WO2022194021A1 - Concurrency control method, network card, computer device, and storage medium - Google Patents

Abstract

The present application belongs to the technical field of concurrency control. Disclosed are a concurrency control method, a network card, a computer device, and a storage medium. The method comprises: a network card acquiring an object, wherein the object carries data that a requesting end requests to read from a destination end, and check information of the data; the network card performing consistency check on the data according to the check information; and when the consistency check is passed, the network card providing the data for a memory of the requesting end. By means of the present application, CPU overheads are effectively saved on.

Description

Concurrency control method, network card, computer equipment, storage medium

This application claims the priority of the Chinese patent application filed on March 16, 2021 with the application number 202110281600.X and the invention titled "Concurrency control method, network card, computer equipment, storage medium", the entire contents of which are incorporated by reference in in this application.

technical field

The present application relates to the technical field of concurrency control, and in particular, to a concurrency control method, a network card, a computer device, and a storage medium.

Background technique

Concurrency control regards the transaction as the basic unit of scheduling. The purpose of concurrency control is to reasonably arrange the concurrent scheduling of transactions to avoid conflicts and improve the correctness and utilization of the system.

In the related art, when the requester needs to read data from the destination, in order to perform concurrency control, it is necessary to perform a consistency check on the data requested by the requester to ensure that the request is not read during the process of reading the data. The fetched data performs an update operation, that is, there is no concurrency conflict in the operation of reading the data.

However, the verification operation is usually performed by a central processing unit (CPU) at the requesting side, resulting in a large CPU overhead.

SUMMARY OF THE INVENTION

The present application provides a concurrency control method, a network card, a computer device, and a storage medium. The present application effectively reduces CPU overhead. The technical solutions provided by this application are as follows:

In a first aspect, the present application provides a concurrency control method. The method includes: a network card obtains an object, and the object carries data requested by a requester to be read from a destination and verification information of the data; Perform consistency check; when the consistency check passes, the network card provides data to the memory of the requester.

By using the network card for consistency check, the CPU in the server does not need to be used for consistency check, and on the basis of realizing the concurrency control of the reading process, the CPU overhead caused by check and memory transfer can be effectively reduced. End-to-end delay is reduced.

In an implementation manner, when the consistency check is passed, the network card provides data to the memory of the requester, including: when the consistency check is passed, the network card performs a verification information stripping operation on the object to obtain data; The memory on the side provides the data.

By using the network card to perform the stripping operation, it is unnecessary to use the CPU in the server to perform the stripping operation. On the basis of realizing the concurrency control of the reading process, the CPU overhead due to checksum memory transfer can be effectively reduced, and the end-to-end end delay.

In an implementation manner, when the network card is deployed in the destination, after the network card performs consistency check on the data according to the verification information, the method further includes: when the consistency check fails, the network card notifies the requester to verify the data. Failed to pass the test. By timely notifying the requester that the consistency check fails, the overhead caused by processing invalid data can be effectively reduced.

In an implementation manner, when the network card is deployed in the destination, after the network card performs consistency check on the data according to the verification information, the method further includes: when the consistency check fails, the network card stops checking the data in the data. The unverified part is checked for consistency, and the part that has passed the verification of the data is stopped to be sent to the requester. In this way, it is possible to reduce the resource overhead and time overhead used in the consistency check and transmission of invalid data, improve the utilization rate of resources and ensure the concurrency of the destination end.

In an implementation manner, when the network card is deployed in the requester, after the network card performs consistency check on the data according to the verification information, the method further includes: when the consistency check fails, the network card stops storing the data in the data. The part that has passed the verification is written to the memory of the requester. In this way, the resource overhead and time overhead of transmitting invalid data can be reduced, the utilization rate of resources can be improved, and the concurrency of the destination end can be guaranteed.

In an implementation manner, when the network card is deployed in the requester, after the network card performs consistency check on the data according to the verification information, the method further includes: when the consistency check fails, the network card stops checking the data in the data. The unverified part is checked for consistency. In this way, it is possible to reduce the resource overhead and time overhead used for the consistency check of invalid data, improve the utilization rate of resources and ensure the concurrency of the destination end.

Optionally, the object can be read from memory when the network card is deployed in the destination. As an achievable manner, obtaining the object by the network card includes: the network card receives a read request sent by the requester, and the read request is used to request to read data; and the network card obtains the object from the memory of the destination terminal according to the read request.

In addition, the network card provides data to the memory of the requester, including: the network card sends data to the network card of the requester, so that the network card of the requester sends data to the memory of the requester.

Optionally, when the network card is deployed in the requester, the object is provided by the destination. As an achievable manner, before the network card obtains the object, the method further includes: the network card sends a read request to the destination, and the read request is used to request to read data; correspondingly, the network card obtains the object, including: the network card receives the network card sent by the destination end. The object is obtained by the network card of the destination based on the read request.

In an implementation manner, when the network card is deployed in the requester, the network card provides data to the memory of the requester, including: the network card directly sends data to the memory of the requester.

Optionally, when the network card provided by this application can support a variety of read and write operations of different read and write scales, the organization forms of the verification information of the data involved in different read and write scales in the object may be the same or different. The verification information can be described by description information. Correspondingly, before the network card performs consistency check on the data according to the verification information, the method further includes: the network card obtains description information of the object, and the description information is used to indicate the organization form of the object; and the network card obtains from the object based on the description information. Check information.

Among them, in order to distinguish the description information used to describe different organizational forms, an identifier can also be assigned to the description information used to describe different organizational forms. Correspondingly, the network card obtains the description information of the object, including: the network card obtains the identifier of the description information; the network card is based on the The identification and mapping relationship of the description information is obtained, and the description information is obtained, and the mapping relationship reflects the corresponding relationship between the description information and the identification.

In an implementation manner, when the network card is deployed in the destination terminal, the network card obtains the identifier of the description information, including: the network card receives a read request sent by the requester, and the read request carries the identifier of the description information; the network card obtains from the read request An identifier that describes the information.

In another implementation manner, when the network card is deployed in the requester, the network card obtains the identifier of the description information, including: the network card receives a read response sent by the destination terminal based on the read request, and the read response carries the identifier of the description information, and the description information The identifier of the id is sent to the destination through the read request; the network card obtains the identifier of the description information from the read response.

In a second aspect, the present application provides a network card, the network card includes: an acquisition unit for acquiring an object, the object carrying data requested by a requester to be read from a destination and verification information of the data; a verification unit for The verification information is used to check the consistency of the data; the interaction unit is used to provide data to the memory of the requester when the consistency check is passed.

Optionally, the network card further includes: a stripping unit, configured to perform a verification information stripping operation on the object to obtain data when the consistency check is passed; an interaction unit, specifically configured to provide data to the memory of the requester.

Optionally, the network card is deployed in the destination, and the interaction unit is further configured to notify the requester that the verification fails when the consistency check fails.

Optionally, the network card is deployed in the destination, and the check unit is also used to stop the consistency check on the unchecked part of the data when the consistency check fails; the interaction unit is also used to check the consistency When the verification fails, stop sending the part of the data that has passed the verification to the requester.

Optionally, the network card is deployed in the requester, and the interaction unit is further configured to stop writing the part of the data that has passed the check into the memory of the requester when the consistency check fails.

Optionally, the network card is deployed in the requester, and the verification unit is further configured to stop the consistency verification of the unverified part of the data when the consistency verification fails.

Optionally, the interaction unit is further configured to receive a read request sent by the requesting end, and the read request is used to request to read data; the obtaining unit is specifically configured to obtain the object from the memory of the destination end according to the read request.

Optionally, the network card is deployed in the destination, and the interaction unit is specifically configured to send data to the network card of the requester, so that the network card of the requester sends data to the memory of the requester.

Optionally, the network card is deployed in the requesting end, and the interaction unit is also used to send a read request to the destination end, and the read request is used to request to read data; the acquiring unit is specifically used to receive the object sent by the network card of the destination end, and the object is sent by the destination end. The network card of the end is obtained in the destination end based on the read request.

Optionally, the network card is deployed in the requester, and the interaction unit is specifically configured to directly send data to the memory of the requester.

Optionally, the acquiring unit is further configured to acquire description information of the object, where the description information is used to indicate the organization form of the object; the verification unit is further configured to acquire verification information from the object based on the description information.

Optionally, the obtaining unit is specifically configured to: obtain the identifier of the description information; and obtain the description information based on the identifier of the description information and a mapping relationship, and the mapping relationship reflects the corresponding relationship between the description information and the identifier.

Optionally, the network card is deployed in the destination terminal, and the interaction unit is further configured to receive a read request sent by the requester, and the read request carries the identifier of the description information; the acquisition unit is specifically configured to obtain the identifier of the description information from the read request.

Optionally, the network card is deployed in the requester, and the interaction unit is further configured to receive a read response sent by the destination based on the read request, the read response carries an identifier of the description information, and the identifier of the description information is sent to the destination through the read request; The unit is specifically used to obtain the identifier of the description information from the read response.

In a third aspect, the present application provides a computer device, the computer device includes: a network card and a memory, and a computer program is stored in the memory; when the network card executes the computer program, the first aspect of the present application and any possible implementation manner thereof are realized. provided method.

In a fourth aspect, the present application provides a non-transitory computer-readable storage medium. When an instruction in the computer-readable storage medium is executed by a network card, the network card executes the first aspect of the present application and any possible implementations thereof. method provided in.

In a fifth aspect, the present application provides a computer program product including instructions, which, when the computer program product runs on a network card, enables the network card to execute the method provided in the first aspect of the present application and any possible implementations thereof.

Description of drawings

1 is a schematic diagram of an application scenario involved in a concurrency control method provided by an embodiment of the present application;

FIG. 2 is a schematic structural diagram of a network card provided by an embodiment of the present application;

3 is a flowchart of a concurrency control method provided by an embodiment of the present application;

4 is a schematic diagram of a storage method of an object provided by an embodiment of the present application;

FIG. 5 is a schematic diagram of another object storage method provided by an embodiment of the present application;

6 is a schematic diagram of a method for performing concurrency control on a network card of a destination end provided by an embodiment of the present application;

7 is a flowchart of another concurrency control method provided by an embodiment of the present application;

8 is a schematic diagram of a method for performing concurrency control on a network card of a requester provided by an embodiment of the present application;

9 is a schematic structural diagram of another network card provided by an embodiment of the present application;

10 is a schematic structural diagram of another network card provided by an embodiment of the present application;

FIG. 11 is a schematic structural diagram of a computer device provided by an embodiment of the present application.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present application clearer, the embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

For ease of understanding, the relevant background knowledge involved in the embodiments of the present application is first introduced below.

A transaction is a logical unit of operation, and a transaction is a sequence of operations consisting of all operations performed between the start of a transaction and the end of a transaction. All operations in a transaction are executed as a whole. All operations in the entire transaction are either successfully executed or all fail. When a transaction fails, an abort or rollback strategy can be used to ensure that the operation objects in the transaction will not be affected.

Concurrency control (CC) is used to serialize multiple concurrent transactions that operate on the same object, and is a mechanism to ensure timely detection and correction of errors caused by concurrent operations. When the concurrent transactions of an object include read operation transactions and write operation transactions, the read operation transaction and the write operation transaction can be serialized through concurrency control to prevent the simultaneous execution of the write operation during the read operation and the simultaneous execution of the read operation during the read operation. Read operations, concurrently perform read operations during a write operation, or perform concurrent operations such as other write operations during a write operation to ensure data consistency.

Most implementation systems implement concurrency control using locking methods. The principle of using the lock method to achieve concurrency control is: when a transaction operates on an object, the object is locked to prevent other transactions from operating on the object during the execution of the transaction, and the object is released after the transaction is completed. lock. However, the locking method will increase the overhead of the implementation system and reduce the potential concurrency of the implementation system.

Another mechanism for implementing concurrency control is optimistic concurrency control (OCC). Optimistic concurrency control is a lock-free concurrency control mechanism, which is based on an optimistic conflict detection strategy for concurrency control, which can maximize parallel efficiency on the basis of concurrency control. When optimistic concurrency control is used, the execution process of a transaction includes an execute phase, a verification phase, and a commit phase. During the execution phase, objects are manipulated according to the logic of the transaction. In the verification phase, it is judged whether the operation of the transaction on the object conflicts with the operations of other transactions on the object. If the operation of the transaction on the object does not conflict with the operations of other transactions on the object, the transaction is allowed to enter the commit phase, otherwise , you need to use a conflict resolution mechanism for conflicting transactions. During the commit phase, the updates generated by the transaction's operations on the object are committed, making the updates persistent.

A distributed transaction is a transaction involving multiple servers, and the realization of the distributed transaction requires the cooperation of the multiple servers. In a distributed transaction, each server applies local concurrency control to the objects it manages to ensure that the transaction is serialized locally, and concurrent control of the entire distributed transaction is required to ensure global serialization .

Among them, the use of optimistic concurrency control can control the concurrency of the lock-free read process in distributed transactions. However, the process of using optimistic concurrency control to achieve lock-free reading will result in a large central processing unit (CPU) overhead in the server.

The embodiments of the present application provide a concurrency control method. The network card obtains the object, performs consistency verification on the data in the object according to the verification information carried by the object, and provides data to the memory of the requester when the consistency verification passes. The object carries the data requested by the requester to be read from the destination and the verification information of the data. In the concurrency control method, by using a network card to perform consistency check, the CPU in the server does not need to be used for consistency check, and the CPU overhead can be effectively reduced on the basis of realizing the concurrency control of the reading process. The data carried by the object is a general term for the content requested by the requester to be read from the destination.

FIG. 1 is an implementation scenario involved in a concurrency control method provided by an embodiment of the present application. The implementation scenario includes: at least two service nodes. A network card is deployed in each of the at least two service nodes, and data transmission between the at least two service nodes is implemented through the network card. In the scenario of reading data, one of the at least two service nodes is used as a destination, and the destination stores data. Among the at least two service nodes, one or more service nodes other than the service node serving as the destination serve as a requester, and the requester is used for requesting to read data from the destination. 1 is a schematic diagram of the application scenario including a first service node 10 and a second service node 20, one of the first service node 10 and the second service node 20 is used as a requester, and the other is used as a destination.

Optionally, the service node in this implementation scenario may be implemented by a physical machine, a physical machine cluster including multiple physical machines, a bare metal server, or a cloud server, or the like.

The process of the requester reading data from the destination includes: the requester sends a read request to the destination, the destination obtains the data read by the read request from the destination according to the read request, and provides the data to the requester. In addition, in order to control the concurrency of the process of reading data, the process of reading data further includes: performing consistency check on the data read by the read request. In this embodiment of the present application, the consistency check process may be performed by a network card at the destination end or by a network card at the requesting end.

In addition, the destination usually stores data in the memory in the unit of an object (object), wherein the object includes data and verification information for verifying the data. The data in the object is usually stored in the form of data blocks, that is, the object usually includes one or more data blocks. In addition, each data block has verification information, and the verification information of any data block can be located at the head or tail of any data block, or the verification information in multiple data blocks in the object can also be performed in other ways. place. What the requester needs to obtain is the data in the object, so when the data passes the verification, the network card is also used to strip (stripping) the verification information of the data in the object, such as stripping the verification information at the head of the data block from the object. information, obtain the data requested by the requester to read, and provide the data to the memory of the requester. Performing the stripping operation through the network card can reduce the CPU and memory overhead caused by the CPU performing the stripping operation.

Further, the network card provided by the embodiment of the present application can support read and write operations of various read and write scales, and the data and verification information of the read and write operations of the various read and write scales are organized in different forms in objects. Among them, the organization form is used to indicate how the data and verification information are stored in the object. Organizational form can be reflected using descriptive information. When data consistency check is required, the network card needs to first obtain the description information of the object that contains the data, and then obtain the data check information from the object according to the description information, and then perform the data check according to the check information. Consistency check.

In an achievable manner, operations such as checksum stripping performed by the network card in the embodiment of the present application may be implemented by program code. Correspondingly, the network card for performing the consistency check may be a programmable network card. The programmable network card is also called a programmable controller (programmable network interface controller, Programmable NIC). When the network card is a programmable network card, the behavior of the network card can be changed by modifying program codes such as firmware or microcode in the programmable network card, so that the network card has the above functions.

In addition, in order to further reduce the delay of the CPU, the network card is configured with a remote direct memory access (RDMA) function. Among them, RDMA is a fast cross-machine memory access technology ubiquitous in the distributed field. RDMA allows a machine to perform memory read and write operations in another target machine's pre-registered memory area, and can bypass the target machine's operating system and CPU, so it has the advantages of high speed, low latency, and low CPU overhead.

Optionally, the above-mentioned functions of the network card may be implemented by multiple functional modules, and the multiple functional modules cooperate to implement the above-mentioned functions of the service node. The following takes Figure 2 as an example to illustrate it:

As shown in FIG. 2 , when the service node includes a network card for performing operations such as checksum stripping, the service node includes: a host system (host system) X01 and a programmable network card X02. The host system X01 runs an application program that performs read and write operations on data. Correspondingly, the host system X01 includes an initialization module (initializing module) X011 that serves the application program. The initialization module X011 is used to perform the initialization configuration required to run the application. And, the service node is configured with a memory X012 for the host system X01. Memory X012 is used to store objects that distributed transactions need to access.

As shown in Figure 2, the programmable network card X02 includes: a network card module X021, a first buffer module X022, a match module (match module) X023, a verification module (validation module) X024, a stripping module (stripping module) X025 and The second cache module X026. The verification information includes version information (version), and performing consistency verification on the data is also referred to as performing version verification on the data, and the verification module X024 is also referred to as a version verification module. The network card module X021 is the hardware that allows the service nodes to communicate on the computer network, and is used to receive data sent by the network cards of other service nodes through the network, and send data to the network cards of other service nodes through the network. The first cache module X022 is used to cache the object to be verified. The matching module X023 is used to obtain the description information of the object. The verification module X024 is configured to acquire the description information of the object from the matching module, acquire the verification information of the data from the object cached by the first cache module according to the description information, and perform consistency verification on the data according to the verification information. The stripping module X025 is used to strip the verification information in the object when the data passes the verification, and obtain the data requested by the requester to be read. The second cache module X026 is used to cache the verification information that has been verified.

Optionally, the network card module X021, the first buffer module X022 and the second buffer module X026 can be implemented by hardware, such as the first buffer module X022 can be a stream buffer (stream buffer), and the second buffer module X026 can be a context buffer. device (contex). The matching module X023, the verification module X024 and the stripping module X025 can be implemented by software. It should be noted that the implementations of the network card module X021, the first cache module X022, the matching module X023, the verification module X024, the stripping module X025 and the second cache module X026 are schematic examples, which are different from the implementations that define the corresponding modules. , the implementation of each module can be adjusted according to application requirements.

As mentioned above, the concurrency control method provided by the embodiment of the present application may be executed by the destination end or by the requesting end. The implementation process of the concurrency control method is described below with respect to the concurrency control method being executed by the destination end and the request end being executed respectively.

FIG. 3 is a method flowchart of a concurrency control method provided by an embodiment of the present application executed by a destination end. As shown in Figure 3, the concurrency control method includes the following steps:

Step 301 , the application program of the destination end registers the description information of the object, and sends the identifier of the description information to the requesting end through the network card of the destination end.

When an application is run in the destination, the object of the application is stored in the memory, and the destination can provide the requesting end with an access function of the object. Among them, the object includes the relevant data of the application and the verification information of the data. In an object, check information is placed at the head, tail, or otherwise of the data. In addition, the data related to the application includes: data read from the disk into the memory, data generated during the running of the application, and data written by other service nodes through the application.

When the network card provided in the embodiment of the present application can support multiple read and write operations of different read and write scales, the organization forms of the verification information of the data involved in different read and write scales in the object may be the same or different. For example, the verification information of data accessed by threads of different applications may be organized in the same or different forms in objects. At this time, in order for the network card to obtain the verification information of the data, the application program needs to register the description information of the object. The description information is used to indicate the organization form of the object, such as the organization form of the indicated data and verification information in the object. The application program registers the description information means that the application program determines the organization form of the data it accesses and its verification information in the object according to its execution logic, and obtains the object description information according to the organization form of the object.

In addition, in order to distinguish the description information used to describe different organizational forms, an identifier can also be assigned to the description information used to describe different organizational forms, and a mapping relationship between the description information and the identifier can be established, so that in the process of subsequent data reading, the network card The description information indicated by the identification can be determined according to the identification. Optionally, the operation of assigning an identifier to the description information can be performed by the application program or the network card of the destination end, as long as it can be ensured that one identifier can uniquely identify one description information. The network card stores a mapping relationship table, and the mapping relationship table is used to describe the mapping relationship between the description information and the identifier.

Since there is usually a lot of data stored in the memory of the destination end, and the destination end cannot predict the data that the requesting end needs to read, correspondingly, the destination end cannot predict the verification information of the data that the requesting end needs to read. Therefore, in order for the destination end to be able to determine the verification information of the data to be read by the requesting end, the read request sent by the requesting end to the destination end needs to carry the identifier or description information of the description information of the object. Correspondingly, after determining the description information of the object, the destination also needs to notify the requester of the description information used to describe different organizational forms, so that when the requester needs to read data from the destination, the requester can carry the corresponding data in the read request. description information. Alternatively, after the destination assigns identifiers to the description information used to describe different organizational forms, the destination also needs to notify the requester of the identifiers used to describe the description information of different organizational forms, so that when the requester needs to read data from the destination , the requester can carry the identifier of the corresponding description information in the read request. Optionally, the application at the destination end may notify the application at the requesting end of description information or its identifier for describing different organizational forms to the application at the requesting end in an in-band or out-of-band manner.

The data in one object may be divided into multiple data blocks, and in order to facilitate verification of each data block, the object also carries verification information of each data block. When the memory caches data in the form of cache lines, each cache line can store one data block, or multiple cache lines can jointly store one data block. The following uses Figure 4 and Figure 5 as examples to illustrate this. Two implementations are described.

As shown in Figure 4, the data in the object is divided into 3 data blocks, each cache line of the 3 cache lines stores a data block in the object, and each cache line also stores the cache line Checksum information of data blocks stored in . When verifying the data in the object, it is necessary to verify the data blocks stored in the corresponding cache lines according to the verification information of the data blocks stored in each cache line. The lengths of the verification information in the multiple cache lines may be the same or different, and/or the contents included in the verification information in the multiple cache lines may be the same or different. For example, the length of each cache line is 256 bytes. According to the instructions of the description information, it can be seen that the check information is located at the head of the data block, and the length of the check information in each cache line is 8 bytes. Both are 248 bytes in length.

As shown in Figure 5, the data in the object is divided into two data blocks, each data block is stored in two cache lines, and the header of each data block also stores the check information of the data block. When verifying the data in the object, the corresponding data block can be verified according to the verification information in the header of each data block.

The storage of one data block in the object by several cache lines can be determined according to application requirements, for example, it can be determined according to the atomic read and write granularity of the network card. When the atomic read/write granularity of the network card is one cache line, each cache line is used to store one data block. When the atomic read/write granularity of the network card is multiple cache lines, multiple cache lines store one data block together.

In an implementation manner, the description information of the object may be represented by a pattern descriptor (pattern descriptor) of the object, and the identifier of the corresponding description information may be represented by a pattern identifier (pattern identify, pattern ID for short). Correspondingly, the aforementioned matching module X023 is also referred to as a pattern matching module (pattern match module), and the mapping relationship table used to describe the pattern and the pattern identifier may be referred to as a pattern table (pattern table).

Optionally, the mode descriptor includes: a length field (length) and an offset field (offset). The length field is used to indicate the valid length of the check information. The offset field is used to indicate the offset of the verification information in the object. When the mode descriptor is used to define the effective length and offset of the verification information, the storage location of the verification information in the object can be determined according to the effective length and the offset. And since the object includes two parts, the data and the verification information, after the storage location of the verification information in the object is determined, the storage location of the data in the object can be determined. Optionally, the mode descriptor further includes: a write lock identification field and an object release field. The write lock identification field is used to indicate whether the data is added with the identification of the write operation. The object deallocation field is used to indicate whether the object has been deallocated by another thread.

By using the description information to indicate the organization form of the object, and maintaining the mapping relationship between the description information and the identifier in the network card, the network card can support one or more applications using objects of different organizational forms, and the adaptability of the network card is improved, and It helps to flexibly adjust the format of the verification information according to the business attributes of the service node, so that the network card can support different business models.

It should be noted that this step 301 is an optional step, and it may be determined whether or not this step 301 needs to be executed according to application requirements. For example, when the multiple verification information to be verified by the network card is organized in different forms in the object, step 301 needs to be executed. When the multiple types of verification information to be verified by the network card are in the same organization form in the object, step 301 does not need to be performed.

Step 302: The network card of the destination end receives the read request sent by the requesting end.

When the requester needs to read data from the destination, the requester needs to send a read request to the destination. The read request is used to indicate the data that the requester needs to read from the destination. The read request carries the logical address of the data requested to be read, so that the data requested to be read is indicated by the logical address. Furthermore, since the destination stores data in the memory in units of objects, the logical address is used to indicate the object including the data requested to be read (the object including the data requested to be read is also called a read object).

In addition, the read request also carries the description information of the object or the identifier of the description information, so that the destination terminal can obtain the verification information in the object according to the description information or the identifier. The requesting end can determine the description information or the identifier of the description information according to its requirement of reading data, and carry the description information or the identifier of the description information in the read request. In an implementation manner, the requester and the destination can pre-agree on the organization form of the verification information of the data accessed by different threads in the object. When a thread on the requester needs to read data from the destination, the requester can Determine the organizational form of the object that the thread needs to read according to the result of the protocol, and carry the identifier of the description information describing the organizational form in the read request, so that the destination can determine the verification information of the data according to the identifier. The read request may determine whether to carry the identifier of the description information of the object according to the application requirements.

At the same time, in order to be different from ordinary read requests, the read request may also carry a check indication identifier, which is used to instruct the network card in the destination to perform consistency check on the data read by the read request request .

In a possible implementation manner, the implementation manner of the identification of the description information and the verification indication identification carried in the read request includes: adding an operation code (operation code) in the message area corresponding to the InfiniBand (IB) transport layer in the read request , OPCode), instruct the network card in the destination end to perform consistency check on the data read by the read request through the opcode, and add an identifier of the description information before the payload of the field corresponding to the IB transport layer. Among them, IB is a computer network communication standard for high-performance computing. It has extremely high throughput and extremely low latency. It is used for data interconnection between computers and can carry RDMA technology. IB also uses It can be used as direct or switched interconnection between servers and storage systems, as well as interconnection between storage systems.

In another achievable manner, the implementation manner in which the read request carries the identifier of the description information and the verification indication identifier includes: using an existing reserved field in the read request, the identifier of the description information and the verification indication identifier are both Embedded in a reserved field in the header of a read request.

Similarly, for the implementation manner of the read request carrying the description information, reference may be made to the implementation manner of the read request carrying the identifier of the description information, which will not be repeated here.

Step 303: The network card of the destination acquires the object from the memory of the destination according to the read request.

After receiving the read request, the network card of the destination end can obtain the object from the memory of the destination end according to the logical address carried by the read request. The object carries the data requested by the requester to be read from the destination and the verification information of the data.

Step 304 , the network card of the destination terminal obtains the description information of the object.

In an implementation manner, when the read request carries the identifier of the description information, the network card of the destination end obtains the description information of the object, including: the network card of the destination end obtains the identifier of the description information from the read request, and based on the description information The identity query step 301 establishes the mapping relationship between the description information and the identity in the network card, so as to obtain the description information of the object. When the read request carries descriptive information, the destination network card can extract the descriptive information in the read request according to the manner in which the read request carries the descriptive information, so as to obtain the descriptive information of the object.

Step 305 , the network card of the destination terminal obtains the verification information from the object based on the description information.

Since the description information is used to indicate the organization form of the object, after obtaining the description information of the object, the network card of the destination end can determine the storage location of the verification information in the object according to the organization form indicated by the description information. Moreover, since the network card of the destination end has acquired the object from the memory, the network card of the destination end can know the storage location of the object, and correspondingly, the network card of the destination end can obtain the verification information from the object in the storage location of the object.

It should be noted that when the verification information of the data involved in the read and write operations supported by the network card is organized in the same form in the object, for example, in some fixed business scenarios, the organization forms of the objects involved in the read and write operations are the same. Similarly, the read request may not carry the identifier of the description information of the object. At this time, the organization form of the object can be negotiated between the requesting end and the destination end in advance, and the network card of the destination end can obtain the verification information from the object according to the pre-negotiated organization form after receiving the read request. Correspondingly, the above step 304 does not need to be performed.

Step 306: The network card of the destination end performs consistency check on the data according to the check information.

The implementation of the consistency check on the data may include: judging whether the check information of multiple data blocks in the object is all the same, and when the check information of the multiple data blocks is all the same, determining that the data passes the consistency check , then step 307 is performed, and when the verification information of the multiple data blocks is not all the same, it is determined that the data does not pass the consistency check, and step 308 is performed.

Optionally, the verification information of the data block includes the version number of the data block, and performing consistency verification on the data according to the verification information can be realized by judging whether the version numbers of the multiple data blocks of the data are all the same. Among them, each time an operation is performed on the data, a version number is generated. Different version numbers are used to identify different operations performed on the data. When the version numbers of any two data blocks in the object are different, it indicates that the operations performed on the two data blocks are different, and it can be determined that the data cannot pass the consistency check. In addition, when judging whether all the multiple verification information are the same, any one of the multiple verification information can be used as a reference, and the other verification information in the multiple verification information can be compared with the referenced one. The verification information is compared, which is not specifically limited in this embodiment of the present application.

Step 307: When the consistency check is passed, the network card of the destination end provides data to the memory of the requesting end.

When the consistency check passes, it means that there is no concurrent conflict and no inconsistency in the process of reading data, and the destination can provide the data in the object to the requester.

Since the data is stored in the object, and the description information is used to describe the organization form of the object, the network card at the destination end can obtain the data from the object according to the description information, and provide the data to the requester. In an implementation manner, the implementation manner of the destination end providing the data to the requesting end includes: the network card of the destination end performs a verification information stripping operation on the object, obtains the data in the object, and provides the data to the memory of the requesting end.

Optionally, because in the object, the verification information of any data block can be located at the head or tail of any data block, or the verification information in multiple data blocks in the object can also be placed in other ways, so The check information indicated by the information and the placement position of the data block in the object can be described, and an operation is performed on the data block to obtain the data block. In an implementation manner, when the check information of any data block is located at the head of any data block, since the data block is located behind the check information, the data block can be left-biased according to the description information. The operation of shifting makes all bytes from the first byte of the object after the offset belong to the payload of the object, and the data after stripping the check information from the object is obtained. For example, when the object includes 4 data blocks, the check information of the data block is located at the head of the data block, and the description information of the object indicates that the valid length of the check information of each data block is 10 bytes, then the The 4 data blocks are all shifted to the left by 10 bytes to obtain the data after stripping the header from the object. When the parity information and the data block are placed in the object in other ways, the parity information stripping operation can be performed according to similar operation logic to obtain the data block.

Wherein, providing data by the network card of the destination end to the memory of the requesting end includes: the network card of the destination end sending the data to the network card of the requesting end, and sending the data to the memory of the requesting end through the network card of the requesting end. Wherein, the network card of the requesting end may use a direct memory access (direct memory access, DMA) method to write the data into the memory of the requesting end. In addition, before the network card of the destination end sends the data to the network card of the requesting end, it is also necessary to encapsulate the relevant header of the data, so as to send the data to the network card of the requesting end by using a message.

It should be noted that the above steps 304 to 307 can be executed in a streaming manner, that is, after some data blocks in the object are read from the memory of the destination end to the network card of the destination end, the above steps 304 to 307 can be executed. By performing the above steps 304 to 307 in a streaming manner, there is no need to wait until all data blocks in the object are read into the network card of the destination end, and then perform the data consistency check operation. In addition, when any data block passes the verification, the stripping operation can be performed on the data block, and then the data block obtained by the stripping operation can be sent to the requester, and it is not necessary to wait until all data blocks in the object have been verified for consistency. After the verification operation, all the data blocks in the object are sent to the requester together. At the same time, when the streaming method is used to point to the above steps 304 to 307, it can be seen that the destination end may perform a consistency check operation on another part of the data while sending a part of the data to the requesting end. It can be seen from this that, by stream execution of the above steps 304 to 307, the concurrency of the memory of the requester is improved, and the storage space of the network card can be effectively utilized.

Among them, when the data consistency check is performed in a streaming manner, it is necessary to wait until at least two data blocks are read to the network card of the destination end, and then start the operation of the consistency check. And the implementation method of performing consistency check on the data in a streaming manner is: judging whether the check information of the at least two data blocks read are all the same, and when the check information of the at least two data blocks are all the same, It is determined that the at least two data blocks pass the consistency check, and when the check information of the at least two data blocks is not all the same, it is determined that the at least two data blocks do not pass the consistency check.

Step 308: When the consistency check fails, the network card of the destination informs the requester that the check fails, so that the requester stops writing the received part of the data into the memory of the requester, and the network card of the destination stops checking the data. The unverified part is checked for consistency, and the part that has passed the verification of the data is stopped to be sent to the requester.

When the consistency check fails, it means that there is a concurrency conflict in the process of reading data, which will lead to inconsistency, and the data in the object cannot be provided to the requester. At this time, the network card of the destination can notify the requester that the verification fails, so that the requester stops writing the received part of the data into the memory of the requester. In addition, when the network card of the requesting end determines that the consistency check fails, it can also notify the application requesting to read the data to perform conflict resolution processing, such as informing the application to roll back and retry or terminate the transaction. By timely notifying the requester that the consistency check fails, the overhead caused by processing invalid data can be effectively reduced.

Among them, the implementation manner of the network card of the requesting end notifying the application program includes: the network card constructs a message for indicating the failure of the consistency check, and puts the message into a queue that communicates with the application program, and the application program obtains the message from the queue, according to This message informs that the consistency check failed.

In an implementation manner, the implementation manner of the destination end notifying the requesting end includes: replying to the requesting end a read response indicating a failure of the consistency check. Optionally, a flag bit can be added to the read response, and the consistency check failure can be identified by the value of the flag bit. Alternatively, an opcode can be added to the read response, and the opcode can be used to identify a consistency check failure. For example, the read response can be an RDMA read response ABORT message.

In addition, when the above steps 303 to 306 are performed in a streaming manner, when it is determined that the consistency check fails, the network card of the destination end may also stop performing consistency check on the unchecked part of the data, and stop sending the data to the requesting end. The part of the sent data that has passed the verification. In this way, it is possible to reduce the resource overhead and time overhead used in the consistency check and transmission of invalid data, improve the utilization rate of resources and ensure the concurrency of the destination end.

It should be noted that, the above steps 301 to 308 may be implemented by functional modules. For example, assuming that the function of the network card in the destination is implemented by the function module shown in Figure 2, then as shown in Figure 6, step 301 can be implemented by the initialization module X011 in Figure 2, step 302 can be implemented by the network card module X021, The first cache module X022 is implemented, and the data read from the memory can be stored in the first cache module X022, step 304 is implemented by the matching module X023, and steps 305 and 306 are implemented by the verification module X024. The verification module X024 sends notification information to the stripping module X025, so that the stripping module X025 performs the verification information stripping operation in step 307 on the objects that have been verified cached in the first cache module X022, and after the stripping operation is completed, the network card Module X021 reads data from the first cache module X022, and performs the operation of sending data to the network card of the requester in step 307. In step 308, the requester is notified that the verification is not realized by the network card module X021. Consistency verification of the unverified part is implemented by the verification module X024, and in step 308, the part that has passed the verification in the data that has passed the verification is stopped by the network card module X021. In addition, in order to distinguish the verified object from the unverified object, the unverified object can be stored in the first cache module X022, the verification information of the verified object can be stored in the second cache module X026, and the After the requester sends the data in the object, the object is deleted from the first cache module X022.

In the concurrency control method provided by the embodiment of the present application, by using the network card to perform consistency check and perform the peel operation, it is unnecessary to use the CPU in the server to perform the consistency check and perform the peel operation, and the concurrency of the reading process can be realized. On the basis of control, the CPU overhead due to checksum memory transfer is effectively reduced, and the end-to-end delay is reduced. Moreover, in the process of concurrency control, the consistency check and the stripping operation are implemented by software, which reduces the dependence of the concurrency control on the hardware, and correspondingly reduces the implementation cost of the concurrency control.

FIG. 7 is a flowchart of a method executed by a requester in a concurrency control method provided by an embodiment of the present application. As shown in Figure 7, the concurrency control method includes the following steps:

Step 701: The application program of the requesting end registers the description information of the object.

For the implementation process of this step 701, please refer to the implementation process of registering the description information of the object by the application program of the destination end in step 301, which will not be repeated here. The difference between step 701 and step 301 is that after the registration of the description information is completed, the application program of the requesting end does not need to send the identifier of the description information to the destination end.

Step 702: A read request sent by the network card of the requesting end to the destination end.

For the implementation process of step 702, please refer to the implementation process of the requesting end sending the read request in step 302, which will not be repeated here. Wherein, whether to carry the identifier of the description information of the object may also be determined in the read request according to application requirements.

Step 703: The network card of the destination acquires the object from the memory of the destination according to the read request, and sends a read response to the requester.

The read response carries the identifier of the object and the description information of the object. The object includes the data requested by the requester to be read from the destination and the verification information of the data. In addition, the identifier of the description information can be transmitted from the requesting end to the destination end through the read request and read response in the form of an accompanying signal, and then transmitted from the destination end to the requesting end, so that the destination end does not need to perceive the identifier of the description information. Wherein, for the implementation process of acquiring the object by the destination end, please refer to the relevant description in step 303 accordingly, which will not be repeated here.

Step 704: The network card of the requesting end obtains the description information of the object according to the read response.

In an implementation manner, since the read response carries the identifier of the description information, the network card of the requesting end obtains the description information of the object, including: the network card of the requesting end obtains the identifier of the description information from the read response, and based on the identifier of the description information The mapping relationship between the description information and the identifier established in the network card in step 701 is queried to obtain the description information of the object.

Step 705: Based on the description information, the network card of the requester obtains the verification information from the object.

Since the description information is used to indicate the storage location of the verification information in the object, after obtaining the description information of the object, the network card of the requesting end can obtain the data of the data read by the read request request in the object according to the storage location indicated by the description information. Check information.

It should be noted that when the verification information of the data involved in the read and write operations supported by the network card is organized in the same form in the object, for example, in some fixed business scenarios, the organization forms of the objects involved in the read and write operations are the same. Similarly, the read request may not carry the identifier of the description information of the object. At this time, the organization form of the object can be negotiated between the requesting end and the destination end in advance, and the network card of the requesting end can obtain the verification information from the object according to the pre-negotiated organization form after receiving the read response. Correspondingly, the above step 704 does not need to be performed.

Step 706: The network card of the requesting end performs consistency check on the data according to the check information.

For the implementation process of this step 706, please refer to the implementation process of step 306, which will not be repeated here.

Step 707: When the consistency check is passed, the network card of the requester sends data to the memory of the requester.

Wherein, the network card of the requesting end sends data to the memory of the requesting end, which may be that the network card of the requesting end directly sends the obtained data to the memory of the requesting end. For the implementation process of this step 707, please refer to the corresponding description in the step 307, which will not be repeated here.

Step 708: When the consistency check fails, the network card of the requester stops writing the part of the data that has passed the check into the memory of the requester, and the network card of the requester stops performing consistency check on the unverified part of the data. .

For the implementation process of step 708, please refer to the relevant description in step 308 accordingly, and details are not repeated here.

Similarly, at least some of the above steps 701 to 708 may be implemented by functional modules. For example, it is assumed that the function of the network card in the requester is realized by the function module shown in Fig. 2, as shown in Fig. 8, step 701 can be realized by the initialization module X011 in Fig. 2, step 702 can be realized by the network card module X021, step 704 can be realized by matching The module X023 is implemented, and the steps 705 and 706 are implemented by the verification module X024. When the verification is passed, the verification module X024 sends notification information to the stripping module X025, so that the stripping module X025 can verify the cached in the first cache module X022. The object performs the verification information stripping operation in step 707, and after the stripping operation is completed, the first cache module X022 performs the operation of sending data to the memory of the requester in step 707, and in step 708, stops the part of the data that has passed the verification Writing to the memory of the requesting end is implemented by the first cache module X022, and in step 708, stopping the consistency verification of the unverified part of the data is implemented by the verification module X024. In addition, in order to distinguish the verified object from the unverified object, the unverified object can also be stored in the first cache module X022, and the verification information of the verified object can be stored in the second cache module X026, and stored in the second cache module X026. After the data in the object is sent to the memory of the requester, the object is deleted from the first cache module X022. Wherein, after receiving the read response sent by the destination terminal, the network card module X021 can obtain the object from the read response, and store the object in the first cache module X022.

To sum up, in the concurrency control method provided by the embodiments of the present application, by using the network card to perform consistency check and perform the peel operation, it is unnecessary to use the CPU in the server to perform the consistency check and perform the peel operation, which can achieve Based on the concurrency control of the reading process, the CPU overhead due to checksum memory transfer is effectively reduced, and the end-to-end delay is reduced. Moreover, in the concurrency control process, the functions implemented by the network card can be implemented by software, thus reducing the concurrency control dependence on hardware and correspondingly reducing the implementation cost of the concurrency control.

It should be noted that, the sequence of steps of the concurrency control method provided by the embodiments of the present application may be appropriately adjusted, and the steps may be correspondingly increased or decreased according to the situation. Any person skilled in the art who is familiar with the technical scope disclosed in the present application can easily think of any variation of the method, which should be covered by the protection scope of the present application, and thus will not be repeated here.

The embodiment of the present application also provides a remote access device. As shown in Figure 9, the network card 90 includes:

The obtaining unit 901 is configured to obtain an object, where the object carries the data requested by the requesting end to be read from the destination end and the verification information of the data.

The verification unit 902 is configured to perform consistency verification on the data according to the verification information.

The interaction unit 903 is configured to provide data to the memory of the requester when the consistency check is passed.

Optionally, as shown in FIG. 10, the network card 90 further includes: a stripping unit 904, configured to perform a stripping operation of verification information on the object to obtain data when the consistency check is passed.

Correspondingly, the interaction unit 903 is specifically configured to provide data to the memory of the requester.

Optionally, the network card is deployed in the destination, and the interaction unit 903 is further configured to notify the requester that the verification fails when the consistency check fails.

Optionally, the network card is deployed in the destination end, and the verification unit 902 is further configured to stop the consistency verification of the unverified part of the data when the consistency verification fails. The interaction unit 903 is further configured to stop sending the part of the data that has passed the check to the requesting end when the consistency check fails.

Optionally, the network card is deployed in the requester, and the interaction unit 903 is further configured to stop writing the part of the data that has passed the check into the memory of the requester when the consistency check fails.

Optionally, the network card is deployed in the requester, and the verification unit 902 is further configured to stop the consistency verification of the unverified part of the data when the consistency verification fails.

Optionally, the interaction unit 903 is further configured to receive a read request sent by the requester, where the read request is used to request to read data. Correspondingly, the obtaining unit 901 is specifically configured to obtain the object from the memory of the destination end according to the read request.

Optionally, the network card is deployed in the destination, and the interaction unit 903 is specifically configured to send data to the network card of the requester, so that the network card of the requester sends data to the memory of the requester.

Optionally, the network card is deployed in the requesting end, and the interaction unit 903 is further configured to send a read request to the destination end, where the read request is used to request to read data. Correspondingly, the obtaining unit 901 is specifically configured to receive the object sent by the network card of the destination end, and the object is obtained by the network card of the destination end in the destination end based on the read request.

Optionally, the network card is deployed in the requester, and the interaction unit 903 is specifically configured to directly send data to the memory of the requester.

Optionally, the obtaining unit 901 is further configured to obtain description information of the object, where the description information is used to indicate the organization form of the object. Correspondingly, the verification unit 902 is further configured to obtain verification information from the object based on the description information.

Optionally, the obtaining unit 901 is specifically configured to: obtain the identifier of the description information; and obtain the description information based on the identifier of the description information and the mapping relationship, and the mapping relationship reflects the corresponding relationship between the description information and the identifier.

Optionally, the network card is deployed in the destination end, and the interaction unit 903 is further configured to receive a read request sent by the requesting end, where the read request carries the identifier of the description information. Correspondingly, the obtaining unit 901 is specifically configured to obtain the identifier of the description information from the read request.

Optionally, the network card is deployed in the requester, and the interaction unit 903 is further configured to receive a read response sent by the destination based on the read request. The read response carries the identifier of the description information, and the identifier of the description information is sent to the destination through the read request. Correspondingly, the obtaining unit 901 is specifically configured to obtain the identifier of the description information from the read response.

To sum up, in the network card provided in the embodiment of the present application, by using the verification unit in the network card to perform consistency check, and the peeling unit to perform the peeling operation, it is unnecessary to use the CPU in the server to perform consistency check and perform peeling. operation, on the basis of realizing the concurrency control of the reading process, the CPU overhead due to checksum memory transfer can be effectively reduced, and the end-to-end delay can be reduced. In addition, the functions of some or all of the units in the network card can be implemented by software, so that the dependence of the concurrency control on the hardware is reduced, and the implementation cost of the concurrency control is correspondingly reduced.

It should be noted that the above division of units in the network card is only an example, and is not used to limit the implementation of the network card provided by the embodiments of the present application. The method of dividing the functional modules in the network card in the embodiments of the present application may be performed according to application requirements. Adjustment. For example, the network card can be further subdivided. Please refer to the division method of the programmable network card in FIG. 2 for the division method. As shown in FIG. 2, the functions implemented by the network card module X021 refer to the functions implemented by the interaction unit in the foregoing The functions implemented by the verification module X024 refer to the functions implemented by the verification unit in the previous embodiment, the functions implemented by the stripping module X025 refer to the functions implemented by the stripping units in the previous embodiments, and the functions implemented by the matching module X023 refer to the acquisition units in the previous embodiments to obtain descriptions For the function of the information, for the function implemented by the first cache module X022, refer to the function of the acquisition unit acquiring the object in the foregoing embodiment.

Those skilled in the art can clearly understand that, for the convenience and brevity of description, for the specific working process of the devices, units and modules described above, reference may be made to the corresponding content in the foregoing method embodiments, which will not be repeated here.

The embodiments of the present application provide a computer device. FIG. 11 exemplarily provides a possible architectural diagram of a computer device. As shown in FIG. 11 , the computer device 110 may include a network card 1101 , a memory 1102 , a communication interface 1103 and a bus 1104 . In a computer device, the number of network cards 1101 may be one or more, and FIG. 11 only illustrates one of the network cards 1101 . Optionally, as shown in FIG. 11 , the computer device 110 may further include a processor 1105, and the processor 1105 may be a central processing unit (Central Processing Unit, CPU). If the computer device has multiple processors, the multiple processors may be of different types or may be the same. Optionally, multiple processors of the computer device may also be integrated into a multi-core processor.

The network card 1101 is used for data interaction with other computer devices. In this embodiment of the present application, the process of the consistency check may be performed by the network card 1101 . Optionally, the network card 1101 is further configured to strip the verification information of the data blocks in the object, obtain the data requested by the requester to read, and provide the data to the memory of the requester. In addition, the network card 1001 can support various read and write operations of different read and write scales, and the data and verification information of the read and write operations of the various read and write scales are organized in different forms in the object.

In an implementation manner, the network card 1001 performs operations such as checksum stripping and the like, which may be implemented by program codes. Correspondingly, the network card 1001 for performing the consistency check may be a programmable network card. When the network card 1001 is a programmable network card, the behavior of the network card 1001 can be changed by modifying program codes such as firmware or microcode in the programmable network card, so that the network card 1001 has the above functions. Moreover, in order to further reduce the time delay of the CPU, the network card 1001 is configured with a remote direct memory access function.

The memory 1102 stores computer instructions and data, and the memory 1102 may store computer instructions and data required to implement the concurrency control method provided by the present application. The memory 1102 can be any one or any combination of the following storage media: non-volatile memory (such as read-only memory (Read-Only Memory, ROM), solid state disk (Solid State Disk, SSD), hard disk (Hard Disk) Drive, HDD), optical disc, etc., volatile memory.

The communication interface 1103 may be any one or any combination of the following devices: a network interface (eg, an Ethernet interface) having a network access function.

The communication interface 1103 is used for data communication between the computer device and other nodes or other computer devices.

FIG. 11 also exemplarily depicts bus 1104 . The bus 1104 can connect the network card 1101 with the memory 1102 and the communication interface 1103 . In this way, through the bus 1104, the network card 1101 can access the memory 1102, and can also use the communication interface 1103 to perform data interaction with other nodes or other computer devices.

In the present application, the computer program is stored in the memory of the computer device, and the network card executes the computer program in the memory 1102 to implement the concurrency control method provided by the present application.

Embodiments of the present application further provide a storage medium, which is a non-transitory computer-readable storage medium. When an instruction in the storage medium is executed by a network card, the network card executes the concurrency control method provided by the embodiments of the present application.

The embodiments of the present application also provide a computer program product including instructions, when the computer program product runs on the network card, the network card is made to execute the concurrency control method provided by the embodiments of the present application.

Those of ordinary skill in the art can understand that all or part of the steps of implementing the above embodiments can be completed by hardware, or can be completed by instructing relevant hardware through a program, and the program can be stored in a computer-readable storage medium. The storage medium mentioned may be a read-only memory, a magnetic disk or an optical disk, etc.

In the embodiments of the present application, the terms "first", "second" and "third" are only used for description purposes, and cannot be understood as indicating or implying relative importance. The term "at least one" refers to one or more, and the term "plurality" refers to two or more, unless expressly limited otherwise.

The term "and/or" in this application is only an association relationship to describe associated objects, indicating that there can be three kinds of relationships, for example, A and/or B, it can mean that A exists alone, A and B exist at the same time, independently There are three cases of B. In addition, the character "/" in this document generally indicates that the related objects are an "or" relationship.

The above are only optional embodiments of the present application, and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the concept and principles of the present application shall be included in the protection of the present application. within the range.

Claims (24)

1. A concurrency control method, characterized in that the method comprises:

   The network card obtains an object, and the object carries the data requested by the requester to be read from the destination and the verification information of the data;

   The network card performs consistency check on the data according to the check information;

   When the consistency check is passed, the network card provides the data to the memory of the requester.
2. The method according to claim 1, wherein when the consistency check is passed, the network card provides the data to the memory of the requester, comprising:

   When the consistency check is passed, the network card performs a verification information stripping operation on the object to obtain the data;

   The network card provides the data to the memory of the requester.
3. The method according to claim 1 or 2, wherein the network card is deployed in the destination terminal, and after the network card performs consistency check on the data according to the verification information, the method Also includes:

   When the consistency check fails, the network card informs the requester that the check fails.
4. The method according to any one of claims 1 to 3, wherein the network card is deployed in the destination terminal, and after the network card performs consistency check on the data according to the verification information, the The method also includes:

   When the consistency check fails, the network card stops performing consistency check on the unchecked part of the data, and stops sending the data that has passed the check to the requesting end.
5. The method according to claim 1 or 2, wherein the network card is deployed in the requester, and after the network card performs consistency check on the data according to the verification information, the method Also includes:

   When the consistency check fails, the network card stops writing the part of the data that has passed the check into the memory of the requester.
6. The method according to claim 1, 2 or 5, wherein the network card is deployed in the requester, and after the network card performs consistency check on the data according to the verification information, the The method also includes:

   When the consistency check fails, the network card stops performing consistency check on the unchecked part of the data.
7. The method according to any one of claims 1 to 4, wherein the network card is deployed in the destination terminal, and the network card obtains an object, comprising:

   The network card receives a read request sent by the requester, and the read request is used to request to read the data;

   The network card acquires the object from the memory of the destination according to the read request.
8. The method according to any one of claims 1 to 4 or 7, wherein the network card is deployed in the destination, and the network card provides the data to the memory of the requester, comprising:

   The network card sends the data to the network card of the requester, so that the network card of the requester sends the data to the memory of the requester.
9. The method according to claim 1, 2, 5 or 6, wherein the network card is deployed in the requester, and before the network card acquires the object, the method further comprises:

   The network card sends a read request to the destination, and the read request is used to request to read the data;

   The network card acquisition object includes:

   The network card receives the object sent by the network card of the destination end, and the object is obtained by the network card of the destination end in the destination end based on the read request.
10. The method according to claim 1, 2, 5, 6 or 9, wherein the network card is deployed in the requester, and the network card provides the data to the memory of the requester, comprising:

    The network card directly sends the data to the memory of the requester.
11. The method according to any one of claims 1 to 10, wherein before the network card performs consistency check on the data according to the check information, the method further comprises:

    obtaining, by the network card, description information of the object, where the description information is used to indicate the organization form of the object;

    The network card acquires the verification information from the object based on the description information.
12. A network card, characterized in that the network card comprises:

    an acquisition unit, configured to acquire an object, the object carrying the data requested by the requester to be read from the destination and the verification information of the data;

    a verification unit, configured to perform consistency verification on the data according to the verification information;

    The interaction unit is configured to provide the data to the memory of the requester when the consistency check is passed.
13. The network card according to claim 12, wherein the network card further comprises:

    a stripping unit, configured to perform a verification information stripping operation on the object to obtain the data when the consistency check is passed;

    The interaction unit is specifically configured to provide the data to the memory of the requester.
14. The network card according to claim 12 or 13, wherein the network card is deployed in the destination terminal, and the interaction unit is further configured to notify the requesting terminal when the consistency check fails The verification fails.
15. The network card according to any one of claims 12 to 14, wherein the network card is deployed in the destination terminal,

    The verification unit is further configured to stop performing consistency verification on the unverified part of the data when the consistency verification fails;

    The interaction unit is further configured to stop sending the part of the data that has passed the check to the requesting end when the consistency check fails.
16. The network card according to claim 12 or 13, wherein the network card is deployed in the requester, and the interaction unit is further configured to stop storing the data in the data when the consistency check fails. The part that has passed the verification is written into the memory of the requester.
17. The network card according to claim 12, 13 or 16, wherein the network card is deployed in the requester, and the verification unit is further configured to stop checking the The unverified part of the data is checked for consistency.
18. The network card according to any one of claims 12 to 15, wherein,

    The interaction unit is further configured to receive a read request sent by the requester, where the read request is used to request to read the data;

    The obtaining unit is specifically configured to obtain the object from the memory of the destination terminal according to the read request.
19. The network card according to any one of claims 12 to 15 or 18, wherein the network card is deployed in the destination end, and the interaction unit is specifically configured to send the data to the network card of the requesting end, so that The network card of the requester sends the data to the memory of the requester.
20. The network card according to claim 12, 13, 16 or 17, wherein the network card is deployed in the requester, and the interaction unit is further configured to send a read request to the destination, and the read request for requesting to read said data;

    The obtaining unit is specifically configured to receive the object sent by the network card of the destination end, where the object is obtained by the network card of the destination end in the destination end based on the read request.
21. The network card according to claim 12, 13, 16, 17 or 20, wherein the network card is deployed in the requester, and the interaction unit is specifically configured to directly send the requester to the memory of the requester. data.
22. The network card according to any one of claims 12 to 21, wherein,

    The obtaining unit is further configured to obtain description information of the object, where the description information is used to indicate the organization form of the object;

    The verification unit is further configured to acquire the verification information from the object based on the description information.
23. A computer device, characterized in that the computer device comprises: a network card and a memory, wherein a computer program is stored in the memory; when the network card executes the computer program, the method according to any one of claims 1 to 11 is implemented.
24. A non-transitory computer-readable storage medium, characterized in that, when an instruction in the computer-readable storage medium is executed by a network card, the network card executes the method of any one of claims 1 to 11.

Images