Hot standby system and method, electronic device, and storage medium

Abstract

The present application provides a hot standby system and method, an electronic device, and a storage medium. The system comprises: accelerator cards, which are connected to master servers in one-to-one correspondence and are in shared connection with a standby server, and each are used for acquiring operation status information, operation log information, memory information, and configuration information of the correspondingly connected master server, when it is determined, on the basis of the operation status information, that the correspondingly connected master server has gone down, acquiring data to be processed that is to be transmitted to the correspondingly connected master server, and transmitting the operation log information, the memory information, the configuration information, and said data to the standby server; and the standby server, which is used for, on the basis of the operation log information, the memory information, and the configuration information, simulating and generating a virtual application environment for the correspondingly connected master server, and on the basis of the virtual application environment and said data, executing an application service of the correspondingly connected master server.

Description

Hot backup system, method, electronic device and storage medium

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to the Chinese patent application filed with the China Patent Office on February 29, 2024, with application number 202410227354.3, and application name “Hot backup system, method, electronic device and storage medium”, all contents of which are incorporated by reference into this application. Technical Field

[0003] The present application relates to a hot backup system, method, electronic equipment and storage medium. Background Art

[0004] In today's information age, server software applications have become the core of enterprise information systems, and the fault tolerance and continuity of critical business systems are particularly important. Failures or data loss in critical server software can lead to business interruptions, data loss, and other serious consequences, impacting the company's reputation and causing significant financial losses. Therefore, efficient and convenient hot backup has become a key technology for ensuring data security and business continuity.

[0005] In the related art, in a multi-server system, it is usually necessary to set up at least one backup server for each main server, so that each group of main and backup servers can store the memory information of the main server in a disk array or hard disk storage area through network sharing through dual-machine hot standby to perform data hot backup; and the inventors realized that this backup method not only requires the configuration of a large number of redundant backup servers, but also requires the storage and transmission of memory information through network sharing, which will introduce higher transmission delays and limited bandwidth, thereby leading to higher hardware deployment and maintenance costs and low master-slave switching efficiency.

[0006] Summary of the Invention

[0007] According to an embodiment of the present application, in a first aspect, a hot backup method is provided, comprising: an accelerator card obtaining operation status information, operation log information, and configuration information of a correspondingly connected master server from a correspondingly connected master server of the accelerator card, mirroring memory information of the correspondingly connected master server in a memory of the correspondingly connected master server to a computing expansion connection memory of the accelerator card through a computing expansion connection interface of the accelerator card, and judging whether the correspondingly connected master server has crashed based on the operation status information; if the correspondingly connected master server has crashed, the correspondingly connected master server that has crashed is used as a crashed master server, obtaining data to be processed to be transmitted to the crashed master server, and transmitting the operation log information, memory information, configuration information, and data to be processed of the crashed master server to a backup server; and

[0008] The standby server simulates and generates a virtual application environment of the downtime main server based on the operation log information, memory information and configuration information of the downtime main server, and executes the application service of the downtime main server based on the virtual application environment and the data to be processed.

[0009] According to an embodiment of the present application, in a second aspect, a hot backup system is further provided, comprising: a plurality of accelerator cards, a plurality of master servers and a backup server;

[0010] The accelerator card is connected to the primary server in a one-to-one correspondence and is also connected to the backup server;

[0011] The accelerator card includes a computing expansion connection interface and a computing expansion connection memory;

[0012] an accelerator card, configured to obtain operating status information, operating log information, and configuration information of the corresponding connected primary server from the corresponding connected primary server, mirror the memory information of the corresponding connected primary server in the memory of the corresponding connected primary server to the computing extension connection memory through the computing extension connection interface, and determine whether the corresponding connected primary server has crashed based on the operating status information. If the corresponding connected primary server has crashed, the corresponding connected primary server that has crashed is designated as the crashed primary server, obtain pending data to be transmitted to the crashed primary server, and transmit the operating log information, memory information, configuration information, and pending data of the crashed primary server to the standby server; and

[0013] The standby server is used to simulate and generate a virtual application environment of the downtime main server based on the operation log information, memory information and configuration information of the downtime main server, and execute the application service of the downtime main server based on the virtual application environment and the data to be processed.

[0014] According to an embodiment of the present application, in a third aspect, an electronic device is also provided, comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, wherein when the processor executes the program, the hot backup method of the first aspect described above is implemented.

[0015] According to an embodiment of the present application, in a fourth aspect, a non-transitory computer-readable storage medium is further provided, on which computer-readable instructions are stored. When the computer-readable instructions are executed by a processor, the hot backup method of the first aspect as described above is implemented.

[0016] According to an embodiment of the present application, in a fifth aspect, a computer program product is further provided, comprising computer-readable instructions, which, when executed by a processor, implement the hot backup method of the first aspect described above.

[0017] The details of one or more embodiments of the present application are set forth in the accompanying drawings and the description below. Other features and advantages of the present application will become apparent from the description, drawings, and claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in this application or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] FIG1 is a schematic structural diagram of a dual-machine hot standby system provided by the related art;

[0020] FIG2 is a schematic diagram of the structure of a hot backup system provided in an embodiment of the present application;

[0021] FIG3 is a schematic diagram of the structure of an accelerator card provided in an embodiment of the present application;

[0022] FIG4 is a flowchart of the work flow of the accelerator card provided in an embodiment of the present application;

[0023] FIG5 is a schematic diagram of a process of data backup management provided by an embodiment of the present application;

[0024] FIG6 is a schematic diagram of a process of master-slave switching provided in an embodiment of the present application;

[0025] 7 is a schematic diagram of a process flow of a standby-active switching step provided in an embodiment of the present application;

[0026] FIG8 is a flow chart of a hot backup method according to an embodiment of the present application;

[0027] FIG9 is a schematic structural diagram of an electronic device provided in an embodiment of the present application;

[0028] FIG10 is a schematic diagram of a non-transitory computer-readable storage medium provided in an embodiment of the present application. DETAILED DESCRIPTION

[0029] To make the objectives, technical solutions, and advantages of this application more clear, the technical solutions of this application will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of this application.

[0030] The terms "first," "second," and the like in this application are used to distinguish similar objects, and are not used to describe a specific order or precedence. It should be understood that such terms are interchangeable where appropriate, so that embodiments of this application can be implemented in an order other than that illustrated or described herein. Furthermore, the terms "first," "second," and the like generally distinguish objects of a class and do not limit the number of objects; for example, the first object can be one or more. Furthermore, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates an "or" relationship between the connected objects.

[0031] In today's information age, server software applications have become the core of enterprise information systems, and the fault tolerance and continuity of critical business systems are particularly important. Failures or data loss in critical server business system software can lead to business interruption, data loss, and other serious consequences, damaging the company's reputation and causing significant economic losses. Therefore, for enterprises that need to ensure information security and provide uninterrupted information services, the fault tolerance and continuity of business systems are particularly important. Ensuring the continuous operation of various critical applications and achieving a virtuous cycle of sustainable operations requires DR (Disaster Recovery) and HB (Hot Backup) to ensure data security and business continuity.

[0032] Disaster recovery refers to a series of measures implemented within server software applications to ensure that the system can be restored to normal operation as quickly as possible after a disaster. The key principle of disaster recovery is to quickly restore server software by backing up critical data and system configuration information, and establishing a backup hardware and software environment.

[0033] Hot backup refers to maintaining data consistency between a primary and backup server in server software applications by synchronizing data and configuration information in real time. If the primary server fails or data is lost, the backup server can immediately take over, ensuring business continuity. Hot backup utilizes real-time data replication and dual-host hot backup technology to synchronize and switch data between the primary and backup servers.

[0034] Hot standby technology is widely used in finance, telecommunications, healthcare, and data centers, where data security and continuity are paramount. It is a mainstream application server fault-tolerance technology, particularly dual-server hot standby. Therefore, related technologies typically rely on dual-server hot standby to achieve data security and business continuity.

[0035] In the related art, there are two implementation modes for hot standby: one is based on a shared storage device, and the other is a method without a shared storage device, generally referred to as a pure software method.

[0036] Figure 1 is a schematic diagram of the structure of a dual-server hot standby system provided by related technologies. As shown in Figure 1, dual-server hot standby based on shared storage is the most standard solution for dual-server hot standby. In this method, two servers 110 communicate with clients 140 via a switch 130 to process business transactions. A shared storage device 120, such as a disk array cabinet or a SAN (Storage Area Network), is used between the two servers 110 for data synchronization and switching. The two servers 110 can adopt various modes, such as mutual backup, master-slave, and parallel operation. During operation, the two servers 110 provide services externally using a virtual IP (Internet Protocol) address. Depending on the operating mode, service requests are sent to one of the servers. Simultaneously, this server monitors the operating status of the other server via a heartbeat line, such as by establishing a private network. If one server fails, the other server makes a judgment based on the heartbeat detection and switches over to take over the service. For users, this process is fully automatic and completes very quickly, without impacting business operations. Since shared storage devices are used, the two servers actually use the same data, which is managed by dual-machine or cluster software.

[0037] For the pure software method, data can be copied to another server in real time through dual-machine software that supports mirroring. In this way, the same data is stored on both servers. If one server fails, it can be switched to the other server in time.

[0038] Therefore, in a multi-server system, it is usually necessary to set up at least one backup server for each main server, and the hot backup server keeps the main and backup servers in the power-on state at all times and maintains configuration synchronization, so that each group of main and backup servers can store the memory information of the main server through network sharing to the disk array or hard disk storage area through dual-machine hot standby, and perform data hot backup to achieve disaster recovery data quickly and avoid data loss. However, the equipment investment is large, the communication cost is high, the communication environment requirements are high, the software maintenance and upgrade, the system hardware upgrade and the daily operation management are relatively complicated, and the storage and transmission of memory information need to be shared through the network, which will introduce high transmission delay and limited bandwidth.

[0039] In summary, the dual-machine hot standby solution in the related art has the disadvantages of high hardware deployment and maintenance costs and low master-slave switching efficiency.

[0040] To address the above-mentioned shortcomings, the present application provides a hot backup system, method, electronic device, and storage medium suitable for systems implementing hot backup for multiple devices simultaneously, and particularly suitable for implementing hot backup functionality when multiple devices are simultaneously operating a single function. This application can reduce the number of backup servers in dual-machine hot backup without compromising fault tolerance, thereby reducing hardware deployment and maintenance costs and increasing the speed of active / standby switching.

[0041] The hot backup system, method, electronic device, and storage medium provided in the embodiments of the present application are described in detail below with reference to the accompanying drawings through specific embodiments and their application scenarios.

[0042] FIG2 is a schematic diagram of the structure of the hot backup system provided by the present application. As shown in FIG2 , the system includes multiple accelerator cards 210, multiple primary servers 220, and one backup server 230. Each accelerator card 210 is connected to each primary server 220 in a one-to-one correspondence, and is also connected to the backup server 230.

[0043] The number of acceleration cards here can be adaptively set according to the number of main servers, such as being the same as the number of main servers, or a preset multiple of the number of main servers; the acceleration card can be a PCIE (Peripheral Component Interconnect Express, high-speed serial computer expansion bus) accelerator (hereinafter also referred to as PCIE smart network card).

[0044] Each accelerator card can be connected to a single primary server, enabling backup sharing between each accelerator card and the corresponding primary server. Each accelerator card can be a small system device plugged into a backup server. Therefore, the power supply and main body of the accelerator card are installed on the backup server. Therefore, a failure or power outage on the primary server will not affect the accelerator card on the backup server. The accelerator card can utilize its intelligent media function as shared storage, triggering the backup server to start taking over the host's services. This system management method maximizes the effectiveness of the backup server.

[0045] The backup server can be switched between different acceleration cards to quickly respond to and repair different faulty main servers through the backup and sharing information between different acceleration cards and the main server, and switch back to the corresponding main server when the faulty main server returns to normal. Therefore, the main and backup servers can have the same or different configurations. It only needs to cover all the configurations of the main server and can be used to adapt to the same system environment as the main server. For example, the backup server here can have a system environment configuration with higher specifications than all the main servers, so as to provide hot standby services for multiple main servers at the same time.

[0046] It should be noted that since the probability of multiple master servers failing at the same time is almost the same as the probability of master and backup servers failing at the same time, the 1-to-many hot backup system provided by this embodiment will not reduce the fault tolerance of the dual-machine hot backup solution. In particular, when the backup server adopts a higher-specification server that supports single-machine fault tolerance technology, the 1-to-N hot backup system provided by this embodiment, that is, a hot backup system including one backup server and N master servers, can achieve 1+1+1 / N reliability, and the fault tolerance rate will be greater than that of traditional dual-machine hot backup, and the cost is also lower than that of multiple backup servers. Therefore, the hot backup system provided by this embodiment has a significant technical improvement compared to related technologies, which can not only reduce costs but also increase fault tolerance.

[0047] In some embodiments, each accelerator card includes a first Ethernet interface, a computing expansion connection interface, and a computing expansion connection memory;

[0048] The first Ethernet interface is connected to the network port of the corresponding connected master server and is used to obtain the operation status information, operation log information and configuration information of the corresponding connected master server from the management module of the corresponding connected master server;

[0049] The computing extension connection interface is connected to the memory interface of the corresponding connected master server and is used to obtain the memory information of the corresponding connected master server from the memory of the corresponding connected master server by mirroring;

[0050] Compute extended connection memory, used to store memory information transmitted by each server mirror, and transmit the memory information mirror to each server;

[0051] In some embodiments, each accelerator card further includes a high-speed serial computer expansion bus interface;

[0052] A high-speed serial computer expansion bus interface is connected to the standby server and is used to transmit the operation log information, memory information, configuration information and pending data of the downed main server to the standby server.

[0053] Figure 3 is a schematic diagram of the structure of the accelerator card provided in this application; as shown in Figure 3, the accelerator card 210 includes multiple interfaces, including at least a first Ethernet interface P0, a CXL (Compute Express Link) interface and a PCIE (Peripheral Component Interconnect Express) interface.

[0054] The first Ethernet interface P0 here can be a dual-interface high-speed Ethernet interface; the CXL interface can be a PCIE X16 high-speed cable expansion interface supporting the CXL2.0 protocol; the PCIE interface can be a PCIE EDGE (PCIE edge interface) gold finger interface supporting PCIE5.0 X16.

[0055] Figure 4 is a flowchart of the accelerator card provided in this application. As shown in Figure 4, the first Ethernet interface P0 is connected to the network port of the corresponding connected master server, such as the management network port or a normal network connection, and performs private network communication through software to monitor and obtain the operating status information, operating log information, and configuration information of the corresponding connected master server. The so-called operating status information includes at least heartbeat information; the so-called configuration information includes at least system configuration information and operating environment configuration information.

[0056] CXL is an open industry interconnect standard that provides high-bandwidth, low-latency connections between host servers and devices such as accelerator cards, memory buffers, and intelligent I / O (Input / Output) devices, thereby meeting the requirements of high-performance heterogeneous computing. It maintains consistency between the CPU (Central Processing Unit) memory space and the memory of connected devices.

[0057] The CXL interface can be connected to the memory interface of a corresponding main server with a CXL expansion interface within 3 meters through an external expansion cable. When the corresponding main server is in normal working condition, the accelerator card will have two slots of DDR (Double Data Rate) memory as shadow memory of the main server connected to the CXL interface, keeping data synchronized with the memory of the corresponding main server at all times.

[0058] The PCIE interface connects and communicates with the backup server. When the corresponding primary server fails, the accelerator card quickly communicates with the backup server through the PCIE interface, initiating the transfer of the downed primary server's operational logs, memory information, and configuration information stored locally to the backup server. The backup server then quickly restores the downed primary server's state to enable service switchover. The PCIE interface used here communicates with the backup server via the PCIE bus protocol. Major memory data is transferred to the backup server's memory in the form of large data files. Compared to the CXL bus protocol, the PCIE protocol offers superior latency and bandwidth for transferring large data memory, especially for large-capacity one-way data reads.

[0059] CXL memory is a dual-channel DDR4 memory expansion slot that supports up to 512GB of memory. Since CXL memory's read and write bandwidth and latency are second only to the server's local memory, and its read and write efficiency is much higher than that of hard drives and Ethernet, using CXL memory as real-time shared storage space can significantly accelerate data synchronization between primary and standby servers compared to traditional hot standby technology. This reduces data loss during switchover between primary and standby servers, resulting in faster switchover efficiency.

[0060] Compared to the shared storage (i.e., hot standby server system operating memory) space in related technologies, which stores operating information over the network in a shared disk array or hard disk storage area, this embodiment mirrors the primary server's memory information in real time to shared CXL memory via the CXL protocol. Compared to network hard disk storage, CXL storage is an enhanced version of the PCIE bus, offering faster data transfer speeds and lower latency, significantly improving the speed of primary and standby server node switching. Therefore, this embodiment uses CXL memory as the shared storage space for hot standby, effectively accelerating primary and standby node switching and improving service continuity. Each accelerator card is used to obtain, through the first Ethernet interface, the operating status information, operating log information and configuration information of the corresponding connected main server from the management module of the main server corresponding to each accelerator card, and store the memory information of the corresponding connected main server in the memory of the corresponding connected main server to the computing extension connection memory through the computing extension connection interface, and judge whether the corresponding connected main server has crashed based on the operating status information. In the case that the corresponding connected main server has crashed, the corresponding connected main server that has crashed is used as the crashed main server, and the pending data to be transmitted to the crashed main server is obtained, and the operating log information, memory information, configuration information and pending data of the crashed main server are transmitted to the standby server;

[0061] The standby server is used to simulate and generate a virtual application environment of the downtime main server based on the operation log information, memory information and configuration information of the downtime main server, and execute the application service of the downtime main server based on the virtual application environment and the data to be processed.

[0062] Optionally, during the hot backup process, each accelerator card can be a management module that monitors the corresponding primary server in real time via the first Ethernet interface to obtain the operating status information, operating log information, and configuration information of the corresponding primary server. Furthermore, through the CXL interface, the accelerator card obtains the corresponding primary server's memory information from the corresponding primary server's memory in real time and stores the corresponding primary server's memory information in real time in the CXL memory. Based on the obtained operating status information, the accelerator card determines whether the corresponding primary server has crashed. If the corresponding primary server has crashed, the accelerator card identifies the crashed primary server as the crashed primary server. The accelerator card promptly receives pending data from the crashed primary server and transmits the crashed primary server's operating log information, configuration information, and pending data, as well as the crashed primary server's memory information stored in the CXL memory, to the backup server in real time.

[0063] When the standby server learns that the primary server has crashed, it switches from sleep mode to active mode. After obtaining the operational log information, memory information, configuration information, and pending data of the crashed primary server, the standby server first configures the virtual application environment of the crashed primary server based on the operational log information, memory information, and configuration information of the crashed primary server, so that the configured virtual application environment is compatible with the application environment of the crashed primary server, thereby better taking over the application services of the crashed primary server. After confirming that the virtual application environment configuration of the crashed primary server is complete, the pending data can be processed in the virtual application environment compatible with the application environment of the crashed primary server, thereby taking over the application services of the crashed primary server.

[0064] The hot backup system provided in this embodiment can realize hot backup of multiple servers sharing one backup server by adding an acceleration card of a small system on the same backup server, and does not require the primary and backup servers to be synchronized at all times. It only needs to transmit the operation log information and configuration information of the down main server shared by the acceleration card and the memory information of the main server stored in real-time mirroring from the main server using CXL memory and the pending data to be transmitted to the down main server quickly and in real time to the backup server when a failure of the corresponding connected main server is detected. In this way, the backup server can quickly restore the data service of the down main server based on the operation log information, memory information and configuration information of the down main server and the pending data. The system has the significant advantages of low data loss, fast recovery and low equipment investment, which can save equipment deployment costs while improving the primary and backup rapid switching speed, recovery function and service continuity, ensuring data fault tolerance and ensuring the online service quality of the server.

[0065] In some embodiments, each accelerator card further includes a second Ethernet interface;

[0066] The second Ethernet interface and the first data transmission interface of the switch 240 are used to obtain the data to be processed from the switch 240 .

[0067] As shown in FIG3 , the accelerator card further includes a second Ethernet interface P1 , which may also be a dual-interface high-speed Ethernet interface.

[0068] As shown in Figure 4, the second Ethernet interface P1 is used to connect to the first data transmission interface of the switch that competes with the corresponding connected main server. The specific competition relationship is configured by the switch. When the corresponding connected main server is operating normally, the second Ethernet interface P1 enters sleep mode, and the switch will give priority to sending the pending data (such as pending user data) to the corresponding connected main server; when the corresponding connected main server goes down, such as being disconnected or unresponsive, the second Ethernet interface P1 enters active mode from sleep mode, and the switch automatically switches to switching the pending data to the backup server through the second Ethernet interface P1; the second Ethernet interface P1 will also always switch from active mode to sleep mode, or from sleep mode to active mode according to the monitored operating status of the corresponding connected main server.

[0069] In some embodiments, each accelerator card further includes a local memory and storage unit;

[0070] The local memory is used to store the operation status information, operation log information and configuration information transmitted by each server during the first time period;

[0071] a storage unit, configured to cyclically store the operation status information, operation log information, and configuration information transmitted by each server during the second time period;

[0072] The time interval between each time in the first time period and the current time is less than or equal to the preset interval, and the time interval between each time in the second time period and the current time is greater than the preset interval.

[0073] As shown in FIG3 , the accelerator card is further configured with a local memory and a storage unit; the local memory may be a memory that supports a local memory cache space of up to 288 GB; and the storage unit may be an expandable SD (Secure Digital card) storage space.

[0074] The accelerator card receives the operation status information, operation log information and configuration information. When it is determined that the corresponding connected main server is normal based on the operation status information, the newly acquired operation status information, operation log information and configuration information are stored in the local memory, and the historically acquired operation status information, operation log information and configuration information are stored in the storage unit in a circular manner from the local memory.

[0075] The accelerator card in the system provided in this embodiment supports multi-memory data storage including CXL memory, local memory and storage unit. Through reasonable storage and time interval control, it can improve data transmission efficiency, quickly store and access server information, and retain historical data for subsequent analysis and processing, thereby accelerating the switching speed of the primary and backup nodes and improving service continuity.

[0076] In some embodiments, each accelerator card further includes an accelerator functional unit;

[0077] The Accelerator Functional Unit (AFU) is the accelerator card's local data scheduling system, implemented using an FPGA (Field-Programmable Gate Array). Thanks to the FPGA's internal structure and high-speed parallel capabilities, it can easily implement distributed algorithm structures and high-speed interface switching, making it particularly suitable for high-speed digital signal processing and parallel data development.

[0078] AFU is used for data scheduling and management of the entire system. It also supports CXL memory and local large-capacity local memory and local SD card storage space, which can be used for local data caching and data storage. It is crucial for hot backup systems and can implement a one-backup-multiple hot backup solution.

[0079] In some embodiments, the AFU can be equipped with 128G of memory space, which can be used to cache server data and user data to be processed (hereinafter also referred to as pending data) received by interface P0 and interface P1, and provide network conversion network card services for the standby server when the standby server starts the service.

[0080] The accelerator functional unit is specifically used to:

[0081] From the running status information, parse and obtain the heartbeat information of the corresponding connected master server;

[0082] When it is determined based on the heartbeat information that the heartbeat of the corresponding connected primary server is abnormal, it is determined that the corresponding connected primary server is down, and a device startup request message is sent to the backup server based on the fault identifier of the downed primary server through the high-speed serial computer expansion bus interface;

[0083] Upon receiving the startup response information returned by the standby server, the operation log information and configuration information of the downed primary server are transmitted to the standby server;

[0084] The startup response information is returned by the standby server when the system resources of the standby server meet the equipment resource configuration requirements of the downed primary server. The equipment resource configuration requirements are obtained based on the fault identifier association.

[0085] It should be noted that since AFU uses FPGA as the main control chip, some of the steps and processes executed by AFU can be carried out in parallel, and unless there are special circumstances triggering, such as the failure of the main server, the steps and processes do not affect each other. The following can configure some of the steps and processes executed in parallel by AFU according to actual needs, and no specific restrictions are made here.

[0086] The following describes the data backup steps between an accelerator card, a primary server connected to the accelerator card, and a backup server.

[0087] FIG5 is a flow chart of data backup management provided by this application. As shown in FIG5 , after each accelerator card is inserted into the backup server, the steps of data backup management include:

[0088] Step 501: The first Ethernet interface P0 of the accelerator functional unit is connected to the network port of the corresponding main server for communication;

[0089] Step 502: The accelerator functional unit obtains the operating status information, operating log information, and configuration information of the corresponding connected master server through the first Ethernet interface P0.

[0090] In step 503, after obtaining the operating status information, the accelerator functional unit may parse the operating status information to obtain the heartbeat information of the corresponding connected primary server, and determine whether the heartbeat of the corresponding connected primary server is abnormal based on the heartbeat information; if normal, it is determined that the corresponding connected primary server is normal, and the data synchronization backup step is executed; if abnormal, it is determined that the corresponding connected primary server is down, and the master-slave switching step is executed.

[0091] FIG6 is a schematic diagram of the master-slave switching process provided by the present application; as shown in FIG6 , when it is determined that the corresponding connected master server is down, the corresponding connected master server that is down is used as the down master server, and the following steps are performed to implement the master-slave switching step:

[0092] Step 601: The accelerator functional unit wakes up the standby server through the PCIE interface, and generates a device startup request message for fault recovery based on the fault identifier of the downed primary server; and sends the device startup request message to the standby server through the PCIE interface.

[0093] In step 602, after receiving the device startup request information, the standby server evaluates its own system resources and responds to the device startup request information for fault recovery; specifically, the standby server parses and obtains the fault identifier from the device startup request information, obtains the device resource configuration requirements of the downtime main server through the fault identifier, evaluates its own system resources through the device resource configuration requirements, that is, determines whether its own system resources meet the device resource configuration requirements of the downtime main server, and returns an abnormal prompt information to the accelerator functional unit when it determines that its own system resources do not meet the device resource configuration requirements of the downtime main server; and returns a startup response information to the accelerator functional unit when it determines that its own system resources meet the device resource configuration requirements of the downtime main server.

[0094] In step 603, after receiving the startup response information returned by the backup server, the accelerator functional unit transmits the operation log information and configuration information obtained from the downed primary server to the backup server, so that the backup server can quickly perform subsequent active-standby switching and take over the application service of the downed primary server.

[0095] In the system provided by this embodiment, when the main server goes down and its heartbeat signal is abnormal, the acceleration card will promptly trigger the interaction and switching process with the backup server, and transmit the operation log information and configuration information shared and obtained from the down main server to the backup server, so that the backup server can quickly perform subsequent master-slave switching and take over the application services of the down main server, thereby achieving maximum data continuity, protecting important data from loss, and realizing fast and secure master-slave switching services.

[0096] In some embodiments, the standby server is specifically configured to:

[0097] Upon receiving the operation log information and configuration information of the downtime main server, the initial application environment of the downtime main server is simulated and generated through the virtual machine according to the operation log information and configuration information of the downtime main server;

[0098] When it is determined that the initial application environment simulation generation is completed, sending a first configuration completion message to the accelerator functional unit through the high-speed serial computer expansion bus interface;

[0099] Receiving the memory information of the downtime main server transmitted by the accelerator functional unit, updating the initial application environment according to the memory information of the downtime main server to obtain a virtual application environment;

[0100] The memory information of the downed main server is transmitted by the accelerator functional unit through the computing extension connection memory when the first configuration completion message is received.

[0101] As shown in FIG6 , the steps of active / standby switching further include:

[0102] In step 604, after receiving the operation log information and configuration information, the standby server can quickly simulate and build the initial application environment of the downed main server through the virtual machine based on the operation log information and configuration information, and feedback to the accelerator functional unit when the simulation generation of the initial application environment is completed; the specific feedback method can be sending a first configuration completion message to the accelerator functional unit through the PCIE interface; the so-called first configuration completion message is used to indicate that the initial application environment configuration of the downed main server is completed.

[0103] In step 605, after receiving the first configuration completion message, the accelerator functional unit may determine that the initial application environment configuration of the downed primary server is complete; at this time, the memory information of the downed primary server shared in the CXL memory may be mirrored and transmitted to the backup server via the PCIE interface.

[0104] In step 606, the backup server mirrors the memory information of the downed primary server stored in the shared CXL memory to its local memory, and updates the initial application environment based on the memory information to obtain the virtual application environment, so as to realize the switching between the primary and backup environments, so as to quickly perform subsequent primary and backup switching and take over the application services of the downed primary server with a lower fault tolerance rate.

[0105] In the system provided by this embodiment, when the primary server crashes and data switching and backup server environment configuration are performed, the backup server will perform an initial application environment simulation based on the operation log information and configuration information of the failed primary server. After the initial application environment simulation is completed, the memory information of the failed primary server shared in the CXL memory will be locally stored, thereby effectively realizing real-time environment backup. When the primary server fails, the virtual application environment can be quickly enabled to ensure the continuity and availability of the system.

[0106] In some embodiments, the standby server is further configured to:

[0107] When it is determined that the configuration of the virtual application environment is completed, sending a second configuration completion message to the accelerator functional unit via the high-speed serial computer expansion bus interface;

[0108] receiving data to be processed sent by the accelerator functional unit; the data to be processed is sent by the accelerator functional unit when the second configuration completion message is received;

[0109] Execute application services of the downed primary server based on the virtual application environment and the data to be processed.

[0110] As shown in Figure 6, step 606 also includes: after determining that the virtual application environment configuration is completed, the backup server feeds back a configuration completion message to the accelerator functional unit; specifically, it can be sending a second configuration completion message to the accelerator functional unit through the PCIE interface; the second configuration completion message here is used to indicate that the virtual application environment configuration is completed.

[0111] The steps of active / standby switching also include:

[0112] Step 607: After receiving the second configuration completion message, the accelerator functional unit sends the to-be-processed user data transmitted by the second Ethernet interface P1 cached in the memory to the standby server.

[0113] In step 608, after receiving the user data to be processed, the backup server uses the accelerator card as the network card and executes the application service of the downed primary server based on the virtual application environment and the data to be processed, so as to take over the application service of the downed primary server.

[0114] In some embodiments, the accelerator functional unit is further configured to:

[0115] When it is determined that the corresponding connected primary server is down, waking up the second Ethernet interface;

[0116] Sending a data transmission request to the switch via the second Ethernet interface according to the fault identification of the downed primary server;

[0117] The data to be processed transmitted by the switch is received through the second Ethernet interface; the data to be processed is transmitted by the switch when receiving a data transmission request.

[0118] As shown in FIG6 , the steps of active / standby switching further include:

[0119] In step 609, when it is determined that the corresponding connected primary server has crashed, the switch enters a dormant state due to the downtime of the corresponding connected primary server. At this time, the accelerator functional unit wakes up the second Ethernet interface P1 and establishes an interface with the first data transmission interface DX0 of the switch to receive the to-be-processed data transmitted by the first data transmission interface DX0 due to the switch entering a dormant state due to the second data transmission interface D0X.

[0120] In step 610 , the accelerator functional unit caches the to-be-processed data of the downed primary server received by the second Ethernet interface P1 in a local memory.

[0121] In the system provided by this embodiment, during the process of data switching and backup server environment configuration, the accelerator card will synchronously cache the received user data. When the backup server is ready to take over the application service of the failed primary server, the cached pending data of the failed primary server will be delivered to the backup server to maximize data continuity, protect important data from being lost, and realize fast and secure primary-backup switching service.

[0122] In some embodiments, the accelerator functional unit is further configured to:

[0123] When it is determined that the corresponding connected main server is down, the connection between the computing extension connection interface and the memory interface of the downed main server is cut off.

[0124] As shown in FIG6 , the steps of active / standby switching further include:

[0125] In step 611, when it is determined that the corresponding connected main server has crashed, the accelerator functional unit cuts off the connection between the memory interface and the CXL interface of the corresponding connected main server that has crashed, so as to isolate the fault, prevent data confusion and inconsistency, and improve backup performance.

[0126] It should be noted that the above steps 601-606, steps 609-610, and step 611 can be executed in parallel to improve backup efficiency.

[0127] In some embodiments, the accelerator functional unit is further configured to:

[0128] When it is determined based on the heartbeat information that the heartbeat of the corresponding connected main server is normal, the corresponding connected main server is determined to be normal, and the first data transmission interface is controlled to enter sleep mode through the second Ethernet interface, and the backup server is controlled to be in standby mode.

[0129] As shown in Figure 5, the data synchronization and backup steps include:

[0130] Step 504: the accelerator functional unit controls the first data transmission interface DX0 to enter the sleep mode via the second Ethernet interface P1;

[0131] Step 505: The standby server is in a standby state; and the standby server and the corresponding primary server are connected to the same database and do not access it.

[0132] In the system provided by this embodiment, a standby server can be inserted with multiple acceleration cards and serve as a standby server for multiple main servers at the same time. When all main servers are in normal status, the standby server is actually in a dormant state, and only the acceleration card is in real-time working status, which can be more beneficial to reduce system power consumption and reduce the deployment of standby servers. In addition, even if the acceleration card fails, the standby server can also serve as a monitoring host for the acceleration card, and perform human-computer interaction to quickly replace the acceleration card, thereby optimizing and ensuring the security of the storage space shared by the main and standby servers, and further strengthening the security of the entire hot backup system.

[0133] In some embodiments, the corresponding connected master server is used to:

[0134] When it is determined based on the heartbeat information that the heartbeat of the corresponding connected primary server is normal and the first data transmission interface has entered the sleep mode, a connection is established with the second data transmission interface of the switch to control the second data transmission interface to enter the active mode;

[0135] When it is determined that the second data transmission interface enters the active mode, the data to be processed is received through the second data transmission interface, and the application service is executed according to the data to be processed.

[0136] As shown in FIG5 , the data synchronization and backup steps further include:

[0137] In step 506, when it is determined based on the heartbeat information that the heartbeat of the corresponding connected primary server is normal and the first data transmission interface DX0 has entered the sleep mode, the switch sets the priority of the second data transmission interface D0X to be higher than the first data transmission interface DX0, so as to achieve competition for user data to be processed by the primary and backup servers.

[0138] Step 507, when it is determined that the second data transmission interface P1 enters the activation mode, the corresponding connected main server receives the to-be-processed data of the corresponding connected main server through the second data transmission interface P1, and accesses the database to execute application services according to the to-be-processed data of the corresponding connected main server, and returns the corresponding data.

[0139] In some embodiments, the data synchronization and backup step further includes:

[0140] Step 508: If the heartbeat is normal, the corresponding master server synchronously mirrors the memory information of the corresponding master server to the CXL memory of the accelerator card through the CXL interface, so that the memory information stored in the CXL memory of the accelerator card is consistent with the memory information stored in the memory of the corresponding master server.

[0141] Step 509 : When the CXL memory of the accelerator card completes storing the memory information, it is determined that the memory loading of the accelerator card and the corresponding host server is completed.

[0142] In some embodiments, the data synchronization and backup step further includes:

[0143] In step 510, when the heartbeat is normal, the accelerator functional unit also synchronously compares the current secondary data of the corresponding connected main server, that is, the operation status information, operation log information and configuration information with the last storage record, discards the data in the current secondary data that is the same as the last storage record, retains the different data, and records the acquisition timestamps of the different data, so as to store them in the local memory or storage unit in a cycle according to the acquisition timestamps.

[0144] In the system provided by this embodiment, when all main servers are in normal state, the backup server is actually in a dormant state. Only the accelerator card and the main server are in real-time working state, and only the main server executes application services. The accelerator card shares and stores the information of the main server, which can be more beneficial to reduce system power consumption and reduce the deployment of backup servers.

[0145] In some embodiments, the accelerator functional unit is further configured to:

[0146] Upon receiving a recovery request from the downed primary server, determining that the downtime event of the downed primary server has been resolved, switching the downed primary server to the primary server to be recovered, and sending a switch request to the standby server;

[0147] Receive switch response information, operation log information, and configuration information from the standby server;

[0148] If it is determined that the standby server allows switching according to the switching response information, sending the first switching instruction, the operation log information and the configuration information of the standby server to the main server to be restored;

[0149] Receive configuration status information of the main server to be restored; the configuration status information is generated by the main server to be restored after receiving the first switching instruction, and performing application environment configuration according to the operation log information and configuration information of the standby server.

[0150] FIG7 is a flowchart of the standby master switching step provided in this application. As shown in FIG7 , the standby master switching step includes:

[0151] Step 701: After the corresponding connected master server that has experienced downtime recovers from the fault and returns to normal, it notifies the accelerator functional unit through the private network to apply for data switching, that is, sends a recovery request to the accelerator functional unit through the first Ethernet interface P0;

[0152] Step 702: After receiving the restoration request of the primary service through the first Ethernet interface P0, the accelerator functional unit determines that the downtime event of the downed primary server has been resolved, and then sends a switch request to the backup server through the PCIE interface;

[0153] Step 703: After receiving the switching request, the standby server sends a switching response message to the accelerator functional unit through the PCIE interface, indicating that the switching is approved, and returns the standby server's operation log information and configuration information to the accelerator functional unit.

[0154] Step 704: After receiving the switching response information, the accelerator functional unit determines that the standby server allows switching;

[0155] Step 705: When the accelerator function unit determines that the standby server allows switching, it sends a first switching instruction, operation log information, and configuration information of the standby server to the corresponding connected master server to be restored (i.e., the master server to be restored) through the first Ethernet interface P0;

[0156] In step 706, after receiving the first switching instruction, the main server to be restored configures the application environment according to the received operation log information and configuration information of the backup server, and feeds back the configuration status information to the accelerator functional unit in real time, so that the subsequent acceleration unit can read back the memory information of the backup server to the main server to be restored, and the main server to be restored continues the application service to realize the backup-master switching.

[0157] In the system provided by this embodiment, after the main server returns to normal, it notifies the acceleration card that it has returned to normal and applies to switch back to the main server. After receiving the request, the acceleration card will send a switching request to the backup server. The backup server will set a breakpoint according to the operating status and return the switching response information, operation log information and configuration information that allow switching to the acceleration card. After the accelerator card parses and allows switching, it will return the first switching instruction, operation log information and configuration information that allow switching to the main server so that the main server can perform system configuration, thereby ensuring that switching is performed at the right time, achieving rapid system recovery, reducing system downtime, and improving the availability and stability of the backup system.

[0158] In some embodiments, the accelerator functional unit is further configured to:

[0159] When it is determined according to the switching response information that the standby server allows switching, sending a second switching instruction to the standby server;

[0160] Acquire memory information in a virtual application environment transmitted by a backup server through a high-speed serial computer expansion bus interface mirror;

[0161] Storing memory information in the virtual application environment in the computing extension connection memory;

[0162] The memory information in the virtual application environment is transmitted by the standby server when the standby server receives the second switching instruction.

[0163] As shown in FIG7 , step 705 further includes: when the accelerator function unit determines that the standby server allows switching, sending a second switching instruction to the standby server and opening the PCIE interface and the CXL memory channel;

[0164] The standby-active switching steps also include:

[0165] Step 707: After receiving the second switching instruction, the standby server transmits the memory information in the virtual application environment constructed by the local virtual machine to the CXL memory of the accelerator card through the PCIE interface mirroring;

[0166] In step 708 , the accelerator functional unit obtains the memory information that needs to be mirrored in the standby server and stores it in the CXL memory.

[0167] It should be noted that steps 707-708 and step 706 can be executed synchronously, that is, after the standby server sends the first switching instruction, operation log information and configuration information allowing switching back to the main server to be restored, while the main server to be restored is performing system configuration, the standby server synchronously writes the memory information into the acceleration card CXL memory through the PCIE interface. When the acceleration card receives the initial configuration of the main server to be restored, it can promptly mirror and read back the memory information of the standby server from the acceleration card CXL memory, thereby avoiding information transmission delays and improving data backup efficiency.

[0168] In some embodiments, the accelerator functional unit is further configured to:

[0169] When it is determined based on the configuration status information that the configuration of the main server to be restored is completed, and based on the storage status information of the computing extension connection memory, it is determined that the storage of memory information in the virtual application environment transmitted by the standby server is completed, a third switching instruction is sent to the main server to be restored, and the communication path between the high-speed serial computer expansion bus interface and the standby server is cut off, and the connection between the computing extension connection interface and the memory interface of the main server to be restored is restored.

[0170] As shown in FIG7 , the standby-active switching step further includes:

[0171] In step 709, the accelerator functional unit determines whether the configuration of the primary server to be restored is complete based on the configuration status information, and determines whether the memory information in the virtual application environment transmitted by the backup server is completely stored based on the storage status information. If the configuration and storage are complete, step 710 is executed.

[0172] In step 710, the accelerator functional unit sends a third switching instruction to the primary server to be recovered through the first Ethernet interface P0 to notify the primary server to be recovered that switching can be performed; at the same time, the communication path between the PCIE interface and the backup server is cut off, and the connection between the CXL interface and the memory interface of the primary server to be recovered is restored.

[0173] It should be noted that the system information of the primary and backup servers (such as operating status information, operating log information, and configuration information) is implemented through a private network. Since the primary and backup server switching requires synchronization of memory data and system information, the amount of system information is very small compared to the amount of memory data, and the delay can be ignored.

[0174] In addition, after the communication path between the standby server and the PCIE interface of the accelerator card is cut off, the standby server can continue to monitor the extended accelerator card after being released, and re-enter the master-slave state switching when any main server is found to have failed through the accelerator card. During the backup process, it may happen that multiple main servers need the standby server to provide application service state switching. At this time, the standby server needs to be a device with better performance and configuration than the main server, which can start multiple virtual machines at the same time to take over the business of more than two main servers in a short period of time, thereby greatly improving the reliability of the multi-server backup system. Generally, the probability of two main servers failing at the same time and the probability of the main and standby servers failing at the same time is the same, and the probability of multiple main servers failing is smaller. Therefore, the security of the hot backup system provided by this application is higher than that of the traditional dual-machine hot standby.

[0175] The system provided in this embodiment closes the channel between the local PCIE and CXL memory after the accelerator card receives information indicating that the configuration of the primary server is complete and the CXL memory mirroring is complete. Then, the memory channel between the external CXL interface and the memory interface of the primary server to be restored is opened, so that the primary server to be restored can read back the memory information of the standby server and mirror it locally, thereby completing the switching between the standby server and the host. This ensures that the memory data of the standby server remains synchronized with that of the primary server, avoiding data loss or inconsistency. After obtaining the latest memory information of the standby server, the primary server can quickly restore application services, reducing system downtime and improving the availability of the hot backup system.

[0176] In some embodiments, the accelerator functional unit is further configured to:

[0177] When it is determined according to the switching response information that the standby server allows switching, the connection with the first data transmission interface of the switch is cut off to control the first data transmission interface to enter the sleep mode.

[0178] As shown in FIG7 , the standby-active switching step further includes:

[0179] In step 711, when it is determined based on the switching response information that the backup server allows switching, the accelerator function unit can synchronously cut off the connection between the second Ethernet interface P1 and the first data transmission interface of the switch, so that the first data transmission interface enters the sleep mode; so that the switch can preferentially transmit the data to be processed to the corresponding main server to be restored through the second data transmission interface of the switch, so that the main server to be restored can quickly restore the application service.

[0180] It should be noted that step 711 can be executed synchronously with steps 707-708 and step 706 to improve the hot backup efficiency.

[0181] In some embodiments, the master server to be restored is used to:

[0182] When it is determined that the first data transmission interface enters the sleep mode, starting a connection with the second data transmission interface of the switch to control the second data transmission interface to enter the active mode;

[0183] When the third switching instruction is received and the second data transmission interface enters the activation mode, receiving the data to be processed of the primary server to be restored through the second data transmission interface;

[0184] Reading memory information in the virtual application environment from the computing extension connection memory through the computing extension connection interface;

[0185] Update the initial environment based on the memory information in the virtual application environment;

[0186] Restore the application services of the master server to be restored based on the updated application environment and the pending data of the master server to be restored;

[0187] The initial environment is generated by configuring the application environment based on the operation log information and configuration information of the standby server.

[0188] As shown in FIG7 , the standby-active switching step further includes:

[0189] Step 712: When the master server to be restored determines that the first data transmission interface has entered the sleep mode, the master server starts a connection with the second data transmission interface D0X of the switch, so that the second data transmission interface enters the active mode.

[0190] In step 713, when the main server to be restored receives the third switching instruction and the second data transmission interface enters the activation mode, it establishes a connection between its own memory interface and the CXL interface to read back the memory information of the standby server from the CXL memory of the accelerator card to the local memory, and receives the data to be processed through the second data transmission interface D0X, thereby updating the initial environment generated by the application environment configuration based on the operation log information and configuration information of the standby server based on the acquired memory information, so as to restore the application service based on the updated application environment and the data to be processed.

[0191] The system provided in this embodiment implements hot backup with one backup and multiple masters through a PCIE accelerator card with expandable CXL memory. During the backup-master switching process, the host computer only needs to keep an eye on the working status of the same backup server to know the operating status of the system. Once the backup server is started, it can quickly respond to repair the primary server, then switch back to the primary server for operation, while the backup server continues to standby. It is particularly suitable for online service systems, can be quickly deployed, and has market innovation and leading advantages.

[0192] As shown in FIG2 , the hot backup system further includes a database 250 ;

[0193] The database 250 shares a connection with each primary server and backup server.

[0194] Optionally, the database is used to share the security data transmitted by the primary server and the backup server, which can be managed by dual-machine or cluster software to store a secure portion of memory, thereby improving the system's availability, performance and management efficiency.

[0195] The hot backup method provided in the present application is described below. The hot backup method described below and the hot backup system described above can be referenced to each other.

[0196] FIG8 is a flow chart of the hot backup method provided by the present application; the method is executed by the hot backup system provided by the above embodiments; as shown in FIG8 , the method includes:

[0197] In step 810, each accelerator card obtains the operating status information, operating log information, and configuration information of the corresponding primary server from the management module of the primary server to which each accelerator card is connected, through the first Ethernet interface. The accelerator card then mirrors the memory information of the corresponding primary server in the memory of the corresponding primary server to the computing extension connection memory through the computing extension connection interface. Based on the operating status information, the accelerator card determines whether the corresponding primary server has crashed. If the corresponding primary server has crashed, the accelerator card designates the crashed primary server as the crashed primary server, obtains pending data to be transmitted to the crashed primary server, and transmits the pending data, operating log information, memory information, configuration information, and pending data of the crashed primary server to the standby server.

[0198] Step 820: The standby server simulates and generates a virtual application environment of the downed primary server based on the operation log information, memory information, and configuration information of the downed primary server, and executes the application service of the downed primary server based on the virtual application environment and the data to be processed.

[0199] Optionally, during the hot backup process, each acceleration card can monitor and obtain the operating status information, operating log information, memory information and configuration information of the corresponding connected main server in real time, and judge whether the corresponding connected main server has crashed based on the acquired operating status information, so that when the corresponding connected main server crashes, it can promptly receive the pending data to be transmitted to the main server, and transmit the operating log information, memory information, configuration information and pending data of the main server to the backup server in real time.

[0200] When the standby server learns that the primary server has crashed, it switches from dormant mode to active mode. Upon obtaining the primary server's operational log information, memory information, configuration information, and pending data, the standby server first configures the primary server's virtual application environment based on the primary server's operational log information, memory information, and configuration information, so that the configured virtual application environment is compatible with the primary server's application environment, thereby better taking over the primary server's application services. After confirming that the virtual application environment configuration of the corresponding connected primary server is complete, the standby server can process the pending data in the virtual application environment compatible with the primary server's application environment, thereby taking over the application services of the corresponding connected primary server.

[0201] The hot backup method provided in this embodiment transmits the operation log information and configuration information of the down master server shared by the acceleration card as a backup, as well as the memory information of the master server stored in real-time mirroring from the master server using CXL memory, and the pending data to be transmitted to the down master server quickly and in real time to the standby server when a failure of the corresponding connected master server is detected. In this way, the standby server can quickly restore the data service of the down master server based on the operation log information, memory information and configuration information of the down master server as well as the pending data to be processed. The method has the significant advantages of low data loss, fast recovery and low equipment investment, which can save equipment deployment costs while improving the master-slave rapid switching speed, recovery function and service continuity, ensuring data fault tolerance and ensuring the online service quality of the server.

[0202] FIG9 illustrates a schematic diagram of the physical structure of an electronic device. As shown in FIG9 , the electronic device may include: a processor 910, a communications interface 920, a memory 930, and a communication bus 940. The processor 910, the communications interface 920, and the memory 930 communicate with each other via the communication bus 940. The processor 910 may call logic instructions in the memory 930 to execute a hot backup method.

[0203] In addition, the logic instructions in the above-mentioned memory 930 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the relevant technology or the 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 for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of one or more embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a disk or an optical disk, etc., which can store program code.

[0204] On the other hand, the present application also provides a computer program product, which includes computer-readable instructions. The computer-readable instructions can be stored on a non-transitory computer-readable storage medium. When the computer-readable instructions are executed by a processor, the computer can execute the hot backup method provided by one or more of the above embodiments.

[0205] On the other hand, referring to FIG10 , the present application further provides a non-transitory computer-readable storage medium having computer-readable instructions stored thereon, which are implemented by a processor to execute the hot backup method provided in one or more of the above embodiments when the computer-readable instructions are executed.

[0206] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units. That is, they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present embodiment without inventive effort.

[0207] Through the description of the above embodiments, those skilled in the art can clearly understand that the embodiments can be implemented by means of software plus the necessary general hardware platform, or of course, by hardware. Based on this understanding, the above technical solution, in essence, or the part that contributes to the relevant technology, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in one or more embodiments or certain parts of one or more embodiments.

[0208] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in one or more of the aforementioned embodiments can still be modified, or some of the technical features thereof can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A hot backup method, characterized in that: The method comprises: The accelerator card obtains the operation status information, operation log information and configuration information of the corresponding main server from the main server to which the accelerator card is connected, and stores the memory information of the corresponding main server in the memory of the corresponding main server in the computing expansion connection memory of the accelerator card through the computing expansion connection interface of the accelerator card, and judges whether the corresponding main server is down based on the operation status information. If the corresponding main server is down, the downed main server is used as the downed main server, and the pending data to be transmitted to the downed main server is obtained, and the operation log information, memory information, configuration information and the pending data of the downed main server are transmitted to the standby server; and The standby server simulates and generates a virtual application environment of the downtime main server according to the operation log information, memory information and configuration information of the downtime main server, and executes the application service of the downtime main server according to the virtual application environment and the data to be processed.

2. The method according to claim 1, characterized in that The transmitting the operation log information, memory information, configuration information and the to-be-processed data of the downed primary server to the standby server includes: The operation log information, memory information, configuration information and the data to be processed of the downed main server are transmitted to the standby server through the high-speed serial computer expansion bus interface of the accelerator card, wherein the high-speed serial computer expansion bus interface is connected to the standby server.

3. The method according to claim 2, characterized in that The method further comprises: The accelerator card parses the running status information to obtain the heartbeat information of the corresponding connected master server; When it is determined based on the heartbeat information that the heartbeat of the corresponding connected primary server is abnormal, determining that the corresponding connected primary server is down, and sending a device startup request message to the standby server through the high-speed serial computer expansion bus interface according to the fault identifier of the downed primary server; Upon receiving the startup response information returned by the standby server, transmitting the operation log information and configuration information of the downed primary server to the standby server; and The startup response information is information returned by the standby server when the system resources of the standby server meet the equipment resource configuration requirements of the downed primary server, and the equipment resource configuration requirements are requirements obtained by associating with the fault identifier.

4. The method according to claim 3, characterized in that The standby server simulates and generates a virtual application environment of the downtime main server based on the operation log information, memory information, and configuration information of the downtime main server, including: The standby server, upon receiving the operation log information and configuration information of the downtime main server, simulates and generates the initial application environment of the downtime main server according to the operation log information and configuration information of the downtime main server; When it is determined that the simulation generation of the initial application environment is complete, sending a first configuration completion message to the accelerator card; receiving the memory information of the downtime main server transmitted by the acceleration card, and updating the initial application environment according to the memory information of the downtime main server to obtain the virtual application environment; and The memory information of the downed main server is information transmitted by the acceleration card through the computing extension connection memory when the acceleration card receives the first configuration completion message.

5. The method according to claim 4, characterized in that The method further comprises: When determining that the configuration of the virtual application environment is completed, the standby server sends a second configuration completion message to the accelerator function unit; receiving the data to be processed sent by the accelerator card; the data to be processed is data sent by the accelerator card when the accelerator card receives the second configuration completion message; and The application service of the downtime main server is executed according to the virtual application environment and the data to be processed.

6. The method according to claim 3, characterized in that The method further comprises: When determining that the corresponding connected main server is down, the accelerator card cuts off the connection between the computing expansion connection interface and the memory interface of the downed main server.

7. The method according to claim 3, characterized in that The obtaining of the data to be transmitted to the downtime main server includes: The data to be processed is obtained from the switch through a second Ethernet interface of the acceleration card, wherein the second Ethernet interface is connected to a first data transmission interface of the switch.

8. The method according to claim 7, characterized in that The method further comprises: The accelerator card wakes up the second Ethernet interface when determining that the corresponding main server is down; sending a data transmission request to the switch through the second Ethernet interface according to the fault identifier of the downed primary server; and The data to be processed transmitted by the switch is received through the second Ethernet interface; the data to be processed is data transmitted by the switch when the switch receives the data transmission request.

9. The method according to claim 7, characterized in that The method further comprises: When the acceleration card determines that the heartbeat of the corresponding connected main server is normal based on the heartbeat information, it determines that the corresponding connected main server is normal, controls the first data transmission interface to enter sleep mode through the second Ethernet interface, and controls the backup server to be in standby mode.

10. The method according to claim 9, characterized in that The method further comprises: When the corresponding connected master server determines, based on the heartbeat information, that the heartbeat of the corresponding connected master server is normal and the first data transmission interface enters the sleep mode, the corresponding connected master server establishes a connection with the second data transmission interface of the switch to control the second data transmission interface to enter the active mode; and When it is determined that the second data transmission interface enters the activation mode, the data to be processed of the corresponding connected main server is received through the second data transmission interface, and the application service of the corresponding connected main server is executed according to the data to be processed of the corresponding connected main server.

11. The method according to claim 3, characterized in that The method further comprises: Upon receiving the recovery request from the down master server, the accelerator card determines that the downtime event of the down master server has been resolved, switches the down master server to a master server to be recovered, and sends a switch request to the standby server; receiving switching response information, operation log information, and configuration information of the standby server; If it is determined according to the switching response information that the standby server allows switching, sending a first switching instruction, the operation log information and the configuration information of the standby server to the primary server to be restored; and Receive configuration status information of the main server to be restored; the configuration status information is information generated by the main server to be restored performing application environment configuration according to the operation log information and configuration information of the standby server when receiving the first switching instruction.

12. The method according to claim 11, characterized in that The method further comprises: Storing the operation status information, operation log information, and configuration information transmitted by each server during the first time period in the local memory of the accelerator card; cyclically storing the operation status information, operation log information, and configuration information transmitted by each server during the second time period through the storage unit of the acceleration card; and The time interval between each time in the first time period and the current time is less than or equal to a preset interval, and the time interval between each time in the second time period and the current time is greater than the preset interval.

13. The method according to claim 12, characterized in that The method further comprises: When the accelerator card determines, based on the switching response information, that the standby server allows switching, it sends a second switching instruction to the standby server; Acquire memory information in the virtual application environment mirrored and transmitted by the standby server through the high-speed serial computer expansion bus interface; Storing memory information in the virtual application environment into the computing extension connection memory; and The memory information in the virtual application environment is information transmitted by the standby server when receiving the second switching instruction.

14. The method according to claim 13, characterized in that The method further comprises: The accelerator card sends a message to the master server to be restored when it is determined based on the configuration status information that the configuration of the master server to be restored is complete and when it is determined based on the storage status information of the computing expansion connection memory that the storage of the memory information in the virtual application environment transmitted by the standby server is complete. The server sends a third switching instruction, cuts off the communication path between the high-speed serial computer expansion bus interface and the standby server, and restores the connection between the computing expansion connection interface and the memory interface of the main server to be restored.

15. The method according to claim 14, characterized in that The method further comprises: When the accelerator card determines, based on the switching response information, that the standby server allows switching, it cuts off the connection with the first data transmission interface of the switch to control the first data transmission interface to enter a sleep mode.

16. The method according to claim 15, characterized in that The method further comprises: When determining that the first data transmission interface has entered the sleep mode, the to-be-restored primary server starts a connection with the second data transmission interface of the switch to control the second data transmission interface to enter the active mode; When the third switching instruction is received and the second data transmission interface enters the activation mode, receiving the data to be processed of the to-be-restored primary server through the second data transmission interface; Reading memory information in the virtual application environment from the computing extension connection memory through the computing extension connection interface; updating the initial environment according to the memory information in the virtual application environment; Restoring the application service of the to-be-restored primary server according to the updated application environment and the to-be-processed data of the to-be-restored primary server; and The initial environment is an environment generated by performing application environment configuration according to the operation log information and configuration information of the standby server.

17. A hot backup system, characterized in that: include: Multiple accelerator cards, multiple master servers and one backup server; The accelerator card is connected to the primary server in a one-to-one correspondence and is also connected to the backup server; The accelerator card includes a computing expansion connection interface and a computing expansion connection memory; The acceleration card is used to obtain the operating status information, operating log information and configuration information of the corresponding main server from the main server corresponding to the acceleration card, and store the memory information of the corresponding main server in the memory of the corresponding main server in the computing expansion connection memory through the computing expansion connection interface, and judge whether the corresponding main server is down based on the operating status information. In the case that the corresponding main server is down, the corresponding main server that is down is used as the down main server, and the pending data to be transmitted to the down main server is obtained, and the operating log information, memory information, configuration information and the pending data of the down main server are transmitted to the standby server; as well as The standby server is used to simulate and generate a virtual application environment of the downtime main server based on the operation log information, memory information and configuration information of the downtime main server, and execute the application service of the downtime main server based on the virtual application environment and the data to be processed.

18. The hot standby system according to claim 17, characterized in that: The system further includes a database; The database is connected to each of the primary server and the backup server.

19. An electronic device comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, wherein: When the processor executes the program, the hot backup method according to any one of claims 1 to 16 is implemented.

20. A non-transitory computer-readable storage medium having computer-readable instructions stored thereon, characterized in that: When the computer-readable instructions are executed by a processor, the hot backup method according to any one of claims 1 to 16 is implemented.

Images