the data transmission process), wherein the first virtualized device of the offload card is directly connected to the cloud instance of the server, and the second virtualized device of the heterogeneous card is directly connected to the offload card. The initialization process of the non-pass-through scenario is first introduced below. Figure 6 A flow chart of a data processing method provided in an embodiment of the present application is shown as follows: Figure 6 As shown, the method includes:
[0097] 601. After the server is started, the second driver is loaded.
[0098] In this embodiment, since the server is connected to one side of the offload card, the offload card and the server can be started synchronously or asynchronously. For example, if the offload card does not have its own power supply, the server can only control the offload card to start after the server is turned on. For another example, if the offload card has its own power supply, the server and the offload card can be started synchronously through their own power supplies, etc., which is not limited here.
[0099] It is worth noting that after the server is started, the server and the cloud instance of the server can load the second driver corresponding to the second protocol supported by the server. Furthermore, after the offload card is started, the processor inside the offload card starts running, thereby starting the software system to support the function of software and hardware collaboration, that is, the hardware system and the software system can work together.
[0100] 602. After determining that the heterogeneous card is started, the unload card loads the first driver.
[0101] Similarly, since the heterogeneous card is connected to the other side of the uninstall card, the uninstall card and the heterogeneous card can be started synchronously or asynchronously. For details, please refer to the relevant description of the startup process of the server and the uninstall card in step 301, which will not be repeated here. After the uninstall card determines that the heterogeneous card is started, that is, after the uninstall card determines that the heterogeneous card is powered on, the uninstall card loads the first driver corresponding to the first protocol supported by the heterogeneous card (the first driver corresponding to the heterogeneous card) through the initialization module.
[0102] 603. The uninstall card obtains device information of the heterogeneous card through the first driver.
[0103] After loading the first driver, the offload card can send an initialization request to the heterogeneous card through the first driver, so that the heterogeneous card performs single-root I / O virtualization (SRIOV) initialization based on the initialization request. During the SRIOV initialization process, the heterogeneous card can create a second virtualization device (including at least one PF and multiple VFs) that is directly connected to the offload card. Thereafter, the heterogeneous card can communicate with the offload card through the second virtualization device. After the heterogeneous card completes the SRIOV initialization, the device information of the heterogeneous card can be returned to the offload card through the second virtualization device. The device information of the heterogeneous card includes information such as the device identification (identity, ID) and device type of each PF and VF.
[0104] 604. The offload card configures the engine module as a first virtualization device based on the configuration information, where the configuration information is preset information.
[0105] The hardware system of the offload card includes an engine module that supports the second protocol, for example, the aforementioned virtioengine, NVMe engine, etc. After loading the first driver, the offload card obtains configuration information for configuring the first virtualized device through the processing module. The configuration information is preset information, including information related to the device type of the first virtualized device (command channel), information related to the number of the first virtualized devices (number of queues), etc. After obtaining the configuration information, the processing module can configure the engine module as the first virtualized device based on the configuration information.
[0106] For example, assume that the server supports the virtio protocol. After loading the first driver, the unload card obtains the configuration information corresponding to the virtio protocol through the processing module. The content contained in the configuration information (i.e., which first virtualization device and how many first virtualization devices are to be configured under the virtio protocol) has been set in advance, such as the configuration information containing the command channel corresponding to virtio-gpu and the number of queues corresponding to virtio-gpu. After obtaining the configuration information, the processing module can determine the virtio engine in the virtio engine, NVMe engine and other modules, and configure the virtio engine as N first virtualization devices such as virtio-gpu1, virtio-gpu2, ..., virtio-gpuN based on the configuration information, where the value of N is related to the number of queues.
[0107] It should be noted that the first virtualization device of the offload card is directly connected to the cloud instance of the server, that is, the offload card can communicate with the cloud instance of the server through the first virtualization device.
[0108] It is worth noting that in the process of the offload card sending information to the cloud instance of the server, the information sent by the offload card through the first virtualization device is forwarded to the second driver of the cloud instance of the server via the second driver of the server (host machine), thereby reaching the cloud instance of the server. This process can be described as the first virtualization device of the offload card sending information to the second driver of the cloud instance of the server, which will not be described in detail later.
[0109] Similarly, when the cloud instance of the server sends information to the offload card, the information sent by the second driver of the cloud instance of the server is forwarded to the first virtualized device of the offload card via the second driver of the server, thereby reaching the offload card. This process can be described as the second driver of the cloud instance of the server sending information to the first virtualized device of the offload card, which will not be described in detail later.
[0110] It should be understood that the above example is only used as an illustrative example in which the second protocol supported by the cloud instance of the server is the virtio protocol, and does not constitute a limitation on the type of the second protocol supported by the cloud instance of the server in this application.
[0111] It should also be understood that in the aforementioned example, under the virtio protocol, only the first virtualized device is virtio-gpu as an example for schematic illustration, and the first virtualized device can also be other types of devices such as virtio-blk and virtio-net.
[0112] It should also be understood that step 603 and step 604 may be executed synchronously or asynchronously. When the two are executed asynchronously, step 603 may be executed before step 604, or after step 604, which is not limited here.
[0113] 605. The offload card receives an enumeration result from the second driver through the first virtualization device, where the enumeration result is used to indicate that the cloud instance of the server has detected the first virtualization device.
[0114] After the processing module of the offload card completes the configuration operation to obtain the first virtualized device, the first virtualized device can be discovered by the cloud instance of the server. When the cloud instance of the server discovers (detects) the first virtualized device, an enumeration result can be sent to the first virtualized device of the offload card through the second driver of the cloud instance, and the enumeration result is used to indicate that the cloud instance of the server has detected the first virtualized device.
[0115] Still as in the above example, after the processing module completes the configuration of the virtio engine, the virtio engine can present the server's cloud instance as virtio-gpu1, virtio-gpu2, ..., virtio-gpuN and other devices. After the server's cloud instance (through the second driver of the cloud instance) detects virtio-gpu1, virtio-gpu2, ..., virtio-gpuN, it can send the enumeration result to the offload card, thereby notifying the offload card that the server's cloud instance has detected virtio-gpu1, virtio-gpu2, ..., virtio-gpuN.
[0116] 606. The offload card constructs a mapping relationship between the second virtualized device of the heterogeneous card and the first virtualized device of the offload card based on the device information of the heterogeneous card and the enumeration result.
[0117] In step 603, after the unloading card receives the device information of the heterogeneous card through the first driver, the device information may be sent to the processing module through the first driver.
[0118] After receiving the enumeration result and the device information of the heterogeneous card, the processing module of the offload card can construct a mapping relationship between the second virtualized device of the heterogeneous card and the first virtualized device of the offload card based on the device information of the heterogeneous card and the enumeration result, and store the mapping relationship in the storage module.
[0119] For example, assume that the device information of the heterogeneous card is used to indicate that the heterogeneous card includes two second virtualized devices (VF1 and VF2 respectively), and the enumeration result is used to indicate that the cloud instance of the server detects three first virtualized devices (virtio-gpu1, virtio-gpu2, and virtio-gpu3 respectively). The processing module can establish the following mapping relationship based on the device information of the heterogeneous card and the enumeration result: virtio-gpu1 corresponds to VF1, virtio-gpu2 corresponds to VF1, and virtio-gpu3 corresponds to VF2.
[0120] Similarly, if another heterogeneous card is connected to the offload card, the initialization process of steps 601 to 606 can also be performed on the heterogeneous card. Then, the offload card is loaded with the first drivers corresponding to multiple heterogeneous cards, and the offload card can record the mapping relationship between multiple first virtualized devices and multiple second virtualized devices of different heterogeneous cards. For example, if heterogeneous card 1 and heterogeneous card 2 are connected to the offload card successively, after completing the initialization respectively, the following mapping relationship can be obtained: virtio-gpu1 corresponds to VF1 of heterogeneous card 1, virtio-gpu2 corresponds to VF1 of heterogeneous card 1, virtio-gpu3 corresponds to VF2 of heterogeneous card 1, virtio-gpu4 corresponds to VF3 of heterogeneous card 2, virtio-gpu5 corresponds to VF4 of heterogeneous card 2, and so on. If the cloud instance of the server sends a certain message to virtio-gpu1, the offload card can forward the message to VF1 based on the correspondence between virtio-gpu1 and VF1, so the message can be successfully sent to heterogeneous card 1.
[0121] Furthermore, the server system of the embodiment of the present application also has a reloading function for the first virtualized device. Take the server supporting the virtio protocol as an example for explanation. Assuming that the virtio protocol supported by the server is the virtio1.0 version, after the aforementioned steps 601 to 606 (wherein, in the configuration process of the virtio engine, the configuration information corresponding to virtio1.0 is used), the virtio engine is presented as the virtio-gpu under virtio1.0. If the virtio protocol supported by the server has undergone a version upgrade, such as updating from virtio1.0 to virtio2.0, then the processing module can obtain the configuration information corresponding to virtio2.0, and reconfigure the configured virtio engine based on the configuration information, so that the virtio engine is presented as the virtio-gpu under virtio2.0. Similarly, if it is necessary to change the device type presented by virtioengine, for example, switch virtio-gpu to virtio-blk, the processing module can obtain new configuration information, and reconfigure the configured virtio engine based on the configuration information, so that the virtio engine is presented as virtio-blk. It should be noted that even if the first virtualized device is reloaded, the mapping relationship between the reloaded first virtualized device and the second virtualized device of the heterogeneous card will not be changed, that is, the mapping relationship between the first virtualized device and the second virtualized device of the heterogeneous card before reloading is retained.
[0122] After the above initialization process, the data transmission process can be carried out. For the server side, the cloud instance of the server can see the first virtualized device presented by the offload card, so the cloud instance of the server can send information to the first virtualized device according to its own needs. After the offload card receives the information through the first virtualized device, it can send the information to the second virtualized device of the heterogeneous card based on the mapping relationship between the first virtualized device of the offload card and the second virtualized device of the heterogeneous card, which is equivalent to the cloud instance of the server actually sending information to the heterogeneous card, thereby realizing the communication between the cloud instance of the server and the heterogeneous card.
[0123] In a non-pass-through scenario, the data transmission process includes the data transmission process in the transmit (TX) direction and the data transmission process in the receive (RX) direction. The data transmission process in the TX direction includes two modes, and the first mode is introduced below. Figure 7 Another flowchart of the data processing method provided in the embodiment of the present application is as follows: Figure 7 As shown, the method includes:
[0124] 701. The offload card receives a second sending request from a second driver through the first virtualization device, where the second sending request complies with the format requirement of the second protocol and is used to indicate that the cloud instance of the server is about to send target data.
[0125] After initialization, the cloud instance of the server can see the first virtualized device presented by the offload card. When the cloud instance of the server needs to send target data to the first virtualized device, a second sending request can be generated. The second sending request meets the format requirements of the second protocol, and the second sending request is used to indicate that the cloud instance of the server is about to send the target data. After receiving the second sending request, the cloud instance of the server sends the second sending request to the first virtualized device through the second driver.
[0126] For example, when the cloud instance of the server wants to send target data to virtio-gpu1, it can generate a second sending request, and the second sending request meets the format requirements of the second protocol (ie, the virtio protocol). Then, the cloud instance of the server sends the second sending request to virtio-gpu1 through the second driver.
[0127] 702. The offload card converts the second sending request into a first sending request, where the first sending request complies with a format requirement of the first protocol.
[0128] After the unload card receives the second sending request through the first virtualization device, the second sending request is sent to the protocol conversion module through the first virtualization device, so that the protocol conversion module obtains the mapping relationship between the first virtualization device and the second virtualization device of the heterogeneous card from the storage module, and converts the second sending request into the first sending request based on the mapping relationship, and the first sending request meets the format requirements of the first protocol.
[0129] For example, if the protocol conversion module receives a second sending request from virtio-gpu1, the mapping relationship between virtio-gpu1 and VF1 of heterogeneous card 1 can be obtained from the storage module. After determining that the second sending request is a request sent to heterogeneous card 1, the data format of the second sending request can be converted to obtain a first sending request, and the first sending request meets the format requirements of the first protocol supported by heterogeneous card 1. For another example, if the protocol conversion module receives a second sending request from virtio-gpu5, the mapping relationship between virtio-gpu5 and VF4 of heterogeneous card 2 can be obtained from the storage module. After determining that the second sending request is a request sent to heterogeneous card 2, the data format of the second sending request can be converted to obtain a first sending request, and the first sending request meets the format requirements of the first protocol supported by heterogeneous card 2, and so on.
[0130] It is worth noting that when the protocol conversion module converts the data format, it needs to obtain the data format mapping table from the storage module, so as to complete the data format conversion based on the mapping table. Specifically, the mapping table records the correspondence between the data format of the first protocol and the data format of the second protocol, so the protocol conversion module can convert the second sending request (which meets the data format requirements of the second protocol) into the first sending request (which meets the data format requirements of the first protocol) based on the mapping table.
[0131] Furthermore, the aforementioned conversion of the data format includes conversion of the address of the target data and the like.
[0132] 703. The offload card sends the first sending request to the second virtualization device through the first driver, so that the heterogeneous card sends the first read request to the offload card through the second virtualization device based on the first sending request.
[0133] After obtaining the first sending request, the offload card sends the first sending request to the first driver through the protocol conversion module, so that the first driver sends the first sending request to the second virtualization device of the heterogeneous card.
[0134] For example, if the protocol conversion module receives the second sending request from virtio-gpu1, the mapping relationship between virtio-gpu1 and VF1 of heterogeneous card 1 can be obtained from the storage module. After determining that the second sending request is a request sent to heterogeneous card 1, the data format of the second sending request can be converted to obtain a first sending request, and the first sending request meets the format requirements of the first protocol supported by heterogeneous card 1. Then, the protocol conversion module sends the first sending request to the first driver corresponding to heterogeneous card 1, so that the first driver sends the first sending request to VF1.
[0135] 704. The offload card receives a first read request from the second virtualization device through the first driver, where the first read request complies with a format requirement of the first protocol.
[0136] After receiving the first sending request, the heterogeneous card can determine that the cloud instance of the server is about to send the target data to the heterogeneous card. Therefore, the heterogeneous card generates a first read request, which meets the format requirements of the first protocol and is used to indicate that the heterogeneous card needs to read the target data. After receiving the first read request, the heterogeneous card sends the first read request to the offload card through the second virtualization device.
[0137] 705. The unloading card converts the first read request into a second read request, where the second read request complies with the format requirement of the second protocol.
[0138] After the first driver of the offload card receives the first read request from the second virtualization device, the first read request can be sent to the protocol conversion module, so that the protocol conversion module obtains the mapping relationship between the first virtualization device of the offload card and the second virtualization device of the heterogeneous card from the storage module, and converts the first read request into a second read request based on the mapping relationship, and the second read request meets the format requirements of the second protocol.
[0139] For example, in the first driver installed by the uninstall card, if the first driver corresponding to the heterogeneous card 1 receives a first read request from the VF1 of the heterogeneous card 1, the first read request can be sent to the protocol conversion module, and the protocol conversion module can be notified that the first read request comes from the VF1 of the heterogeneous card 1 (for example, the VF1 identifier can be carried on the first read request, etc.). The protocol conversion module can obtain the mapping relationship between virtio-gpu1 and the VF1 of the heterogeneous card 1 from the storage module, and after determining that the first read request is a request sent to the cloud instance through virtio-gpu1, the data format of the first read request can be converted to obtain a second read request, and the second read request meets the format requirements of the second protocol (i.e., the virtio protocol) supported by the cloud instance of the server. Then, the protocol conversion module sends the second read request to virtio-gpu1.
[0140] 706. The offload card sends the second read request to the second driver through the first virtualization device.
[0141] After receiving the second read request, the offload card sends the second read request to the first virtualization device through the protocol conversion module, so that the first virtualization device sends the second read request to the second driver of the cloud instance of the server.
[0142] 707 . The cloud instance of the server reads the target data from the memory of the cloud instance of the server based on the second read request.
[0143] After receiving the second read request, the cloud instance of the server may read the target data from the memory of the cloud instance of the server based on the second read request.
[0144] 708. The offload card receives target data from the second driver through the first virtualization device.
[0145] After the cloud instance of the server obtains the target data, it can return the target data to the first virtualized device through the second driver.
[0146] 709. The offload card sends the target data to the second virtualization device through the first driver.
[0147] After the unload card receives the target data through the first virtualization device, it sends the target data to the protocol conversion module through the first virtualization device, so that the protocol conversion module obtains the mapping relationship between the first virtualization device and the second virtualization device of the heterogeneous card from the storage module, but does not convert the data format of the target data. Based on the mapping relationship, the target data is directly sent to the first driver corresponding to the heterogeneous card, and then the target data is sent to the second virtualization device of the heterogeneous card through the first driver, that is, the heterogeneous card successfully reads the target data. At this point, the data transmission process in the TX direction is completed.
[0148] For example, if the protocol conversion module receives target data from virtio-gpu1, it can obtain the mapping relationship between virtio-gpu1 and VF1 of heterogeneous card 1 from the storage module. After determining that the target data is data sent to heterogeneous card 1, the target data is not converted into a data format. Then, the protocol conversion module sends the target data to the first driver corresponding to heterogeneous card 1, so that the first driver sends the target data to VF1.
[0149] The above is an explanation of the first mode of the data transmission process in the TX direction, and the second mode will be introduced below. Figure 8 Another flowchart of the data processing method provided in the embodiment of the present application is as follows: Figure 8 As shown, the method includes:
[0150] 801. The offload card receives a second sending request from a second driver through a first virtualization device, where the second sending request complies with a format requirement of a second protocol and is used to indicate that a cloud instance of a server is about to send target data.
[0151] 802. The offload card converts the second sending request into a first sending request, where the first sending request complies with a format requirement of the first protocol.
[0152] 803. The offload card sends the first sending request to the second virtualization device through the first driver, so that the heterogeneous card sends the first read request to the offload card through the second virtualization device based on the first sending request.
[0153] For the description of steps 701 to 703 , please refer to the relevant description of steps 801 to 803 above, which will not be repeated here.
[0154] 804. The offload card sends the second read request to the second driver through the first virtualization device.
[0155] After the offload card sends the first send request to the second virtualization device of the heterogeneous card, the offload card can generate a second read request, which meets the format requirements of the second protocol and is used to indicate that the offload card needs to read the target data. Then, the offload card sends the second read request to the second driver of the cloud instance of the server through the first virtualization device.
[0156] 805. The cloud instance of the server reads the target data from the memory of the cloud instance of the server based on the second read request.
[0157] After the cloud instance of the server receives the second read request, it reads the target data from the memory of the cloud instance of the server based on the second read request.
[0158] 806. The cloud instance of the server sends the target data to the first virtualized device through the second driver.
[0159] After obtaining the target data, the cloud instance of the server returns the target data to the first virtualized device through the second driver.
[0160] 807. The offload card stores the target data in the memory of the offload card.
[0161] After receiving the target data through the first virtualization device, the unloading card sends the target data to the operation module (included in the hardware system, Figure 5 (not shown in the figure), so that the operating module stores the target data in the storage module of the unloading card.
[0162] 808. The offload card receives a first read request from the second virtualization device through the first driver.
[0163] After receiving the second sending request, the heterogeneous card can determine that the cloud instance of the server is about to send the target data, so it can generate a first read request, which meets the format requirements of the first protocol. Then, the heterogeneous card sends the first read request to the first driver of the offload card through the second virtualization device.
[0164] 809. The offload card reads target data from the memory of the offload card based on the first read request.
[0165] Since the storage module of the unloading card has pre-stored the target data, after receiving the first read request, the first driver of the unloading card can send the first read request to the operation module, so that the operation module obtains the target data in the storage module based on the first read request.
[0166] 810. The offload card sends the target data to the second virtualization device through the first driver.
[0167] After obtaining the target data, the offload card sends the target data to the first driver through the operation module, so that the first driver returns the target data to the second virtualization device of the heterogeneous card, that is, the heterogeneous card successfully reads the target data. At this point, the data transmission process in the TX direction is completed.
[0168] The above is an explanation of the data transmission process in the TX direction. The following is an introduction to the data transmission process in the RX direction. Fig. 9 Another flowchart of the data processing method provided in the embodiment of the present application is as follows: Fig. 9 As shown, the method includes:
[0169] 901. The offload card receives a first receiving request from the second virtualization device through a first driver. The first receiving request complies with the format requirement of the first protocol, and the first receiving request is used to indicate that the heterogeneous card requests to obtain available memory space information of the cloud instance of the server.
[0170] After initialization, the heterogeneous card can actively request the available memory space information from the cloud instance of the server, so as to determine which part of the memory in the cloud instance of the server can be used to write the target data. Specifically, the heterogeneous card can first generate a first receiving request, which meets the format requirements of the first protocol and is used to instruct the heterogeneous card to request to obtain the available memory space information of the cloud instance of the server. Then, the heterogeneous card sends the first receiving request to the offload card through the second virtualization device.
[0171] 902. The offload card converts the first receiving request into a second receiving request, where the second receiving request complies with a format requirement of the second protocol.
[0172] 903. The offload card sends the second receiving request to the second driver through the first virtualization device.
[0173] For the description of step 902 and step 903, please refer to the relevant description of step 705 and step 706 above, which will not be repeated here.
[0174] 904. The offload card receives available memory space information from the second driver through the first virtualization device.
[0175] After receiving the second receiving request, the cloud instance of the server can obtain its own available memory space information based on the second receiving request. Then, the cloud instance of the server returns the available memory space information to the first virtualization device through the second driver.
[0176] 905. The offload card sends the available memory space information to the second virtualization device through the first driver.
[0177] After the offload card receives the available memory space information through the first virtualization device, the available memory space information can be sent to the protocol conversion module through the first virtualization device, so that the protocol conversion module obtains the mapping relationship between the first virtualization device and the second virtualization device of the heterogeneous card from the storage module, but does not convert the data format of the available memory space information. Based on the mapping relationship, the available memory space information is directly sent to the first driver corresponding to the heterogeneous card, and then the available memory space information is sent to the second virtualization device of the heterogeneous card through the first driver, that is, the heterogeneous card successfully obtains the available memory space information of the cloud instance of the server.
[0178] After the heterogeneous card obtains the available memory space information of the cloud instance of the server, it can determine which memory location of the cloud instance to write the target data to based on the available memory space information, thereby generating a first write request, which meets the format requirements of the first protocol and is used to indicate that the heterogeneous card needs to write the target data. Then, the heterogeneous card sends the first write request to the offload card through the second virtualization device.
[0179] 906. The offload card receives a first write request from the second virtualization device through the first driver, where the first write request complies with a format requirement of the first protocol.
[0180] 907. The offload card converts the first write request into a second write request, where the second write request complies with a format requirement of the second protocol.
[0181] 908. The offload card sends the second write request to the second driver through the first virtualization device.
[0182] 909. The cloud instance of the server writes the target data to the memory of the cloud instance of the server based on the second write request.
[0183] For the description of step 906 to step 909, please refer to the relevant description of step 704 and step 707, which will not be repeated here. At this point, the cloud instance of the server successfully writes the target data, and the data transmission process in the RX direction is completed.
[0184] In the server system provided by the embodiment of the present application, the offload card is connected between the server and the heterogeneous card. When the server system works in a non-through scenario, the cloud instance of the server is loaded with a second driver corresponding to the second protocol supported by the server, and the offload card is loaded with a first driver corresponding to the first protocol supported by the heterogeneous card. Specifically, after the offload card receives the first operation request from the second virtualization device of the heterogeneous card through the first driver, the first operation request can be converted into a second operation request, the first operation request meets the format requirements of the first protocol, and the second operation request meets the format requirements of the second protocol. Then, the offload card sends the first operation request to the second driver of the cloud instance of the server through the first virtualization device, so that the cloud instance of the server completes the corresponding operation based on the first operation request. Therefore, in the server system provided by the embodiment of the present application, even if the cloud instance of the server only loads the second driver, the communication between the cloud instance of the server and the heterogeneous card can be realized.
[0185] Furthermore, since the offload card is directly connected to the expansion of the heterogeneous card, the offload card can be responsible for processing the control plane of the second virtualization device of the heterogeneous card, without occupying any additional resources of the front-end server, which can save server resources and help promote the lightweight of the server.
[0186] Furthermore, security issues can also be well resolved. For example, in the event of an abnormal situation in which a virtual machine escapes security, the management plane of the unloading card isolates and shields the abnormal queue channel to prevent impact on other virtual machines, and saves the abnormal error information in local memory, so that a new workflow can be started immediately after the fault is recovered.
[0187] Furthermore, the offload card that supports software and hardware collaboration can exist in the form of FPGA+CPU, making full use of the dynamic reconfiguration capability of FPGA to design the engine module more flexibly. It not only supports all current virtualized I / O protocol processing, but also supports future diversified scenario requirements by modifying the design and reloading the image files of some modules without changing the hardware architecture.
[0188] The above is a detailed description of the non-through scenario. The following will introduce the through scenario. The through scenario also includes the initialization process and the data transmission process. The following will first introduce the initialization process of the through scenario. Fig.10 Another flowchart of the data processing method provided in the embodiment of the present application is as follows: Fig.10 As shown, the method includes:
[0189] In this embodiment, since the server is connected to one side of the offload card, the offload card and the server can be started synchronously or asynchronously. For example, if the offload card does not have its own power supply, the server can only control the offload card to start after the server is turned on. For another example, if the offload card has its own power supply, the server and the offload card can be started synchronously through their own power supplies, etc., which is not limited here.
[0190] 1000. After determining that the heterogeneous card is started, the server and the offload card both load the first driver.
[0191] When both the server and the offload card determine that the heterogeneous card is started, the server and the server's cloud instance can load the first driver corresponding to the first protocol supported by the heterogeneous card. Similarly, the offload card can also load the first driver corresponding to the first protocol supported by the heterogeneous card through the initialization module.
[0192] 1001. The offload card receives an initialization request of a first driver from a cloud instance of a server through a direct channel.
[0193] When the offload card works in the pass-through scenario, the pass-through channel of the offload card (i.e. Figure 5 One end of the PF bypass in the offload card is directly connected to the cloud instance of the server, and the other end is directly connected to the heterogeneous card. That is, the offload card can directly transmit the information from the cloud instance of the server to the heterogeneous card through the direct channel, or directly transmit the information from the heterogeneous card to the cloud instance of the server.
[0194] Specifically, after the cloud instance of the server loads the first driver, an initialization request may be sent to the direct channel of the offload card.
[0195] 1002. The offload card sends an initialization request to the heterogeneous card through a first driver of the offload card.
[0196] After receiving the initialization request through the direct channel, the offload card may send the initialization request to the first driver of the offload card through the direct channel, so that the first driver of the offload card sends the initialization request to the heterogeneous card.
[0197] 1003. The heterogeneous card sends device information of the heterogeneous card to the first driver of the uninstall card based on the initialization request.
[0198] After receiving the initialization request, the heterogeneous card can start initialization. After completing the initialization, the device information of the heterogeneous card can be returned to the first driver of the uninstall card. The device information of the heterogeneous card includes the ID of the heterogeneous card, the device type and other information.
[0199] 1004. After receiving the device information of the heterogeneous card through the first driver of the offload card, the offload card sends the device information of the heterogeneous card to the first driver of the cloud instance of the server through the direct channel.
[0200] After the unload card receives the device information of the heterogeneous card through the first driver of the unload card, the device information of the heterogeneous card can be sent to the direct channel through the first driver of the unload card, so that the direct channel returns the device information of the heterogeneous card to the first driver of the cloud instance of the server, that is, the cloud instance of the server obtains the device information of the heterogeneous card. At this point, the initialization process in the direct scenario is completed.
[0201] In the direct pass scenario, the data transmission process also includes the data transmission process in the TX direction and the data transmission process in the RX direction. The data transmission process in the TX direction includes two modes, and the first mode is introduced below. Fig.11 Another flowchart of the data processing method provided in the embodiment of the present application is as follows: Fig.11 As shown, the method includes:
[0202] 1101. The offload card receives a sending request from a first driver of a cloud instance of a server through a direct channel. The sending request complies with a format requirement of a first protocol, and the sending request is used to indicate that the cloud instance of the server is about to send target data.
[0203] After initialization, since the heterogeneous card is directly connected to the cloud instance of the server through the direct channel inside the offload card, the cloud instance of the server can directly see the heterogeneous card from the server side. When the cloud instance of the server needs to send target data to the heterogeneous card, a send request can be generated. The send request conforms to the format of the first protocol and is used to indicate that the cloud instance of the server is about to send the target data. Then, the cloud instance of the server sends the send request to the direct channel of the offload card through its own first driver.
[0204] 1102. The offload card sends a sending request to the heterogeneous card through the first driver of the offload card.
[0205] After receiving the sending request through the direct channel, the offload card can send the sending request to the first driver of the offload card through the direct channel, so that the first driver of the offload card sends the sending request to the heterogeneous card.
[0206] 1103. The heterogeneous card sends a read request to the first driver of the unload card based on the sending request, where the read request complies with the format requirement of the first protocol.
[0207] After receiving the sending request, the heterogeneous card can determine based on the sending request that the cloud instance of the server is about to send the target data, so the heterogeneous card generates a read request, which meets the format requirements of the first protocol and is used to indicate that the heterogeneous card needs to read the target data. Then, the heterogeneous card sends the read request to the first driver of the unload card.
[0208] 1104. The offload card sends the read request to the first driver of the cloud instance of the server through the direct channel.
[0209] After the offload card receives the read request of the heterogeneous card through the first driver of the offload card, the read request can be sent to the direct channel through the first driver of the offload card, so that the direct channel sends the read request to the first driver of the cloud instance of the server.
[0210] 1105. The cloud instance of the server reads the target data from the memory of the cloud instance of the server based on the read request.
[0211] After the cloud instance of the server receives the read request, it reads the target data from its own memory based on the read request.
[0212] 1106. The offload card receives target data of the first driver from the cloud instance of the server through the direct channel.
[0213] After the cloud instance of the server obtains the target data, it can send the target data to the direct channel of the offload card through its own first driver.
[0214] 1107. The offload card sends the target data to the heterogeneous card through the first driver of the offload card.
[0215] After the unloading card receives the target data through the direct channel, it can send the target data to the first driver of the unloading card through the direct channel, so that the first driver of the unloading card sends the target data to the heterogeneous card, that is, the heterogeneous card successfully reads the target data. At this point, the data transmission process in the TX direction is completed.
[0216] The above is an explanation of the first mode of the data transmission process in the TX direction, and the second mode will be introduced below. Fig.12 Another flowchart of the data processing method provided in the embodiment of the present application is as follows: Fig.12 As shown, the method includes:
[0217] 1201. The offload card receives a sending request from a first driver of a cloud instance of a server through a direct channel. The sending request complies with the format requirements of the first protocol, and the sending request is used to indicate that the cloud instance of the server is about to send target data.
[0218] 1202. The offload card sends a sending request to the heterogeneous card through the first driver of the offload card.
[0219] For the description of step 1201 to step 1202, please refer to the relevant description of step 1101 and step 1102 above, which will not be repeated here.
[0220] 1203. The offload card is further configured to send the read request to the first driver of the cloud instance of the server through the pass-through channel.
[0221] After the offload card sends the send request to the heterogeneous card, it can first generate a read request that meets the format requirements of the first protocol and is used to indicate that the offload card needs to read the target data. Then, the offload card sends the read request to the first driver of the cloud instance of the server through the direct channel.
[0222] 1204. The cloud instance of the server reads target data from the memory of the cloud instance of the server based on the read request from the offload card.
[0223] 1205. The offload card receives target data of the first driver from the cloud instance of the server through the direct channel.
[0224] For the description of steps 1204 to 1205, please refer to the relevant description of the above steps 1105 and 1106, which will not be repeated here.
[0225] 1206. The offload card stores the target data in the memory of the offload card.
[0226] After receiving the target data through the direct channel, the offload card can send the target data to the operation module of the offload card through the direct channel, so that the operation module stores the target data in the memory of the offload card.
[0227] 1207. The offload card receives a read request from the heterogeneous card through the first driver of the offload card.
[0228] After receiving the sending request, the heterogeneous card can determine based on the sending request that the cloud instance of the server is about to send the target data, so the heterogeneous card generates a read request, which meets the format requirements of the first protocol and is used to indicate that the heterogeneous card needs to read the target data. Then, the heterogeneous card sends the read request to the first driver of the unload card.
[0229] 1208. The offload card reads target data from the memory of the offload card based on the read request from the heterogeneous card.
[0230] After the unloading card receives a read request from the heterogeneous card through the first driver of the unloading card, since the target data is stored in the memory of the unloading card, the read request can be sent to the operation module through the first driver, so that the operation module reads the target data from the memory of the unloading card based on the read request.
[0231] 1209. The offload card sends the target data to the heterogeneous card through the first driver of the offload card.
[0232] After the operation module of the unloading card obtains the target data, it sends the target data to the first driver of the unloading card, so that the first driver sends the target data to the heterogeneous card, that is, the heterogeneous card successfully reads the target data. At this point, the data transmission process in the TX direction is completed.
[0233] The above is an explanation of the data transmission process in the TX direction. The following is an introduction to the data transmission process in the RX direction. Fig.13 Another flowchart of the data processing method provided in the embodiment of the present application is as follows: Fig.13 As shown, the method includes:
[0234] 1301. The offload card receives a receiving request from the heterogeneous card through the first driver of the offload card. The receiving request complies with the format requirement of the first protocol, and the receiving request is used to indicate that the heterogeneous card requests to obtain available memory space information of the cloud instance of the server.
[0235] After initialization, the heterogeneous card can actively request the available memory space information from the cloud instance of the server, so as to determine which part of the memory in the cloud instance of the server can be used to write the target data. Specifically, the heterogeneous card can first generate a receiving request, which meets the format requirements of the first protocol and is used to instruct the heterogeneous card to request to obtain the available memory space information of the cloud instance of the server. Then, the heterogeneous card sends the first receiving request to the offload card.
[0236] 1302. The offload card sends the receiving request to the first driver of the cloud instance of the server through the direct channel.
[0237] After the unload card receives the receiving request from the heterogeneous card through the first driver of the unload card, the receiving request can be sent to the direct channel through the first driver of the unload card, so that the direct channel sends the receiving request to the first driver of the cloud instance of the server.
[0238] 1303. The offload card receives available memory space information of the first driver of the cloud instance from the server through the direct channel.
[0239] After receiving the request, the cloud instance of the server can obtain its own available memory space information based on the received request. Then, the cloud instance of the server sends the available memory space information to the direct channel of the offload card through its own first driver.
[0240] 1304. The offload card sends the available memory space information to the heterogeneous card through the first driver of the offload card.
[0241] After receiving the available memory space information through the direct channel, the offload card can send the available memory space information to the first driver of the offload card through the direct channel, so that the first driver of the offload card sends the available memory space information to the heterogeneous card.
[0242] 1305. The heterogeneous card sends a write request to the offload card based on the available memory space information, where the write request complies with the format requirement of the first protocol.
[0243] After obtaining the available memory space information, the heterogeneous card can determine the memory location of the cloud instance to which the target data is to be written based on the available memory space information, thereby generating a write request, which complies with the format requirements of the first protocol and is used to indicate that the heterogeneous card needs to write the target data. Then, the heterogeneous card sends the write request to the offload card.
[0244] 1306. The offload card sends the write request to the first driver of the cloud instance of the server through the direct channel.
[0245] 1307. The cloud instance of the server writes the target data to the memory of the cloud instance of the server based on the write request.
[0246] For the description of step 1306 to step 1307, please refer to the relevant description of step 1104 and step 1105, which will not be repeated here. At this point, the cloud instance of the server successfully writes the target data, and the data transmission process in the RX direction is completed.
[0247] In the server system provided in the embodiment of the present application, the offload card is connected between the server and the heterogeneous card. When the server system operates in a pass-through scenario, the cloud instance of the server and the offload card are both loaded with a first driver corresponding to the first protocol supported by the heterogeneous card. Specifically, after the offload card receives an operation request from the heterogeneous card through the first driver, the offload card sends the operation request to the first driver of the cloud instance of the server through a pass-through channel, so that the cloud instance of the server completes the corresponding operation based on the operation request. Therefore, the offload card can serve as a bridge between the cloud instance of the server and the heterogeneous card, providing a pass-through channel between the two, thereby realizing communication between the cloud instance of the server and the heterogeneous card.
[0248] Fig.14 Another structural diagram of the uninstall card provided in the embodiment of the present application is as follows: Fig.14 As shown, one end of the offload card is connected to the server, and the other end of the offload card is connected to the heterogeneous card. The offload card includes: a recording module 1401, which is used to record the mapping relationship between the first virtualization device of the offload card and the second virtualization device of the heterogeneous card, and a communication module 1402, which is used to realize the communication between the cloud instance of the server and the second virtualization device of the heterogeneous card according to the mapping relationship, wherein the first virtualization device of the offload card is directly connected to the cloud instance of the server, and the second virtual device of the heterogeneous card is directly connected to the offload card.
[0249] In a possible implementation, the server includes a first interface, the offload card includes a second interface and a third interface, and the heterogeneous card includes a fourth interface; the first interface is connected to the second interface, and the third interface is connected to the fourth interface.
[0250] In one possible implementation, the server system also includes an expansion card, the server includes a first interface, the expansion card includes a second interface, a third interface, a fourth interface and a fifth interface, the uninstall card includes a sixth interface and a seventh interface, and the heterogeneous card includes an eighth interface; the first interface is connected to the second interface, the second interface is connected to the third interface, the fourth interface is connected to the fifth interface, the sixth interface is connected to the third interface, the seventh interface is connected to the fourth interface, and the eighth interface is connected to the fifth interface.
[0251] In one possible implementation, the heterogeneous card supports a first protocol, the offload card is loaded with a first driver corresponding to the first protocol, the server supports a second protocol, and the cloud instance of the server is loaded with a second driver corresponding to the second protocol; the communication module 1402 is specifically used to: receive a first operation request from a second virtualization device through the first driver, the first operation request conforming to the format requirements of the first protocol; convert the first operation request into a second operation request, the second operation request conforming to the format requirements of the second protocol; send the second operation request to the second driver through the first virtualization device, so that the cloud instance of the server reads target data from the memory of the cloud instance of the server or writes target data to the memory of the cloud instance of the server based on the second operation request.
[0252] In a possible implementation, the offload card includes an engine module that supports the second protocol; the recording module 1401 is specifically used to: load the first driver after determining that the heterogeneous card is started; obtain device information of the heterogeneous card through the first driver; configure the engine module as a first virtualized device based on configuration information, and the configuration information is preset information; receive an enumeration result from the second driver through the first virtualized device, and the enumeration result is used to indicate that the cloud instance of the server has detected the first virtualized device; and construct a mapping relationship between the second virtualized device of the heterogeneous card and the first virtualized device of the offload card based on the device information and the enumeration result.
[0253] In one possible implementation, if the first operation request is a first read request, the communication module 1402 is specifically used to: receive a second send request from a second driver through a first virtualization device, the second send request complies with the format requirements of a second protocol, and the second send request is used to indicate that the cloud instance of the server is about to send target data; convert the second send request into a first send request, the first send request complies with the format requirements of the first protocol; send the first send request to the second virtualization device through the first driver, so that the heterogeneous card sends a first read request to the offload card through the second virtualization device based on the first send request.
[0254] In a possible implementation, the communication module 1402 is specifically configured to: receive target data from the second driver through the first virtualization device; and send the target data to the second virtualization device through the first driver.
[0255] In one possible implementation, if the first operation request is a first read request and the second operation request is a second read request, the communication module 1402 is also used to: send the second read request to the second driver through the first virtualization device, so that the cloud instance of the server reads the target data from the memory of the cloud instance of the server based on the second read request and transmits it to the first virtualization device; stores the target data in the memory of the offload card; receives the first read request from the second virtualization device through the first driver; reads the target data from the memory of the offload card based on the first read request; and sends the target data to the second virtualization device through the first driver.
[0256] In one possible implementation, if the first operation request is a first write request, the communication module 1402 is specifically used to: receive a first receive request from the second virtualization device through the first driver, the first receive request complies with the format requirements of the first protocol, and the first receive request is used to indicate that the heterogeneous card requests to obtain the available memory space information of the cloud instance of the server; convert the first receive request into a second receive request, the second receive request complies with the format requirements of the second protocol; send the second receive request to the second driver through the first virtualization device; receive the available memory space information from the second driver through the first virtualization device; send the available memory space information to the second virtualization device through the first driver, so that the heterogeneous card sends the first write request to the offload card through the second virtualization device based on the available memory space information.
[0257] Fig.15 Another structural diagram of the uninstall card provided in the embodiment of the present application is as follows: Fig.15 As shown, one end of the offload card is connected to the server, and the other end of the offload card is connected to the heterogeneous card, the heterogeneous card supports the first protocol, and the server and the offload card are both loaded with a first driver corresponding to the first protocol; the offload card includes: a communication module 1501, which is used to receive an operation request from the heterogeneous card through the first driver of the offload card, and the operation request complies with the format requirements of the first protocol; the communication module 1501 is also used to send the operation request to the first driver of the cloud instance of the server through a direct channel, so that the cloud instance of the server reads the target data from the memory of the cloud instance of the server or writes the target data to the memory of the cloud instance of the server based on the operation request.
[0258] In a possible implementation, the server includes a first interface, the offload card includes a second interface and a third interface, and the heterogeneous card includes a fourth interface; the first interface is connected to the second interface, and the third interface is connected to the fourth interface.
[0259] In one possible implementation, the server system also includes an expansion card, the server includes a first interface, the expansion card includes a second interface, a third interface, a fourth interface and a fifth interface, the uninstall card includes a sixth interface and a seventh interface, and the heterogeneous card includes an eighth interface; the first interface is connected to the second interface, the second interface is connected to the third interface, the fourth interface is connected to the fifth interface, the sixth interface is connected to the third interface, the seventh interface is connected to the fourth interface, and the eighth interface is connected to the fifth interface.
[0260] In a possible implementation, the communication module 1501 is also used to: receive an initialization request from the first driver of the cloud instance of the server through a direct channel; send the initialization request to the heterogeneous card through the first driver of the unloaded card, so that the heterogeneous card sends the device information of the heterogeneous card to the first driver of the unloaded card based on the initialization request; after receiving the device information of the heterogeneous card through the first driver of the unloaded card, send the device information of the heterogeneous card to the first driver of the cloud instance of the server through the direct channel.
[0261] In one possible implementation, if the operation request is a read request, the communication module 1501 is also used to: receive a send request from the first driver of the cloud instance of the server through a direct channel, the send request complies with the format requirements of the first protocol, and the send request is used to indicate that the cloud instance of the server is about to send target data; send the send request to the heterogeneous card through the first driver of the unload card, so that the heterogeneous card sends a read request to the first driver of the unload card based on the send request.
[0262] In a possible implementation, the communication module 1501 is further used to: receive target data of a first driver from a cloud instance of a server through a through channel; and send the target data to a heterogeneous card through the first driver of the offload card.
[0263] In one possible implementation, the communication module 1501 is also used to: send a read request to the first driver of the cloud instance of the server through a direct channel, so that the cloud instance of the server reads the target data from the memory of the cloud instance of the server based on the read request of the offload card, and sends the target data to the direct channel through the first driver of the cloud instance of the server; stores the target data in the memory of the offload card; receives the read request from the heterogeneous card through the first driver of the offload card; reads the target data from the memory of the offload card based on the read request of the heterogeneous card; and sends the target data to the heterogeneous card through the first driver of the offload card.
[0264] In one possible implementation, if the operation request is a write request, the communication module 1501 is also used to: receive a receive request from a heterogeneous card through the first driver of the unload card, the receive request complies with the format requirements of the first protocol, and the receive request is used to indicate that the heterogeneous card requests to obtain available memory space information of the cloud instance of the server; send the receive request to the first driver of the cloud instance of the server through a direct channel; receive available memory space information from the first driver of the cloud instance of the server through the direct channel; send the available memory space information to the heterogeneous card through the first driver of the unload card, so that the heterogeneous card sends a write request to the unload card based on the available memory space information.
[0265] It should be noted that the information interaction, execution process, etc. between the modules / units of the above-mentioned device are based on the same concept as the method embodiment of the present application, and the technical effects they bring are the same as those of the method embodiment of the present application. For specific contents, please refer to the description in the method embodiment shown above in the embodiment of the present application, and will not be repeated here.
[0266] Fig.16 A schematic diagram of the structure of the uninstall card provided in the embodiment of the present application. Fig.16 As shown, an embodiment of the uninstall card in the present application may include one or more central processors 1601, a memory 1602, an input and output interface 1603, a wired or wireless network interface 1604, and a power supply 1605.
[0267] The memory 1602 may be a temporary storage or a permanent storage. Furthermore, the central processor 1601 may be configured to communicate with the memory 1602 and execute a series of instruction operations in the memory 1602 on the offload card.
[0268] In this embodiment, the CPU 1601 may execute the aforementioned Figures 6 to 13 The operations performed by the uninstall card in the illustrated embodiment will not be described in detail here.
[0269] In this embodiment, the specific functional module division in the central processor 1601 can be the same as the above Fig.14 The division method of the recording module and communication module described in is similar and will not be repeated here.
[0270] In this embodiment, the specific functional module division in the central processor 1601 can also be the same as the above Figure 6 The division method of the communication module and other modules described in is similar and will not be repeated here.
[0271] The embodiment of the present application also relates to a computer storage medium, including computer readable instructions. When the computer readable instructions are executed, the following is implemented: Figures 6 to 13 The method described.
[0272] The present application also relates to a computer program product comprising instructions, which, when executed on a computer, causes the computer to execute Figures 6 to 13 The method described.
[0273] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0274] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.
[0275] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0276] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0277] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk.
Claims
1. A server system, characterized in that: The server system comprises: a server, an offload card and a heterogeneous card, wherein the offload card is connected to the server, and the heterogeneous card is connected to the offload card; The offload card is used to record a mapping relationship between a first virtualized device of the offload card and a second virtualized device of the heterogeneous card, and to implement communication between a cloud instance of a server and the second virtualized device of the heterogeneous card according to the mapping relationship, wherein the first virtualized device of the offload card is directly connected to the cloud instance of the server, and the second virtualized device of the heterogeneous card is directly connected to the offload card; The heterogeneous card supports a first protocol, the offload card is loaded with a first driver corresponding to the first protocol, the server supports a second protocol, and the cloud instance of the server is loaded with a second driver corresponding to the second protocol; The offload card is specifically used to receive a first operation request from the second virtualization device through the first driver, where the first operation request complies with the format requirement of the first protocol; The offload card is specifically used to convert the first operation request into a second operation request, where the second operation request complies with the format requirement of the second protocol; The offload card is specifically used to send the second operation request to the second driver through the first virtualization device, so that the cloud instance of the server reads the target data from the memory of the cloud instance of the server or writes the target data to the memory of the cloud instance of the server based on the second operation request.
2. The server system according to claim 1, characterized in that: The server includes a first interface, the offload card includes a second interface and a third interface, and the heterogeneous card includes a fourth interface; The first interface is connected to the second interface, and the third interface is connected to the fourth interface.
3. The server system according to claim 1, characterized in that: The server system further comprises an expansion card, the server comprises a first interface, the expansion card comprises a second interface, a third interface, a fourth interface and a fifth interface, the uninstall card comprises a sixth interface and a seventh interface, and the heterogeneous card comprises an eighth interface; The first interface is connected to the second interface, the second interface is connected to the third interface, the fourth interface is connected to the fifth interface, the sixth interface is connected to the third interface, the seventh interface is connected to the fourth interface, and the eighth interface is connected to the fifth interface.
4. The server system according to claim 1, characterized in that: The offload card includes an engine module supporting the second protocol; The offload card is specifically used to load the first driver after determining that the heterogeneous card is started; The offload card is specifically used to obtain the device information of the heterogeneous card through the first driver; The offload card is specifically used to configure the engine module as the first virtualization device based on configuration information, and the configuration information is preset information; The offload card is specifically used to receive an enumeration result from the second driver through the first virtualization device, where the enumeration result is used to indicate that the cloud instance of the server has detected the first virtualization device; The offload card is specifically used to construct a mapping relationship between the second virtualized device of the heterogeneous card and the first virtualized device of the offload card based on the device information and the enumeration result.
5. The server system according to any one of claims 1 to 4, characterized in that: If the first operation request is a first read request, the offload card is specifically used to receive a second sending request from the second driver through the first virtualization device, the second sending request conforms to the format requirement of the second protocol, and the second sending request is used to indicate that the cloud instance of the server is about to send the target data; The offload card is specifically used to convert the second sending request into a first sending request, and the first sending request complies with the format requirement of the first protocol; The offload card is specifically configured to send the first sending request to the second virtualization device through the first driver, so that the heterogeneous card sends the first read request to the offload card through the second virtualization device based on the first sending request.
6. The server system according to any one of claims 1 to 4, characterized in that: The offload card is specifically used to receive target data from the second driver through the first virtualization device; The offload card is specifically configured to send the target data to the second virtualization device through the first driver.
7. The server system according to claim 5, characterized in that: If the first operation request is a first read request and the second operation request is a second read request, the offload card is further used to send the second read request to the second driver through the first virtualization device, so that the cloud instance of the server reads the target data from the memory of the cloud instance of the server based on the second read request and transmits the target data to the first virtualization device; The offloading card is also used to store the target data in the memory of the offloading card; The offload card is further configured to receive the first read request from the second virtualization device through the first driver; The offload card is further used to read the target data from the memory of the offload card based on the first read request; The offload card is further configured to send the target data to the second virtualization device through the first driver.
8. The server system according to any one of claims 1 to 4, characterized in that: If the first operation request is a first write request, the offload card is specifically used to receive a first receive request from the second virtualization device through the first driver, the first receive request conforms to the format requirement of the first protocol, and the first receive request is used to instruct the heterogeneous card to request to obtain the available memory space information of the cloud instance of the server; The offload card is specifically used to convert the first receiving request into a second receiving request, where the second receiving request complies with the format requirement of the second protocol; The offload card is specifically configured to send the second receiving request to the second driver through the first virtualization device; The offload card is specifically used to receive the available memory space information from the second driver through the first virtualization device; The offload card is specifically used to send the available memory space information to the second virtualization device through the first driver, so that the heterogeneous card sends the first write request to the offload card through the second virtualization device based on the available memory space information.
9. A method of data processing, characterized in that: The method is applied to a server system, the server system comprising: a server, an offload card and a heterogeneous card, the offload card is connected to the server, and the heterogeneous card is connected to the offload card; The offload card records a mapping relationship between a first virtualized device of the offload card and a second virtualized device of the heterogeneous card; The offload card implements communication between the cloud instance of the server and the second virtualized device of the heterogeneous card according to the mapping relationship, wherein the first virtualized device of the offload card is directly connected to the cloud instance of the server, and the second virtualized device of the heterogeneous card is directly connected to the offload card; The heterogeneous card supports a first protocol, the offload card is loaded with a first driver corresponding to the first protocol, the server supports a second protocol, and the cloud instance of the server is loaded with a second driver corresponding to the second protocol, and the offload card implements communication between the cloud instance of the server and the second virtualized device of the heterogeneous card according to the mapping relationship, including: The offload card receives a first operation request from the second virtualization device through the first driver, where the first operation request complies with a format requirement of the first protocol; The offloading card converts the first operation request into a second operation request, where the second operation request complies with the format requirement of the second protocol; The offload card sends the second operation request to the second driver through the first virtualization device, so that the cloud instance of the server reads the target data from the memory of the cloud instance of the server or writes the target data to the memory of the cloud instance of the server based on the second operation request.
10. The method according to claim 9, characterized in that The offload card includes an engine module supporting the second protocol, and the offload card records a mapping relationship between a first virtualized device of the offload card and a second virtualized device of the heterogeneous card, including: After determining that the heterogeneous card is started, the offload card loads the first driver; The offload card obtains the device information of the heterogeneous card through the first driver; The offload card configures the engine module as the first virtualization device based on configuration information, where the configuration information is preset information; The offload card receives an enumeration result from the second driver through the first virtualization device, where the enumeration result is used to indicate that the cloud instance of the server has detected the first virtualization device; The offload card constructs a mapping relationship between the second virtualized device of the heterogeneous card and the first virtualized device of the offload card based on the device information and the enumeration result.
11. The method according to claim 9 or 10, characterized in that: If the first operation request is a first read request, before the offload card receives the first operation request from the second virtualization device through the first driver, the method further includes: The offload card receives a second sending request from the second driver through the first virtualization device, where the second sending request complies with the format requirement of the second protocol, and the second sending request is used to indicate that the cloud instance of the server is about to send the target data; The offloading card converts the second sending request into a first sending request, where the first sending request complies with the format requirement of the first protocol; The offload card sends the first sending request to the second virtualization device through the first driver, so that the heterogeneous card sends the first read request to the offload card through the second virtualization device based on the first sending request.
12. The method according to claim 9 or 10, characterized in that: The method further comprises: The offload card receives target data from the second drive through the first virtualization device; The offload card sends the target data to the second virtualization device through the first driver.
13. The method according to claim 11, characterized in that If the first operation request is a first read request and the second operation request is a second read request, after the offload card sends the first sending request to the second virtualization device through the first driver, the method further includes: The offload card sends the second read request to the second driver through the first virtualization device, so that the cloud instance of the server reads the target data from the memory of the cloud instance of the server based on the second read request and transmits the target data to the first virtualization device; The offloading card stores the target data in the memory of the offloading card; The offload card receives the first read request from the second virtualization device through the first driver; The offload card reads the target data from the memory of the offload card based on the first read request; The offload card sends the target data to the second virtualization device through the first driver.
14. The method according to claim 9 or 10, characterized in that If the first operation request is a first write request, before the offload card receives the first operation request from the second virtualization device through the first driver, the method further includes: The offload card receives a first receiving request from the second virtualization device through the first driver, the first receiving request conforms to the format requirement of the first protocol, and the first receiving request is used to instruct the heterogeneous card to request to obtain the available memory space information of the cloud instance of the server; The offloading card converts the first receiving request into a second receiving request, where the second receiving request complies with the format requirement of the second protocol; The offload card sends the second receiving request to the second driver through the first virtualization device; The offload card receives the available memory space information from the second driver through the first virtualization device; The offload card sends the available memory space information to the second virtualization device through the first driver, so that the heterogeneous card sends the first write request to the offload card through the second virtualization device based on the available memory space information.
15. An uninstall card, characterized in that: One end of the offload card is connected to the server, and the other end of the offload card is connected to the heterogeneous card. The offload card includes: A recording module, used for recording a mapping relationship between the first virtualized device of the offload card and the second virtualized device of the heterogeneous card; A communication module, used for realizing communication between the cloud instance of the server and the second virtualized device of the heterogeneous card according to the mapping relationship, wherein the first virtualized device of the offload card is directly connected to the cloud instance of the server, and the second virtualized device of the heterogeneous card is directly connected to the offload card; The heterogeneous card supports a first protocol, the offload card is loaded with a first driver corresponding to the first protocol, the server supports a second protocol, and the cloud instance of the server is loaded with a second driver corresponding to the second protocol, and the communication module is used to: receiving, through the first driver, a first operation request from the second virtualization device, where the first operation request complies with a format requirement of the first protocol; Converting the first operation request into a second operation request, where the second operation request complies with the format requirement of the second protocol; The second operation request is sent to the second driver through the first virtualization device, so that the cloud instance of the server reads the target data from the memory of the cloud instance of the server or writes the target data to the memory of the cloud instance of the server based on the second operation request.
16. An uninstall card, characterized in that: The offloading card comprises: a processor and a memory; The memory is used to store computer-executable instructions; The processor is used to execute the computer-executable instructions stored in the memory, so that the unloading card implements the method as described in any one of claims 9 to 14.
17. A computer storage medium, characterized in that: The method comprises computer readable instructions, which, when executed, implement the method according to any one of claims 9 to 14.
18. A computer program product comprising instructions, which, when executed on a computer, causes the computer to perform the method according to any one of claims 9 to 14.
