WO2023246241A1 - 数据处理系统、数据处理方法、装置及相关设备 - Google Patents

数据处理系统、数据处理方法、装置及相关设备 Download PDF

Info

Publication number
WO2023246241A1
WO2023246241A1 PCT/CN2023/087566 CN2023087566W WO2023246241A1 WO 2023246241 A1 WO2023246241 A1 WO 2023246241A1 CN 2023087566 W CN2023087566 W CN 2023087566W WO 2023246241 A1 WO2023246241 A1 WO 2023246241A1
Authority
WO
WIPO (PCT)
Prior art keywords
subsystem
mode
data processing
processing system
host
Prior art date
Application number
PCT/CN2023/087566
Other languages
English (en)
French (fr)
Inventor
曾红丽
王秀武
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Publication of WO2023246241A1 publication Critical patent/WO2023246241A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/07Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/14Error detection or correction of the data by redundancy in operation
    • G06F11/1479Generic software techniques for error detection or fault masking
    • G06F11/1492Generic software techniques for error detection or fault masking by run-time replication performed by the application software
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/07Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/14Error detection or correction of the data by redundancy in operation

Definitions

  • the present application relates to the field of data processing technology, and in particular, to a data processing system, data processing method, device and related equipment.
  • a dual-active data processing system means that both subsystems in the data processing system are running, and the data in the two subsystems remains consistent. In this way, when one subsystem cannot provide business services due to fire, flood, etc., another subsystem in the data processing system can take over the business in time, so that business services are not interrupted and data is not lost, thereby improving data processing. system reliability.
  • data processing systems usually run two subsystems in active-passive (AP) mode. Among them, only the subsystem in the active state can provide business services (such as data reading and writing services, etc.) to the host.
  • the subsystem in the passive state serves as the disaster recovery end. It usually backs up the data in the active subsystem in real time and does not provide business services when no exception occurs in the data processing system.
  • the host may require access to both subsystems in the data processing system, which makes it difficult for the data processing system running in AP mode to meet the needs of the host, thus affecting the use of the data processing system by tenants. experience.
  • inventions of the present application provide a data processing system.
  • the data processing system includes a first subsystem and a second subsystem, and the data processing system also includes a receiving unit and a processing unit; wherein the receiving unit is For receiving configuration instructions, for example, it can receive configuration instructions issued by the administrator, or it can detect changes in the networking mode of the host connected to the data processing system, etc.
  • the processing unit is used to perform the processing according to the configuration instructions received by the receiving unit. Switching between the first subsystem and the second subsystem from the first mode to the second mode.
  • the first mode is one of the AP mode and the AA mode between the first subsystem and the second subsystem.
  • the second mode is another mode between the AP mode and the AA mode between the first subsystem and the second subsystem.
  • the first subsystem and the second subsystem can be switched from the AA mode to the AP mode. mode, or switch from AP mode to AA mode, etc.
  • both the first subsystem and the second subsystem can provide data processing services for the host.
  • the AP mode only one subsystem (the subsystem in the active state) can provide data processing services for the host.
  • the data processing system can automatically switch the working mode between the first subsystem and the second subsystem according to changes in the tenant's needs (that is, changes in the working modes of the two subsystems matched by the tenant's host), thereby achieving
  • the data processing system can adapt to changes in tenant needs. That is, when the tenant requires AA mode, the data processing system can switch the working mode of the two subsystems to the AA mode required by the tenant. When the tenant requires AP mode again, the data processing system can Then switch the working mode of these two subsystems to AP mode, which can effectively improve the tenant's experience of using the data processing system.
  • the first subsystem and the second subsystem provide data processing services, such as data storage services, data computing services, etc., for the host of the target tenant based on the first mode, and, after the first After the working mode between the subsystem and the second subsystem is switched to the second mode, the operation between the first subsystem and the second subsystem is based on the second mode.
  • the receiving unit is specifically configured to detect the networking mode of the host connected to the data processing system and switch from the first networking mode to the second networking mode. At this time, the receiving unit can trigger a process of switching the working modes of the first subsystem and the second subsystem, so as to adapt to changes in the host's networking mode (different networking modes of the host) by switching the working modes of the two subsystems.
  • the data processing system may be required to provide data processing services based on different working modes).
  • the first networking mode of the host may be one of a parallel networking mode and a full-crossover networking mode
  • the second networking mode of the host may be a parallel networking mode or a full-crossover networking mode.
  • Another networking mode in cross-networking mode in cross-networking mode. In this way, when the host's networking mode switches between parallel networking mode and full-crossover networking mode, the data processing system can also adaptively switch the working mode between the two subsystems to adapt to the host networking mode. Variety.
  • the data processing system further includes a third subsystem and a fourth subsystem, and when the first subsystem and the second subsystem work in the second mode, the third subsystem and The fourth subsystem works in the first mode.
  • the data processing system can be based on multiple subsystems and provide AA mode and AP mode for different hosts at the same time. This allows the data processing system to select the subsystem of the corresponding operating mode according to the tenant's needs to provide business services for the tenant's host. This can meet the business needs of different hosts in actual application scenarios and improve the universality of business services provided by the data processing system.
  • the data processing system when the data processing system includes first to fourth subsystems at the same time, the first subsystem and the third subsystem belong to the first site, and the second subsystem and the fourth subsystem belong to Second site.
  • the first to fourth subsystems may respectively belong to different sites.
  • the data processing services provided by the first subsystem and the second subsystem for the target tenant are specifically data storage services.
  • the first subsystem includes at least one storage array to utilize the storage array. Store the data of the target tenant and provide data storage services.
  • the first subsystem and the second subsystem communicate through at least one of IP protocol, RDMA protocol, and FC.
  • the data processing service provided by the first subsystem and the second subsystem is specifically a data storage service, and the data storage service may specifically be a file storage service.
  • the data processing service provided by the first subsystem and the second subsystem is specifically a data storage service, and the data storage service may specifically be a block storage service.
  • the data processing service provided by the first subsystem and the second subsystem is specifically a data storage service, and the data storage service may specifically be an object storage service.
  • this application provides a data processing method, which is applied to a data processing system.
  • the data processing system includes a first subsystem, a second subsystem, a receiving unit and a processing unit.
  • the receiving unit receives the configuration instruction and provides the configuration instruction to the processing unit;
  • the processing unit switches the first subsystem and the second subsystem from the first mode to the second mode according to the configuration instruction; wherein the first mode is the second mode.
  • the first subsystem and the second subsystem work in one of the AP mode and the AA mode; the second mode is the other of the AP mode and the AA mode when the first subsystem and the second subsystem work. model.
  • the data processing system can automatically switch the working mode between the first subsystem and the second subsystem according to the changes in the tenant's needs, so that the data processing system can adapt to the changes in the tenant's needs, which can effectively improve the tenant's use of data processing. System experience.
  • the first subsystem and the second subsystem provide data processing services for the host of the target tenant based on the first mode
  • the connection between the first subsystem and the second subsystem is After the working mode is switched to the second mode, the first subsystem and the second subsystem provide data processing services for the host of the target tenant based on the second mode.
  • the working mode can be adaptively switched to continue to provide corresponding data processing services for the target tenant's host, which can effectively improve the tenant's experience of using the data processing system.
  • the receiving unit receiving the configuration instruction includes: the receiving unit detects the networking mode of the host connected to the data processing system, and switches from the first networking mode to the second networking mode. In this way, the receiving unit can trigger the process of switching the working modes of the first subsystem and the second subsystem by detecting changes in the networking mode of the host, so as to adapt to the networking of the host by switching the working modes of the two subsystems. Pattern changes
  • the first networking mode is one of a parallel networking mode and a full-crossover networking mode
  • the second networking mode is one of a parallel networking mode and a full-crossover networking mode. Another networking mode.
  • the data processing system can also adaptively switch the working mode between the two subsystems to adapt to the host networking mode. Variety.
  • the data processing system further includes a third subsystem and a fourth subsystem.
  • the third subsystem and the fourth subsystem work in the first mode.
  • the data processing system can be based on multiple subsystems and provide AA mode and AP mode for different hosts at the same time. This allows the data processing system to select the subsystem of the corresponding operating mode according to the tenant's needs to provide business services for the tenant's host. This can meet the business needs of different hosts in actual application scenarios and improve the universality of business services provided by the data processing system.
  • the first subsystem and the third subsystem belong to the first site, and the second subsystem and the fourth subsystem belong to the second site.
  • the first subsystem to the fourth subsystem may also belong to different sites respectively.
  • the first subsystem and the second subsystem are used to provide data storage services for target tenants, and the first subsystem includes at least one storage array.
  • the data processing system can use the storage array included in the subsystem to store the data of the target tenant to provide data storage services.
  • the first subsystem and the second subsystem communicate through at least one of IP protocol, RDMA protocol, and FC.
  • the data storage service provided by the first subsystem and the second subsystem is a file storage service.
  • the data storage service provided by the first subsystem and the second subsystem is a block storage service.
  • the data storage service provided by the first subsystem and the second subsystem is an object storage service.
  • inventions of the present application provide a data processing device.
  • the data processing device is applied to a data processing system.
  • the data processing system includes a first subsystem and a second subsystem.
  • the data processing device includes: a receiving unit for receiving Configuration instruction; processing unit, configured to switch between the first subsystem and the second subsystem from the first mode to the second mode according to the configuration instruction; wherein the first mode is between the first subsystem and the second subsystem One mode of working in active-passive AP mode and active-active AA mode; the second mode is another mode of working in AP mode and AA mode between the first subsystem and the second subsystem.
  • the first subsystem and the second subsystem provide data processing services for the host of the target tenant based on the first mode
  • the connection between the first subsystem and the second subsystem is After the working mode is switched to the second mode, the first subsystem and the second subsystem provide data processing services for the host of the target tenant based on the second mode.
  • the receiving unit is specifically configured to detect the networking mode of the host connected to the data processing system and switch from the first networking mode to the second networking mode.
  • the first networking mode is one of a parallel networking mode and a full-crossover networking mode
  • the second networking mode is one of a parallel networking mode and a full-crossover networking mode. Another networking mode.
  • the data processing system further includes a third subsystem and a fourth subsystem.
  • the third subsystem and the fourth subsystem work in the first mode.
  • the first subsystem and the third subsystem belong to the first site, and the second subsystem and the fourth subsystem belong to the second site.
  • the first subsystem and the second subsystem are used to provide data storage services for target tenants, and the first subsystem includes at least one storage array.
  • the first subsystem and the second subsystem communicate through at least one of Internet Protocol IP, Remote Direct Data Access RDMA protocol, and Fiber Channel FC.
  • the data storage service provided by the first subsystem and the second subsystem is a file storage service.
  • the data storage service provided by the first subsystem and the second subsystem is a block storage service.
  • the data storage service provided by the first subsystem and the second subsystem is an object storage service.
  • the technical effects of the third aspect and each embodiment in the third aspect can be found in The technical effects of the corresponding first aspect and each embodiment in the first aspect, or you can refer to the corresponding second aspect and the technical effects of each embodiment in the second aspect, which will not be described again here.
  • embodiments of the present application provide a computing device, including: a processor and a memory; the memory is used to store instructions, and when the computing device is running, the processor executes the instructions stored in the memory, so that the calculation The device executes the data processing method provided by the above second aspect or any implementation of the second aspect.
  • the memory can be integrated into the processor or independent of the processor.
  • the computing device may also include a bus. Among them, the processor is connected to the memory through a bus.
  • the memory may include readable memory and random access memory.
  • embodiments of the present application further provide a readable storage medium in which a program or instructions are stored, which, when run on a computer, enable the above second aspect or any one of the second aspects
  • the data processing methods provided in the implementation are executed.
  • embodiments of the present application also provide a computer program product containing instructions that, when run on a computer, cause the computer to execute the data processing method provided in the above second aspect or any implementation of the second aspect.
  • Figure 1 is a schematic structural diagram of an exemplary data processing system provided by an embodiment of the present application.
  • Figure 2 is a schematic diagram of the subsystem in the data storage system 100 provided by the embodiment of the present application operating based on AP mode;
  • Figure 3 is a schematic diagram of subsystem 102 continuing to provide business services after subsystem 101 fails in AP mode;
  • Figure 4 is a schematic diagram of the subsystem in the data storage system 100 provided by the embodiment of the present application operating based on the AA mode;
  • Figure 5 is a schematic diagram of subsystem 104 continuing to provide business services after subsystem 103 fails in AA mode;
  • Figure 6 is a schematic diagram of the working mode of the two subsystems switched from AP mode to AA mode according to the embodiment of the present application;
  • Figure 7 is a schematic diagram of the working mode of the two subsystems switched from AA mode to AP mode according to the embodiment of the present application;
  • Figure 8 is a schematic structural diagram of another exemplary data processing system provided by an embodiment of the present application.
  • Figure 9 is a schematic structural diagram of another exemplary data processing system provided by an embodiment of the present application.
  • Figure 10 is a schematic flow chart of a data processing method provided by an embodiment of the present application.
  • Figure 11 is a schematic flowchart of a data processing device provided by an embodiment of the present application.
  • Figure 12 is a schematic diagram of the hardware structure of a computing device provided by an embodiment of the present application.
  • the data processing system 100 may include multiple subsystems.
  • subsystem 101 and subsystem 102 are taken as an example for illustration. In actual applications, the data processing system 100 may include a larger number of subsystems.
  • subsystem 101 belongs to site (site) A
  • subsystem 102 belongs to site B.
  • Each site can be used to store files provided by tenants (or tenants).
  • Subsystems belonging to different sites can communicate with each other through the Internet Protocol (Internet Protocol). protocol, IP), remote direct data access (remote direct memory access, RDMA) protocol, fiber channel (fibre channel, FC) for communication, or data communication can be carried out through other methods.
  • Internet Protocol Internet Protocol
  • protocol, IP remote direct data access
  • RDMA remote direct memory access
  • FC fiber channel
  • This embodiment provides This is not limiting.
  • Each subsystem can include one or more devices to provide corresponding data computing or storage capabilities.
  • each subsystem includes multiple computing devices, multiple storage devices, or both computing devices and storage devices.
  • the multiple subsystems in the data processing system 100 may be multiple storage arrays, which may be used to support the data storage type services and data reading types requested by the host. business etc.
  • multiple subsystems in the data processing system 100 can be different storage domains in the same storage array, such as storage domains that can be divided according to tenants, or storage domains divided according to services (file domain, block domain, object domain, etc.) .
  • the subsystem 101 and the subsystem 102 in the data processing system 100 can operate based on the AP mode and provide services, such as data storage services, data computing services, etc., for the hosts 201 in the host cluster 200.
  • the host cluster 200 Host 202 in can be used as disaster recovery for host 201.
  • the requirements of the host 201 and the host 202 for the working mode of the data processing system 100 may change.
  • the tenant P may expect the data processing system 100 to operate in AP mode, such as only allowing data A subsystem in the processing system 100 performs data communication and the like with the host 201 of the tenant P.
  • tenant P may require the data processing system 100 to operate in an active-active (AA) mode to facilitate the change of tenant P's primary and secondary identities.
  • AA active-active
  • Both the host 201 and the host 202 can request services from the data processing system.
  • the data processing system 100 only supports running the AP mode, it will be difficult to adapt to changes in the tenant P's demand for the working mode of the data processing system 100 .
  • the data processing system 100 only supports running the AA mode, when tenant P requires the working mode of the data processing system 100 to be the AP mode, it is also difficult to adapt to the change in the tenant P's demand for the working mode of the data processing system 100 .
  • the data processing system 100 can support switching of the working mode between AP mode and AA mode.
  • a data processing device 400 can be configured.
  • the data processing device 400 includes a receiving unit 401 and a processing unit 402, and the data processing device can be connected with the subsystem 101 and the subsystem 101.
  • System 102 communicates.
  • the receiving unit 401 can receive a configuration instruction, which may be issued by a tenant or an administrator, for example, and provide the configuration instruction to the processing unit 402, so that the processing unit 402 can connect the subsystem 101 to the subsystem according to the configuration instruction.
  • the working mode between 102 is to switch between AP mode and AA mode.
  • the data processing system 100 can automatically switch the working mode between the subsystem 101 and the subsystem 102 according to the changes in the needs of the tenant's host, so that the data processing system 100 can adapt to the changes in the needs of the tenant's host and improve the tenant's use of the data processing system. 100% experience.
  • the data processing system 100 may be, for example, a system that provides storage services, a system that provides computing services, or other types of systems.
  • a data storage system as an example to introduce in detail the process of the data storage system operating in AP mode and operating in AA mode to provide business services to the host.
  • the subsystem 101 belonging to site A may include one or more controllers, and the controller may provide multiple external ports for communicating with the host through the multiple ports.
  • Hosts 201 in cluster 200 communicate.
  • controller 0 in subsystem 101 can provide port No. 0 and port No. 1
  • controller 1 can provide port No. 2 and port No. 3
  • subsystem 101 can provide port No. 0 and port No. 2.
  • Perform data communication with the host 201 such as receiving service requests issued by the host 201, etc.
  • any number of ports in the subsystem 101 can be allocated to the host 201, which is not limited in this embodiment.
  • the subsystem 102 belonging to site B includes controller 2 and controller 3, and port 0' in controller 2 and port 2' in controller 3 can be assigned to host 201 in case subsystem 101 fails.
  • the host 201 can access the subsystem 102 based on the port 0' or the port 2'.
  • the host cluster 200 may also include a host 202, which can be used as a hot backup (ie, online backup) of the host 201, and port 0 of the controller 0 and port 2 of the controller 1 in the subsystem 101 Port number 202 can be assigned to host.
  • the host 202 when the host 201 fails, the host 202 continues to access the subsystem 101 through port 0 and port 2 to take over the business on the host 201, thereby improving the reliability of the business on the host. Further, the host 202 can also be assigned to the port 0' in the controller 2 of the subsystem 102 and the port 2' in the controller 3, so that when the subsystem 101 fails, the host 202 can use the port 0'. , or port 2′ to access subsystem 102.
  • site A can serve as the main site that provides services to the host 201, and the host 201 can only request business services from site A.
  • the subsystem 101 included in site A may be provided with a file system (FS), and the file system has a root directory, and the root directory may be stored in the memory of the subsystem 101.
  • the subsystem 101 can mount the file system to the host 201 and provide the root directory of the file system to the host 201. In this way, the host 201 can access the file system in the subsystem 101 according to the root directory of the file system and read the data in the file system.
  • the subsystem 101 can create a subdirectory under the root directory and save the data provided by the host 201 in the subdirectory in a file format.
  • the subsystem 101 can be configured with multiple IP addresses, and when the host 201 accesses the subsystem 101, the subsystem 101 can select one or more IP addresses from the multiple IP addresses and assign them. to the host 201 so that the host 201 can access the file system in the subsystem 101 based on the assigned IP address.
  • each IP address can be used to access all files in each subdirectory in the file system, and can support the host 201 to create a subdirectory in the file system to subsequently store files in the subdirectory.
  • the host 201 uses the assigned IP address 1 to create subdirectory d5 in the file system, uses the assigned IP address 2 to create subdirectory d7 in the file system, and so on.
  • the subsystem 101 can be configured with a preset number of shards (shards), such as 4096 shards (ie, shard 0 to shard 4095) are configured in Figure 2. These shards are used to combine the file system. Directories and files are distributed and stored in this subsystem Different physical nodes in it can be used to control the distribution and access of data. Each physical node can be used to store one or more sharded data. Based on the data distribution of multiple shards, a shard view (or a data distribution view) can be generated, and the shard view can indicate the distribution of data on different physical nodes.
  • shards such as 4096 shards (ie, shard 0 to shard 4095) are configured in Figure 2. These shards are used to combine the file system. Directories and files are distributed and stored in this subsystem Different physical nodes in it can be used to control the distribution and access of data. Each physical node can be used to store one or more s
  • the subsystem 101 can periodically store data in the memory persistently. Specifically, it can deliver data such as directories, files, and fragmented views in the memory to persistent storage.
  • the storage medium is stored, and the corresponding controller supports data reading and writing to the persistent storage medium. In this way, the subsystem 101 can continue to receive new data written by the host 201 based on the released memory space.
  • Subsystem 102 included in site B can also be configured with a file system, and the data in the file system of subsystem 102 (including directories, files under directories, and shard views, etc.) can be Data synchronization is completed from the file system in subsystem 101.
  • the tenant when the tenant configures the IP address of the host 201 to access the subsystem, the tenant can configure the IP address of the host 201 to access the subsystem 102.
  • the subsystem 102 in site B can take over the business on the host 201 based on the target, file, shard view and other data stored on it, so that it can continue to provide business services to the host 201 .
  • FIG 3 a schematic diagram is shown in which subsystem 102 continues to provide business services after subsystem 101 fails.
  • the subsystem 102 can periodically monitor the heartbeat of the subsystem 101, specifically, it can receive heartbeat messages sent by the subsystem 101.
  • the subsystem 102 may determine that the subsystem 101 has failed.
  • the subsystem 102 may also determine that the subsystem 101 has failed after receiving a notification message sent by the arbitration server (not shown in FIG. 3 ). This embodiment is not limited to this.
  • the subsystem 102 can drift the IP address of the host 201 to access the subsystem 101 to the subsystem 102, and enable the port 0' assigned to the host 201 to support the host 201 to access the subsystem 102. Or port 2', so that the host 201 can access the data in the subsystem 102 based on the IP address and port. Since when the subsystem 101 is running normally, in the process of providing business services to the host 201, it will not only save the new data generated locally, but also synchronize the new data to the subsystem 102. Therefore, in the subsystem 102 The data in the memory and the persistent storage medium can be consistent with the memory in the subsystem 101 and the data in the persistent storage medium, so the subsystem 102 can continue to provide business services to the host 201 based on these data.
  • FIG. 4 there is shown a schematic diagram of a subsystem in the data storage system 100 provided by the embodiment of the present application operating based on the AA mode.
  • the subsystem 101 belonging to site A may include one or more controllers, and the controller may provide multiple external ports for communicating with the host through the multiple ports.
  • Hosts 201 in cluster 200 communicate.
  • controller 0 in subsystem 101 can provide port No. 0 and port No. 1
  • controller 1 can provide port No. 2 and port No. 3
  • subsystem 101 can provide port No. 0 and port No. 2.
  • Perform data communication with the host 201 such as receiving service requests issued by the host 201, etc.
  • the subsystem 102 belonging to site B includes controller 2 and controller 3, and port 1' in controller 2 and port 3' in controller 3 can communicate with the host 202 in the host cluster 200. communication.
  • the host 202 can be used as a hot backup (ie, online backup) of the host 201 to improve the reliability of services on the host.
  • port 0' and port 2' in the subsystem 102 can also be allocated to the host 201, so that when the subsystem 101 fails, the host 201 accesses the subsystem 102 through port 0' and port 2' to continue. Request business services.
  • port No. 1 and port No. 3 in the subsystem 101 can also be assigned to the host 202, so that when the subsystem 102 fails, the host 202 can Port No. 3 and port No. 3 access the subsystem 101 to continue requesting business services.
  • both the host 201 and the host 202 can request the site to provide business services. Specifically, host 201 can request site A to provide services, and host 202 can request site B to provide services. In this way, while utilizing the subsystem 102 for disaster recovery to meet disaster recovery requirements, the subsystem 102 provides a part of business services to the host, which can improve the overall service capability and system resource utilization of the data processing system 100 .
  • the subsystem 101 included in site A may be provided with a file system, and the file system has a root directory, and the root directory may be stored in the memory of the subsystem 101.
  • the subsystem 101 can mount the file system to the host 201 and provide the root directory of the file system to the host 201. In this way, the host 201 can access the file system in the subsystem 101 according to the root directory of the file system and read the data in the file system.
  • the subsystem 101 can create a subdirectory under the root directory and save the data provided by the host 201 in the subdirectory in a file format.
  • Subsystem 102 in site B is also provided with a file system, and the file system in subsystem 102 and the file system in subsystem 101 may have a unified root directory.
  • the host 202 can access the file system in the subsystem 101 and read and write data in the file system.
  • host 201 and host 202 can be assigned the same multiple IP addresses.
  • the IP address is used to support host 201/host 202 to access all files in each subdirectory in the file system, and can create corresponding directories in the file system to subsequently store files in the created subdirectories, such as host 201
  • the assigned IP address a can be used to create subdirectory d5 in the file system, and the host 202 can use the assigned IP address b to create subdirectory d7 in the file system, etc.
  • the home sites of host 201 and host 202 can also be configured respectively.
  • host 201 is configured to belong to site A
  • host 202 is configured to belong to site B
  • host 201 and host 202 give priority to the file system of the home site.
  • the data in the host is accessed to avoid data transmission between different sites for the files requested by the host as much as possible, and a part of the IP addresses on each of the two sites are in working condition.
  • Different shards may be configured in subsystem 101 and subsystem 102 respectively.
  • 2048 shards are configured in subsystem 101 and subsystem 102 respectively.
  • the shards in subsystem 101 are shards 0 to 2047
  • the shards in subsystem 102 are shards.
  • These shards are used to store the directory and file distribution of the file system to different physical nodes in the subsystem, and can be used to control the distribution and access of data.
  • Each physical node can be used to store one or more sharded data. .
  • a shard view can be generated, and the shard view can indicate the distribution of data on different physical nodes.
  • the split view is evenly distributed on the subsystem 101 and the subsystem 102 as an example. In actual application, the split view may not be evenly distributed on the two subsystems.
  • subsystem 101 and the subsystem 102 can periodically persistently store data in the memory. Since subsystem 101 and subsystem 102 respectively store data in some shard views, subsystem 101 can send the data in shards 0 to 2047 before persisting the data in the memory. To the subsystem 102, the subsystem 102 can send the data in slices 2048 to 4095 to the subsystem 101. In this way, both subsystem 101 and subsystem 102 can obtain complete 4096 fragmented data.
  • subsystem 101 and subsystem 102 can respectively deliver the data in all shards (such as complete directories, files, shard views, etc.) to the persistent storage medium for storage, and be supported by the corresponding controller. Read and write data to persistent storage media. In this way, the subsystem 101 and the subsystem 102 can continue to receive new data written by the host 201 and the host 202 based on the released memory space.
  • shards such as complete directories, files, shard views, etc.
  • subsystem 101 or subsystem 102 may fail during operation. In this case, you can Another subsystem that is running normally takes over the services on the failed subsystem. Taking the failure of subsystem 101 in site A as an example, see FIG. 5 , which shows a schematic diagram of subsystem 102 continuing to provide business services.
  • subsystem 101 and subsystem 102 can monitor the heartbeat of the peer end at the same time.
  • the time interval since the last time subsystem 102 received a heartbeat message from subsystem 101 will usually exceed the preset interval.
  • the subsystem 102 may determine that the subsystem 101 has failed.
  • the subsystem 101 may also determine that the subsystem 101 has failed after receiving a notification message sent by the arbitration server (not shown in FIG. 5 ). This embodiment is not limited to this.
  • the subsystem 102 can request the subsystem 101 to provide the data in shards 0 to 2047 stored in the memory to the subsystem 102. In this way, the shard view on the subsystem 102 can be updated as Fragment 0 ⁇ Fragment 4095.
  • the subsystem 102 can drift the IP address of the host 201 accessing the subsystem 101 to the subsystem 102, and enable port 0' and port 2' to receive the service request issued by the host 201. In this way, when the host 201 requests access to a file in the file system, the subsystem 102 can feed back the corresponding files stored in slices 0 to 4095 to the host 201, thereby continuing to provide business services to the host 201.
  • the subsystem 101 and the subsystem 102 in the data processing system 100 The working mode is supported to switch between AP mode and AA mode to meet the requirements of the host cluster for the working mode between subsystem 101 and subsystem 102 and improve the universality of business services provided by the data processing system 100.
  • the data processing system 100 can support the working mode of the subsystem 101 and the subsystem 102 to be switched from the AP mode to the AA mode, and can be switched from the AA mode to the AP mode.
  • the data processing device 400 in the data processing system 100 can switch the working mode between the subsystem 101 and the subsystem 102 .
  • the receiving unit 401 may receive a configuration instruction, which is used to instruct the data processing device 400 to switch the working mode between the subsystem 101 and the subsystem 102 from AP mode to AA mode, or from AA mode to AP. model.
  • this embodiment provides the following implementation methods for receiving configuration instructions.
  • the data processing device 400 can provide a client to the outside, so that when the tenant or administrator determines that it is necessary to switch the working mode between the subsystem 101 and the subsystem 102, the data processing device 400 can use the client to request data processing.
  • the device 400 issues a configuration instruction to instruct the data processing device 400 to switch the working mode between the subsystem 101 and the subsystem 102 from the current mode to another mode, and the receiving unit 401 receives the configuration instruction.
  • the data processing device 400 can detect the networking mode of the host currently connected to the data processing system 100.
  • the networking mode of the host can be a parallel networking or a full-crossover networking.
  • parallel networking means that cross-connections do not need to be established between different hosts in the network. For example, data communication channels do not need to be directly established between different hosts; and, when hosts using parallel networking are connected to the data processing system
  • the subsystem 101 and the subsystem 102 usually operate based on the AP mode to provide corresponding services to the host.
  • Full-crossover networking means that connections can be established between different hosts in the network, so that different hosts can communicate with each other through the communication connection; and, when a host using full-crossover networking is connected to the data processing system 100 , subsystem 101 and subsystem 102 usually run based on AA mode to provide corresponding services to the host.
  • the networking mode of the host may also be other modes, which is not limited in this embodiment.
  • the receiving unit 401 can determine whether the networking mode of the host currently connected to the data processing system 100 matches the working mode between the subsystem 101 and the subsystem 102 by detecting the networking mode of the host. If it does not match, , then the receiving unit 401 can generate a configuration instruction to trigger the working mode between the subsystem 101 and the subsystem 102 switch. If they match, the data processing device 400 does not need to switch the working mode between the subsystem 101 and the subsystem 102 .
  • the receiving unit may also obtain the configuration instructions based on other methods, which is not limited in this embodiment.
  • the receiving unit 401 may send the configuration instruction to the processing unit 402, so that the processing unit 402 performs the operation of switching the working mode between the subsystem 101 and the subsystem 102.
  • FIG. 6 a schematic diagram of switching the working mode of subsystem 101 and subsystem 102 from AP mode to AA mode is shown.
  • the subsystem 102 can support the host 202 requesting services from the subsystem 102 .
  • the processing unit 402 can instruct the subsystem 102 in site B to allocate port 1' in controller 2 and port 3' in controller 3 to the host 202, so that the host 202 can pass the port 1'.
  • Port No. or port No. 3′ accesses the file system in subsystem 102.
  • the processing unit 402 may instruct the subsystem 101 to migrate data in some shards to the subsystem 102, and update the shard views in the subsystem 101 and the subsystem 102 respectively. For example, assuming that subsystem 101 migrates data in shards 2048 to 4095 to subsystem 102, the updated shard views in subsystem 101 and subsystem 102 are as shown in Figure 6 .
  • the processing unit 402 can instruct the subsystem 102 to allocate an IP address for the host 202 to access files in the subdirectory. Further, the subsystem 102 can also designate the home site of the host 202 as site B. In this way, when the host 202 accesses files under the subdirectory (such as accessing files under the subdirectories d6 and d7, etc.), the host 202 can give priority to the home site of the host 202. Subsystem 102 in site B accesses and can obtain the data it needs in subsystem 102 according to the shard view on subsystem 102 (such as shard 2048 to shard 4095).
  • the host 202 accesses the subsystem 101 and causes data transmission of files between different sites.
  • the subsystem 101 can designate the home site of the host 201 as site A, so that the host 201 preferentially accesses the subsystem 101 in site A.
  • the working mode of the subsystem 101 and the subsystem 102 can be switched from the AP mode to the AA mode.
  • FIG. 7 a schematic diagram of switching the working mode of subsystem 101 and subsystem 102 from AA mode to AP mode is shown.
  • subsystem 102 when the working modes of subsystem 101 and subsystem 102 are switched to AP mode, subsystem 102 does not support the host 202 requesting services from subsystem 102 .
  • the processing unit 402 can instruct the subsystem 102 in site B to allocate port 0' and port 2' to the host 201, so that when the subsystem 101 fails, the host 201 can use the port 0'. Or port 2′ to access the file system in subsystem 102. Since in AP mode, the data processing system 100 only supports the host 201 to request services from the site, the subsystem 102 can disable the IP address previously assigned to the host 202 for accessing files in the subdirectory.
  • the processing unit 402 can instruct the subsystem 101 backs up the data in all shards to subsystem 102.
  • Subsystem 102 can back up the data in all shards to subsystem 101, and updates the shard views in subsystem 101 and subsystem 102 respectively.
  • subsystem 101 and subsystem 102 are fragment 0 to fragment 2047 and fragment 2048 to fragment 4095 respectively, then when switching to AP mode, subsystem 101 and subsystem 102 After systems 102 back up sharded data with each other, subsystems 101 and The updated shard view in subsystem 102 is shown in Figure 7.
  • the working modes of different subsystems in the data processing system 100 can be flexibly switched between the AP mode and the AA mode to meet the needs of actual application scenarios.
  • system architecture of the data processing system 100 shown in FIG. 1 is only an example, and is not intended to limit its specific implementation to this example.
  • the data processing system 100 shown in Figure 8 in addition to subsystem 101 and subsystem 102, it can also include subsystem 103 and subsystem 104, and both subsystem 101 and subsystem 103 belong to site A.
  • subsystem 102 and subsystem 104 both belong to site B.
  • subsystems 101 to 104 can be used to support simultaneous provision of services corresponding to the AP mode and the AA mode.
  • the subsystem 101 and the subsystem 102 can operate based on the AP mode, and are used to provide services corresponding to the AP mode for the host 201 and the host 202 in the host cluster 200; the subsystem 103 and the subsystem 104 can operate based on the AA mode, using To provide services corresponding to the AA mode for the host 301 and the host 302 in the host cluster 300 .
  • the data processing system 100 shown in Figure 9 is similar to the data processing system 100 shown in Figure 8 and also includes subsystems 101 to 104, and simultaneously supports the provision of services corresponding to the AP mode and the AA mode. .
  • subsystems 101 to 104 belong to different sites respectively. Specifically, subsystem 101 belongs to site A, subsystem 102 belongs to site B, and subsystem 102 belongs to site B. 103 belongs to site C, and subsystem 104 belongs to site D.
  • This application does not limit the specific architecture of the data processing system.
  • the data processing system 100 can use the subsystem 101 and the subsystem 102 to process the services requested by the tenant's host;
  • the data processing system 100 can use subsystem 103 and subsystem 104 to process services requested by the tenant's host. Since different subsystems in the data processing system 100 run AP mode and AA mode respectively, the data processing system 100 can select the subsystem of the corresponding working mode according to the tenant's needs to provide business services for the tenant's host, so as to meet the actual requirements.
  • the business needs of different hosts in application scenarios improve the universality of business services provided by the data processing system 100.
  • the above-mentioned data processing system 100 shown in FIGS. 2 to 9 is used to provide data storage services for tenants and the data storage service is specifically a file storage service. That is, the data processing system 100 stores The data may be saved in the form of files, and in other possible implementations, the data storage service provided by the data processing system 100 may also be a block storage service or an object storage service.
  • the block storage service means that when the data processing system 100 stores data, the data is divided into blocks according to a fixed size. The data amount of each block may be, for example, 512 bytes or 4 kilobytes (KB). wait.
  • the object storage service means that the data processing system 100 stores data as objects.
  • the object can be the basic unit of data storage in the data processing system 100, and each object can be a complex of data and data attributes.
  • This data attribute can be set according to the needs of the application, including data distribution, service quality, etc.
  • the data processing system 100 may also provide other types of services, such as data calculation services, model training services, data statistics services, etc., which are not limited in this embodiment.
  • the data processing system 100 shown in FIGS. 1 to 9 may be suitable for centralized storage application scenarios or distributed storage application scenarios.
  • each subsystem can be composed of one or more computing nodes to form a central node, and all data processing services of the entire subsystem can be deployed centrally on this central node.
  • a disk-control-separated architecture can be used between computing nodes and storage devices, that is, computing nodes and storage devices are deployed independently; or, a disk-control integrated architecture can be used between computing nodes and storage devices, that is, computing nodes can have slots.
  • the storage device is placed in the computing node through the slot and deployed integrated with the computing node.
  • the data in each subsystem can be distributed and stored on multiple independent storage nodes.
  • the computing node can be integrated and deployed with the storage device, so that the computing node has both computing capabilities and storage capabilities, and a virtual machine can be created on the computing node, or a virtual machine does not need to be created.
  • a storage-computing separation architecture can be adopted between computing nodes and storage devices, that is, computing nodes and storage devices are deployed independently and communicate through the network.
  • the storage device may include one or more different storage media, which is not limited in this embodiment.
  • the administrator can configure the tenant's host to access a subsystem running in AP mode or access a subsystem running in AA mode according to the needs of the tenant.
  • the data processing system 100 can also automatically connect the host to a suitable subsystem according to the host's networking mode.
  • FIG 10 it is a schematic flowchart of a data processing method in an embodiment of the present application.
  • This method can be applied to the data processing system 100 shown in Figure 1, Figure 8, or Figure 9. In practical applications, this method can also be applied to other applicable data processing systems.
  • the following is an example of an application to the data processing system 100 shown in Figure 8.
  • the method may specifically include:
  • the receiving unit 401 obtains the networking form of the host connected to the data processing system 100.
  • the networking form of the host may be a parallel networking or a full-crossover networking, or other types of networking.
  • parallel networking means that cross-connections do not need to be established between different hosts in the network.
  • data communication channels do not need to be directly established between different hosts.
  • Full cross-connection networking means that connections can be established between different hosts in the network, so that different hosts can communicate with each other through this communication connection.
  • S902 When the networking form of the host is parallel networking, the processing unit 402 uses the subsystem 101 and the subsystem 102 in the data processing system 100 to process the service requested by the host.
  • the working mode is AP mode.
  • S903 When the networking form of the host is a full-crossover network, the processing unit 402 uses the subsystem 103 and the subsystem 104 in the data processing system 100 to process the service requested by the host. Between the subsystem 103 and the subsystem 104 The working mode is AA mode.
  • a host with parallel networking can be connected to the subsystem 101 running in AP mode, and the access port on the subsystem 102 and the access port on the subsystem 102 can be automatically configured for the host. IP address, so that after the subsystem 101 fails, the host can access the subsystem 102 to achieve uninterrupted service processing.
  • the working mode of the subsystem connected to the host can be automatically switched, for example, when the host requires the working mode of the subsystem 101
  • the subsystem 101 can automatically switch the working modes of the subsystem 101 and the subsystem 102 to the AA mode based on the method shown in Figure 6 above to meet the actual application needs of the host.
  • the subsystem connected to the host can automatically perform the switching process of the working mode, or the working mode of the subsystem can be automatically switched under the control of the arbitration server or other devices, which is not limited in this embodiment. .
  • the data processing device 1100 is applied to a data processing system, such as the data processing system 100 shown in Figures 1 to 10, etc.
  • the data processing system includes a first subsystem and a second subsystem, such as the above subsystem.
  • System 101, subsystem 102, etc., the data processing device 100 shown in Figure 11 includes:
  • the processing unit 1102 is configured to switch the first subsystem and the second subsystem from the first mode to the second mode according to the configuration instruction; wherein the first mode is when the first subsystem and the second subsystem work in One of the active-passive AP mode and the active-active AA mode; the second mode is another mode of the AP mode and the AA mode working between the first subsystem and the second subsystem.
  • the first subsystem and the second subsystem provide data processing services for the host of the target tenant based on the first mode
  • the connection between the first subsystem and the second subsystem is After the working mode is switched to the second mode, the first subsystem and the second subsystem provide data processing services for the host of the target tenant based on the second mode.
  • the receiving unit 1101 is specifically configured to detect the networking mode of the host connected to the data processing system, and switch from the first networking mode to the second networking mode.
  • the first networking mode is one of a parallel networking mode and a full-crossover networking mode
  • the second networking mode is one of a parallel networking mode and a full-crossover networking mode. Another networking mode.
  • the data processing system further includes a third subsystem and a fourth subsystem.
  • the third subsystem and the fourth subsystem work in the first mode.
  • the first subsystem and the third subsystem belong to the first site, and the second subsystem and the fourth subsystem belong to the second site.
  • the first subsystem and the second subsystem are used to provide data storage services for target tenants, and the first subsystem includes at least one storage array.
  • the first subsystem and the second subsystem communicate through at least one of IP protocol, RDMA protocol, and FC.
  • the data storage service provided by the first subsystem and the second subsystem is a file storage service.
  • the data storage service provided by the first subsystem and the second subsystem is a block storage service.
  • the data storage service provided by the first subsystem and the second subsystem is an object storage service.
  • the data processing device 1100 shown in Figure 11 corresponds to the data processing device 400 in the previous embodiment. Therefore, the specific implementation of the data processing device 1100 and its technical effects can be found in the relevant descriptions of the previous embodiments. This will not be described in detail.
  • the computing device 1200 may include a communication interface 1210 and a processor 1220.
  • the computing device 1200 may also include a memory 1230.
  • the memory 1230 may be disposed inside the computing device 1200 or may be disposed outside the computing device 1200 .
  • each action performed by the data processing device 400 in the above embodiments shown in FIGS. 1 to 10 can be implemented by the processor 1220.
  • the processor 1220 can obtain configuration instructions through the communication interface 1210 and be used to implement the method described in FIG. 9 above.
  • each step of the processing flow can complete the method described in Figure 9 through instructions in the form of hardware integrated logic circuits or software in the processor 1220.
  • the processor 1220 is used to implement the program code executed by the above method.
  • the code may be stored in memory 1230.
  • the memory 1230 and the processor 1220 are connected, such as coupling connection, etc.
  • Some features of the embodiments of the present application may be implemented/supported by the processor 1220 executing program instructions or software codes in the memory 1230.
  • the software components loaded on the memory 1230 may be summarized functionally or logically, for example, the processing unit 1102 shown in FIG. 11 .
  • the function of the receiving unit 1101 can be implemented by the communication interface 1210.
  • Any communication interface involved in the embodiments of this application may be a circuit, bus, transceiver, or any other device that can be used for information exchange.
  • the communication interface 1210 in the computing device 1200 for example, the other device may be a device connected to the computing device 1200, or the like.
  • the processor involved in the embodiments of this application may be a general-purpose processor, a digital signal processor, an application-specific integrated circuit, a field programmable gate array or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component, which may implement or Execute each method, step and logical block diagram disclosed in the embodiment of this application.
  • a general-purpose processor may be a microprocessor or any conventional processor, etc.
  • the steps of the methods disclosed in conjunction with the embodiments of the present application can be directly implemented by a hardware processor for execution, or can be executed by a combination of hardware and software units in the processor.
  • the coupling in the embodiment of this application is an indirect coupling or communication connection between devices, units or units, which may be in electrical, mechanical or other forms, and is used for information interaction between devices, units or units.
  • the processor may operate in conjunction with the memory.
  • the memory can be a non-volatile memory, such as a hard disk or a solid state drive, or a volatile memory, such as a random access memory.
  • Memory is, but is not limited to, any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer.
  • the embodiments of the present application do not limit the specific connection medium between the above communication interface, processor and memory.
  • the memory, processor and communication interface can be connected through a bus.
  • the bus can be divided into address bus, data bus, control bus, etc.
  • embodiments of the present application also provide a computer storage medium, which stores a software program.
  • the software program can implement any one or more of the above.
  • the embodiment provides a method performed by the data processing apparatus 400.
  • the computer storage medium may include: U disk, mobile hard disk, read-only memory, random access memory, magnetic disk or optical disk and other various media that can store program codes.
  • embodiments of the present application also provide a chip, which includes a processor and is used to implement the functions of the data processing device 400 involved in the above embodiments, for example, to implement the method executed in FIG. 9 .
  • the chip further includes a memory, and the memory is used for necessary program instructions and data executed by the processor.
  • the chip may be composed of chips or may include chips and other discrete devices.
  • embodiments of the present application may be provided as methods, systems, or computer program products. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment that combines software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
  • computer-usable storage media including, but not limited to, disk storage, CD-ROM, optical storage, etc.
  • These computer program instructions may also be stored in a computer-readable memory that causes a computer or other programmable data processing apparatus to operate in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including the instruction means, the instructions
  • the device implements the functions specified in a process or processes of the flowchart and/or a block or blocks of the block diagram.
  • These computer program instructions may also be loaded onto a computer or other programmable data processing device, causing a series of operating steps to be performed on the computer or other programmable device to produce computer-implemented processing, thereby executing on the computer or other programmable device.
  • Instructions provide steps for implementing the functions specified in a process or processes of a flowchart diagram and/or a block or blocks of a block diagram.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

提供一种数据处理系统,该数据处理系统包括第一子系统、第二子系统、接收单元和处理单元;接收单元,用于接收配置指令,处理单元,用于根据接收单元所接收到的配置指令,将第一子系统和第二子系统之间由第一模式切换为第二模式,该第一模式为第一子系统与第二子系统之间工作在AP模式和AA模式中的一种模式,第二模式为第一子系统与第二子系统之间工作在AP模式和AA模式中的另一种模式。如此,数据处理系统可以自动切换第一子系统与第二子系统之间的工作模式,以此实现数据处理系统能够适应租户的需求变化,从而可以有效提高租户使用数据处理系统的体验。此外,还提供了相应的数据处理方法、装置及相关设备。

Description

数据处理系统、数据处理方法、装置及相关设备 技术领域
本申请涉及数据处理技术领域,尤其涉及一种数据处理系统、数据处理方法、装置及相关设备。
背景技术
双活数据处理系统,是指数据处理系统中的两个子系统均处于运行状态,并且,两个子系统中的数据保持一致。这样,当一个子系统因火灾、洪水等情况下无法提供业务服务时,可以由该数据处理系统中的另一个子系统及时接管业务,从而实现业务服务不中断、数据不丢失,进而提高数据处理系统的可靠性。
目前,数据处理系统通常采用活动-消极(active-passive,AP)模式运行两个子系统,其中,只有处于活动(active)状态的子系统能够给主机提供业务服务(如数据读写服务等),处于消极(passive)状态的子系统作为容灾端,通常会实时备份活动状态的子系统中的数据,并在数据处理系统未发生异常时不提供业务服务。
但是,实际应用场景中,主机可能会要求对数据处理系统中的两个子系统均能进行访问,这使得基于AP模式运行的数据处理系统难以满足该主机的需求,从而影响租户使用该数据处理系统的体验。
发明内容
提供一种数据处理系统、数据处理方法、装置、计算设备、存储介质以及计算机程序产品,以使得数据处理系统能够满足主机对于访问子系统的需求变化,提高租户使用数据处理系统的体验。
第一方面,本申请实施例提供一种数据处理系统,该数据处理系统包括第一子系统以及第二子系统,并且,该数据处理系统还包括接收单元和处理单元;其中,接收单元,用于接收配置指令,例如可以接收管理员下发的配置指令,或者可以检测到接入数据处理系统的主机的组网模式发生变化等,处理单元,用于根据接收单元所接收到的配置指令,将第一子系统和第二子系统之间由第一模式切换为第二模式,该第一模式为第一子系统与第二子系统之间工作在AP模式和AA模式中的一种模式,第二模式为第一子系统与第二子系统之间工作在AP模式和AA模式中的另一种模式,如可以将第一子系统和第二子系统之间由AA模式切换为AP模式,或者由AP模式切换为AA模式等。其中,AA模式中第一子系统与第二子系统均可以为主机提供数据处理服务,AP模式中尽可以由一个子系统(处于active状态的子系统)为主机提供数据处理服务。
如此,数据处理系统可以根据租户的需求变化(也即该租户的主机所匹配的两个子系统工作模式发生变化),自动切换第一子系统与第二子系统之间的工作模式,以此实现数据处理系统能够适应租户的需求变化,即在租户需求AA模式时,数据处理系统可以将两个子系统的工作模式切换为租户所需的AA模式,在租户再次需求AP模式时,数据处理系统可以再将这两个子系统的工作模式切换为AP模式,从而可以有效提高租户使用数据处理系统的体验。
在一种可能的实施方式中,第一子系统与第二子系统之间基于第一模式为目标租户的主机提供数据处理服务,如数据存储服务、数据计算服务等,并且,在将第一子系统与第二子系统之间的工作模式切换为第二模式后,第一子系统与第二子系统之间基于第二模式为该目 标租户的主机提供数据处理服务。如此,可以根据目标租户的主机所需求的工作模式的变化,自适应切换工作模式以继续为目标租户的主机提供相应的数据处理服务,以此可以有效提高租户使用数据处理系统的体验。
在一种可能的实施方式中,接收单元,具体用于检测到接入数据处理系统的主机的组网模式,由第一组网模式切换为第二组网模式。此时,接收单元可以触发执行切换第一子系统与第二子系统的工作模式的过程,以便通过切换两个子系统的工作模式,来适应主机的组网模式的变化(主机不同的组网模式可能要求数据处理系统基于不同的工作模式提供数据处理服务)。
在一种可能的实施方式中,主机的第一组网模式可以是平行组网模式和全交叉组网模式中的一种组网模式,主机的第二组网模式为平行组网模式和全交叉组网模式中的另一种组网模式。如此,当主机的组网模式在平行组网模式和全交叉组网模式之间进行切换时,数据处理系统也可以自适应的切换两个子系统之间的工作模式,以适应主机组网模式的变化。
在一种可能的实施方式中,数据处理系统还包括第三子系统以及第四子系统,并且,在第一子系统与第二子系统之间工作在第二模式时,第三子系统与第四子系统之间工作在第一模式。如此,数据处理系统可以基于多个子系统,同时为不同的主机提供AA模式以及AP模式,这使得数据处理系统可以根据租户的需求选择相应运行模式的子系统为该租户的主机提供业务服务,以此可以满足实际应用场景中不同主机的业务需求,提高数据处理系统提供业务服务的普适性。
在一种可能的实施方式中,当数据处理系统同时包括第一子系统至第四子系统时,第一子系统与第三子系统属于第一站点,第二子系统与第四子系统属于第二站点。
可选地,第一子系统至第四子系统可以分别属于不同的站点。
在一种可能的实施方式中,第一子系统与第二子系统为目标租户提供的数据处理服务具体为数据存储服务,此时,第一子系统包括至少一个存储阵列,以便利用该存储阵列存储该目标租户的数据,实现提供数据存储服务。
在一种可能的实施方式中,第一子系统与第二子系统通过IP协议、RDMA协议、FC中的至少一种进行通信。
在一种可能的实施方式中,第一子系统与第二子系统提供的数据处理服务具体为数据存储服务,并且,该数据存储服务具体可以是文件存储服务。
在一种可能的实施方式中,第一子系统与第二子系统提供的数据处理服务具体为数据存储服务,并且,该数据存储服务具体可以是块存储服务。
在一种可能的实施方式中,第一子系统与第二子系统提供的数据处理服务具体为数据存储服务,并且,该数据存储服务具体可以是对象存储服务。
第二方面,本申请提供一种数据处理方法,该数据处理方法应用于数据处理系统,该数据处理系统包括第一子系统、第二子系统、接收单元和处理单元,则在进行数据处理时,接收单元接收配置指令,并将配置指令提供给处理单元;处理单元根据配置指令将第一子系统和第二子系统之间由第一模式切换为第二模式;其中,第一模式为第一子系统与第二子系统之间工作在AP模式和AA模式中的一种模式;第二模式为第一子系统与第二子系统之间工作在AP模式和AA模式中的另一种模式。如此,数据处理系统可以根据租户的需求变化,自动切换第一子系统与第二子系统之间的工作模式,以此实现数据处理系统能够适应租户的需求变化,从而可以有效提高租户使用数据处理系统的体验。
在一种可能的实施方式中,第一子系统与第二子系统之间基于第一模式为目标租户的主机提供数据处理服务,并且,在将第一子系统与第二子系统之间的工作模式切换为第二模式后,第一子系统与第二子系统之间基于第二模式为目标租户的主机提供数据处理服务。如此,可以根据目标租户的主机所需求的工作模式的变化,自适应切换工作模式以继续为目标租户的主机提供相应的数据处理服务,以此可以有效提高租户使用数据处理系统的体验。
在一种可能的实施方式中,接收单元接收配置指令,包括:接收单元检测到接入数据处理系统的主机的组网模式,由第一组网模式切换为第二组网模式。如此,接收单元可以通过检测主机的组网模式的变化,来触发执行切换第一子系统与第二子系统的工作模式的过程,以便通过切换两个子系统的工作模式,来适应主机的组网模式的变化
在一种可能的实施方式中,第一组网模式为平行组网模式和全交叉组网模式中的一种组网模式,第二组网模式为平行组网模式和全交叉组网模式中的另一种组网模式。如此,当主机的组网模式在平行组网模式和全交叉组网模式之间进行切换时,数据处理系统也可以自适应的切换两个子系统之间的工作模式,以适应主机组网模式的变化。
在一种可能的实施方式中,数据处理系统还包括第三子系统以及第四子系统,在第一子系统与第二子系统之间工作在第二模式时,第三子系统与第四子系统之间工作在第一模式。如此,数据处理系统可以基于多个子系统,同时为不同的主机提供AA模式以及AP模式,这使得数据处理系统可以根据租户的需求选择相应运行模式的子系统为该租户的主机提供业务服务,以此可以满足实际应用场景中不同主机的业务需求,提高数据处理系统提供业务服务的普适性。
在一种可能的实施方式中,第一子系统与第三子系统属于第一站点,第二子系统与第四子系统属于第二站点。实际应用时,第一子系统至第四子系统也可以分别属于不同的站点。
在一种可能的实施方式中,第一子系统与第二子系统用于为目标租户提供数据存储服务,第一子系统包括至少一个存储阵列。如此,数据处理系统可以利用子系统包括的存储阵列来存储该目标租户的数据,实现提供数据存储服务。
在一种可能的实施方式中,第一子系统与第二子系统通过IP协议、RDMA协议、FC中的至少一种进行通信。
在一种可能的实施方式中,第一子系统与第二子系统提供的数据存储服务为文件存储服务。
在一种可能的实施方式中,第一子系统与第二子系统提供的数据存储服务为块存储服务。
在一种可能的实施方式中,第一子系统与第二子系统提供的数据存储服务为对象存储服务。
第三方面,本申请实施例提供了一种数据处理装置,数据处理装置应用于数据处理系统,数据处理系统包括第一子系统以及第二子系统,数据处理装置包括:接收单元,用于接收配置指令;处理单元,用于根据配置指令将第一子系统和第二子系统之间由第一模式切换为第二模式;其中,第一模式为第一子系统与第二子系统之间工作在活动-消极AP模式和活动-活动AA模式中的一种模式;第二模式为第一子系统与第二子系统之间工作在AP模式和AA模式中的另一种模式。
在一种可能的实施方式中,第一子系统与第二子系统之间基于第一模式为目标租户的主机提供数据处理服务,并且,在将第一子系统与第二子系统之间的工作模式切换为第二模式后,第一子系统与第二子系统之间基于第二模式为目标租户的主机提供数据处理服务。
在一种可能的实施方式中,接收单元,具体用于检测到接入数据处理系统的主机的组网模式,由第一组网模式切换为第二组网模式。
在一种可能的实施方式中,第一组网模式为平行组网模式和全交叉组网模式中的一种组网模式,第二组网模式为平行组网模式和全交叉组网模式中的另一种组网模式。
在一种可能的实施方式中,数据处理系统还包括第三子系统以及第四子系统,在第一子系统与第二子系统之间工作在第二模式时,第三子系统与第四子系统之间工作在第一模式。
在一种可能的实施方式中,第一子系统与第三子系统属于第一站点,第二子系统与第四子系统属于第二站点。
在一种可能的实施方式中,第一子系统与第二子系统用于为目标租户提供数据存储服务,第一子系统包括至少一个存储阵列。
在一种可能的实施方式中,第一子系统与第二子系统通过网际互联协议IP、远程直接数据存取RDMA协议、光纤通道FC中的至少一种进行通信。
在一种可能的实施方式中,第一子系统与第二子系统提供的数据存储服务为文件存储服务。
在一种可能的实施方式中,第一子系统与第二子系统提供的数据存储服务为块存储服务。
在一种可能的实施方式中,第一子系统与第二子系统提供的数据存储服务为对象存储服务。
由于第三方面提供的数据处理装置,对应于第一方面提供的数据处理系统以及第二方面提供的数据处理方法,因此,第三方面以及第三方面中各实施方式所具有技术效果,可以参见相应的第一方面以及第一方面中各实施方式所具有的技术效果,或者可以参见相应的第二方面以及第二方面中各实施方式所具有的技术效果,在此不做赘述。
第四方面,本申请实施例提供一种计算设备,包括:处理器和存储器;该存储器用于存储指令,当该计算设备运行时,该处理器执行该存储器存储的该指令,以使该计算设备执行上述第二方面或第二方面的任一实现方式提供的数据处理方法。需要说明的是,该存储器可以集成于处理器中,也可以是独立于处理器之外。计算设备还可以包括总线。其中,处理器通过总线连接存储器。其中,存储器可以包括可读存储器以及随机存取存储器。
第五方面,本申请实施例还提供一种可读存储介质,所述可读存储介质中存储有程序或指令,当其在计算机上运行时,使得上述第二方面或第二方面的任一实现方式中提供的数据处理方法被执行。
第六方面,本申请实施例还提供一种包含指令的计算机程序产品,当其在计算机上运行时,使得计算机执行上述第二方面或第二方面的任一实现方式中提供的数据处理方法。
另外,第二方面至六方面中任一种实现方式所带来的技术效果可参见第一方面中不同实现方式所带来的技术效果,此处不再赘述。
附图说明
为了更清楚地说明本申请实施例中的技术方案,下面将对实施例描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本申请中记载的一些实施例,对于本领域普通技术人员来讲,还可以根据这些附图获得其他的附图。
图1为本申请实施例提供的一示例性数据处理系统的结构示意图;
图2为本申请实施例提供的数据存储系统100中的子系统基于AP模式运行的示意图;
图3为在AP模式下子系统101故障后由子系统102继续提供业务服务的示意图;
图4为本申请实施例提供的数据存储系统100中的子系统基于AA模式运行的示意图;
图5为在AA模式下子系统103故障后由子系统104继续提供业务服务的示意图;
图6为本申请实施例提供的两个子系统的工作模式由AP模式切换至AA模式的示意图;
图7为本申请实施例提供的两个子系统的工作模式由AA模式切换至AP模式的示意图;
图8为本申请实施例提供的另一示例性数据处理系统的结构示意图;
图9为本申请实施例提供的又一示例性数据处理系统的结构示意图;
图10为本申请实施例提供的一种数据处理方法的流程示意图;
图11为本申请实施例提供的一种数据处理装置的流程示意图;
图12为本申请实施例提供的一种计算设备的硬件结构示意图。
具体实施方式
参见图1,为一示例性数据处理系统的结构示意图。如图1所示,数据处理系统100可以包括多个子系统,图1中以包括子系统101以及子系统102为例进行示例性说明。实际应用时,数据处理系统100可以包括更多数量的子系统。
其中,子系统101属于站点(site)A,子系统102属于站点B,每个站点可以用于存储租户(或者租户)提供的文件,属于不同站点的子系统之间可以通过网际互联协议(internet protocol,IP)、远程直接数据存取(remote direct memory access,RDMA)协议、光纤通道(fibre channel,FC)中的至少一种进行通信,或者可以通过其它方式进行数据通信等,本实施例对此并不进行限定。
每个子系统,可以包括一个或者多个设备,用于提供相应的数据计算或者存储能力。如,每个子系统包括多个计算设备,或者包括多个存储设备,或者同时包括计算设备以及存储设备等。示例性地,当数据处理系统100具体为存储系统时,数据处理系统100中的多个子系统具体可以是多个存储阵列,可以用于支持主机所请求的数据存储类型的业务、数据读取类型的业务等。或者,数据处理系统100中的多个子系统可以是同一存储阵列中的不同存储域,如可以按照租户划分的存储域,或者是按照业务划分的存储域(文件域、块域、对象域等)。
如图1所示,数据处理系统100中的子系统101以及子系统102可以基于AP模式运行,并为主机集群200中的主机201提供服务,如数据存储服务、数据计算服务等,主机集群200中的主机202可以作为主机201的灾备。但是,实际应用场景中,主机201以及主机202对于数据处理系统100的工作模式的要求可能发生变化,比如,在初始状态下,租户P可能期望数据处理系统100采用AP模式运行,如仅允许数据处理系统100中的一个子系统与租户P的主机201进行数据通信等。但是,随着租户P的主机组网模式或者主机运行业务的变化,租户P可能要求数据处理系统100采用活动-活动(active-active,AA)模式运行,以便于租户P的主、备身份的主机201以及主机202均能向数据处理系统请求业务等。此时,数据处理系统100如果仅支持运行AP模式,则难以适应该租户P对于数据处理系统100的工作模式的需求变化。类似的,如果数据处理系统100仅支持运行AA模式,则当租户P要求数据处理系统100的工作模式为AP模式时,同样难以适应该租户P对于数据处理系统100的工作模式的需求变化。
基于此,本申请实施例提供的数据处理系统100可以支持工作模式在AP模式之间AA模式之间进行切换。具体地,在图1所示的数据处理系统100中,可以配置有数据处理装置400,该数据处理装置400包括接收单元401以及处理单元402,并且,该数据处理装置可以与子系统101以及子系统102进行通信。当租户的主机要求数据处理系统100切换工作模式时, 接收单元401可以接收配置指令,该配置指令例如可以是由租户或者管理员下发,并将该配置指令提供给处理单元402,从而处理单元402可以根据该配置指令,将子系统101与子系统102之间的工作模式在AP模式与AA模式之间进行切换。
如此,数据处理系统100可以根据租户主机的需求变化,自动切换子系统101与子系统102之间的工作模式,以此实现数据处理系统100能够适应租户主机的需求变化,提高租户使用数据处理系统100的体验。
实际应用时,数据处理系统100例如可以是提供存储服务的系统,或者可以是提供计算服务的系统,又或者是其它类型的系统等。为便于理解,下面以数据存储系统为例,对数据存储系统分别基于AP模式运行、基于AA模式运行为主机提供业务服务的过程进行详细介绍。
参阅图2,示出了本申请实施例提供的数据存储系统100中的子系统基于AP模式运行的示意图。如图2所示,在AP模式中,归属于站点A的子系统101中可以包括一个或者多个控制器,并且,该控制器可以对外提供多个端口,用于通过该多个端口与主机集群200中的主机201进行通信。如图2所示,子系统101中控制器0可以提供0号端口以及1号端口、控制器1可以提供2号端口以及3号端口,并且,子系统101可以通过0号端口以及2号端口与主机201进行数据通信,如接收主机201下发的业务请求等。实际应用时,可以将子系统101中的任意数量的端口分配给主机201,本实施例对此并不进行限定。
归属于站点B的子系统102中包括控制器2和控制器3,并且,控制器2中的0′号端口以及控制器3中的2′号端口可以分配给主机201,以便子系统101故障时,主机201可以基于该0′号端口或者2′号端口访问子系统102。实际应用时,主机集群200中的还可以包括主机202,该主机202可以作为主机201的热备(即在线备份),并且,子系统101中控制器0的0号端口、控制器1的2号端口可以分配给主机202。这样,在主机201发生故障时,主机202通过0号端口以及2号端口继续访问子系统101,以实现接管主机201上的业务,以此提高主机上的业务的可靠性。进一步地,主机202还可以分配到子系统102的控制器2中的0′号端口以及控制器3中的2′号端口,用于在子系统101故障时,主机202能够利用0′号端口、或者2′号端口访问子系统102。
当数据处理系统100正常运行时,站点A可以作为向主机201提供服务的主站点,并且,主机201仅能向站点A请求提供业务服务。
站点A包括的子系统101中可以设置有文件系统(file system,FS),并且,该文件系统具有根目录,该根目录可以保存在子系统101的内存中。当主机201与子系统101建立通信连接后,子系统101可以将该文件系统挂载至主机201,并将该文件系统的根目录提供给主机201。这样,主机201可以根据文件系统的根目录访问子系统101中的文件系统,并读取该文件系统中的数据。当主机201请求向子系统101写入数据时,子系统101可以在该根目录下创建子目录,并将主机201提供的数据通过文件格式保存在该子目录下。实际应用时,子系统101可以配置有多个IP地址,并且,当主机201接入该子系统101时,子系统101可以从多个IP地址中选择一个或者多个IP地址,并将其分配给主机201,从而主机201可以基于分配到的IP地址,访问子系统101中的文件系统。并且,每个IP地址可以用于访问文件系统中各个子目录下的所有文件,并可以支持主机201在文件系统中创建子目录,以便后续将文件存储至该子目录下。如,主机201利用分配到的IP地址1在文件系统中创建子目录d5,利用分配到的IP地址2在文件系统中创建子目录d7等。
并且,子系统101中可以配置有预设数量的分片(shard),如图2中配置有4096个分片等(即分片0至分片4095),这些分片用于将文件系统的目录以及文件分布存储至该子系统 中的不同物理节点,可以用于控制数据的分布和访问,每个物理节点可以用于存储一个或者多个分片的数据。基于多个分片的数据分布,可以生成分片视图(或者可以成为数据分布视图),并且,该分片视图可以指示数据在不同物理节点的分布情况。
由于子系统101的内存空间通常有限,因此,子系统101可以周期性的将内存中的数据进行持久化存储,具体可以是将内存中的目录、文件以及分片视图等数据下发至持久化存储介质中进行存储,并由相应的控制器支持对持久化存储介质进行数据读写。如此,子系统101可以基于释放的内存空间继续接收主机201写入的新数据。
站点B包括的子系统102,作为子系统101的灾备,可以同样配置有文件系统,并且,子系统102的文件系统中的数据(包括目录、目录下的文件以及分片视图等),可以从子系统101中的文件系统完成数据同步。并且,租户在配置主机201访问子系统的IP地址时,可以为该主机201配置访问子系统102的IP地址。
当站点A中的子系统101故障时,站点B中的子系统102可以基于其上存储的目标、文件以及分片视图等数据,接管主机201上的业务,从而可以继续为主机201提供业务服务。参见图3,示出了子系统101故障后由子系统102继续提供业务服务的示意图。
实际应用时,子系统102可以周期性的监测子系统101的心跳,具体可以是接收子系统101发送的心跳消息。当子系统102距离上一次接收子系统101的心跳消息的时间间隔超出预设间隔时,子系统102可以确定该子系统101发生故障。或者,子系统102也可以是在接收到仲裁服务器(图3中未示出)发送的通知消息后,确定子系统101发生故障等,本实施例对此并不进行限定。
子系统102在确定子系统101故障后,可以将主机201访问子系统101的IP地址漂移至子系统102,并启用为主机201分配的、用于支持主机201访问子系统102的0′号端口或者2′号端口,从而主机201可以基于该IP地址以及端口访问子系统102中的数据。由于子系统101正常运行时,在为主机201提供业务服务的过程中,不仅会将所产生的新数据保存在本地,也会将该新数据同步至子系统102中,因此,子系统102中的内存以及持久化存储介质中的数据,与子系统101中的内存、持久化存储介质中的数据可以保持一致,从而子系统102可以基于这些数据继续向主机201提供业务服务。
需要说明的是,上述图2以及图3,主要介绍了数据处理系统100中的子系统101以及子系统102基于AP模式运行以及进行灾备的实现方式。下面对数据处理系统100中的子系统101以及子系统102基于AP模式运行以及进行灾备的实现方式进行介绍。
参阅图4,示出了本申请实施例提供的数据存储系统100中的子系统基于AA模式运行的示意图。如图4所示,在AA模式中,归属于站点A的子系统101中可以包括一个或者多个控制器,并且,该控制器可以对外提供多个端口,用于通过该多个端口与主机集群200中的主机201进行通信。如图4所示,子系统101中控制器0可以提供0号端口以及1号端口、控制器1可以提供2号端口以及3号端口,并且,子系统101可以通过0号端口以及2号端口与主机201进行数据通信,如接收主机201下发的业务请求等。
归属于站点B的子系统102中包括控制器2和控制器3,并且,控制器2中的1′号端口以及控制器3中的3′号端口可以与主机集群200中的主机202进行数据通信。其中,主机202,可以作为主机201的热备(即在线备份),用于提高主机上的业务的可靠性。进一步地,子系统102中的0′号端口以及2′号端口还可以分配给主机201,以便在子系统101故障时,主机201通过0′号端口、2′号端口访问子系统102以继续请求业务服务。并且,子系统101中1号端口以及3号端口还可以分配给主机202,以便在子系统102故障时,主机202通过1 号端口、3号端口访问子系统101以继续请求业务服务。
值得注意的是,在AA模式中,当数据处理系统100正常运行时,主机201以及主机202均可以向站点请求提供业务服务。具体的,主机201可以向站点A请求提供业务,主机202可以向站点B请求提供业务。如此,在利用作为灾备的子系统102满足容灾要求的同时,通过子系统102为主机提供一部分的业务服务,可以提高数据处理系统100的整体服务能力和系统资源利用率。
站点A包括的子系统101中可以设置有文件系统,并且,该文件系统具有根目录,该根目录可以保存在子系统101的内存中。当主机201与子系统101建立通信连接后,子系统101可以将该文件系统挂载至主机201,并将该文件系统的根目录提供给主机201。这样,主机201可以根据文件系统的根目录访问子系统101中的文件系统,并读取该文件系统中的数据。当主机201请求向子系统101写入数据时,子系统101可以在该根目录下创建子目录,并将主机201提供的数据通过文件格式保存在该子目录下。
站点B中的子系统102中同样设置有文件系统,并且,子系统102中的文件系统与子系统101中的文件系统可以具有统一的根目录。当主机202与子系统102建立通信连接后,主机202可以访问子系统101中的文件系统,并在该文件系统中读写数据。
由于子系统101以及子系统102均可以提供业务服务,因此,在为主机201以及主机202配置访问子系统的IP地址时,主机201以及主机202可以分配到相同的多个IP地址,该多个IP地址用于支持主机201/主机202访问文件系统中各个子目录下的所有文件,并可以在该文件系统中创建相应的目录,以便后续将文件存储至所创建的子目录下,如主机201可以利用分配到的IP地址a在文件系统中创建子目录d5、主机202可以利用分配到的IP地址b在文件系统中创建子目录d7等。实际应用时,还可以配置主机201以及主机202分别所述的归属站点,如配置主机201归属于站点A、配置主机202归属于站点B,并且,主机201以及主机202优先对归属站点的文件系统中的数据进行访问,以此尽可能避免主机所请求访问的文件在不同站点之间进行数据传输,并且,两个站点上各有一部分IP地址处于工作状态。
子系统101以及子系统102中,分别可以配置有不同的分片。如图4所示,子系统101中以及子系统102中分别配置有2048个分片,其中,子系统101中的分片为分片0~分片2047,子系统102中的分片为分片2048~分片4095。这些分片用于将文件系统的目录以及文件分布存储至该子系统中的不同物理节点,可以用于控制数据的分布和访问,每个物理节点可以用于存储一个或者多个分片的数据。基于多个分片的数据分布,可以生成分片视图,并且,该分片视图可以指示数据在不同物理节点的分布情况。本实施例中,是以将分片视图均分在子系统101以及子系统102上为例,实际应用时,分片视图也可以不均分于两个子系统中。
通常情况下,子系统101以及子系统102的内存空间通常有限,因此,子系统101以及子系统102可以周期性的将内存中的数据进行持久化存储。由于子系统101以及子系统102中分别存储有部分分片视图中的数据,因此,在对内存中的数据进行持久化存储之前,子系统101可以将分片0~分片2047中的数据发送给子系统102,子系统102可以将分片2048~分片4095中的数据发送给子系统101。这样,子系统101以及子系统102中均能够获得完整的4096个分片的数据。然后,子系统101以及子系统102可以分别将所有分片中的数据(如完整的目录、文件以及分片视图等数据)下发至持久化存储介质中进行存储,并由相应的控制器支持对持久化存储介质进行数据读写。如此,子系统101以及子系统102可以基于释放的内存空间继续接收主机201以及主机202写入的新数据。
实际应用场景中,子系统101或者子系统102在运行过程中可能发生故障,此时,可以 由正常运行的另一子系统接管故障子系统上的业务。以站点A中的子系统101发生故障为例,参见图5,示出了子系统102继续提供业务服务的示意图。
具体地,子系统101以及子系统102可以同时监测对端的心跳,当子系统101发生故障时,子系统102距离上一次接收子系统101的心跳消息的时间间隔通常会超出预设间隔,此时,子系统102可以确定该子系统101发生故障。或者,子系统101也可以是在接收到仲裁服务器(图5中未示出)发送的通知消息后,确定子系统101发生故障等,本实施例对此并不进行限定。
子系统102在确定子系统101故障后,可以要求子系统101将内存中存储的分片0~分片2047中的数据提供给子系统102,如此,子系统102上的分片视图可以更新为分片0~分片4095。
并且,子系统102可以将主机201访问子系统101的IP地址漂移至子系统102,并启用0′号端口以及2′号端口接收主机201下发的业务请求。这样,主机201在请求访问文件系统中的文件时,子系统102可以将分片0~分片4095中存储的相应文件反馈给主机201,从而实现为主机201继续提供业务服务。
由于实际应用场景中,主机集群200或者主机集群200中的主机对于数据处理系统100的工作模式的要求可能发生变化,为此,本实施例中数据处理系统100中的子系统101以及子系统102支持工作模式在AP模式以及AA模式之间进行切换,以满足主机集群对于子系统101以及子系统102之间的工作模式的要求,提高数据处理系统100提供业务服务的普适性。其中,数据处理系统100可以支持子系统101与子系统102的工作模式由AP模式切换为AA模式,可以由AA模式切换为AP模式。
本实施例中,可以由数据处理系统100中的数据处理装置400切换子系统101与子系统102之间的工作模式。具体实现时,接收单元401可以接收配置指令,该配置指令用于指示数据处理装置400对子系统101与子系统102之间的工作模式由AP模式切换为AA模式,或者由AA模式切换为AP模式。
作为一些示例,本实施例提供了以下几种接收配置指令的实现方式。
在第一种可能的实施方式中,数据处理装置400可以对外提供客户端,从而租户或者管理员在确定需要切换子系统101与子系统102之间的工作模式时,通过该客户端向数据处理装置400下发配置指令,以指示数据处理装置400将子系统101与子系统102之间的工作模式由当前的模式切换至另一种模式,并由接收单元401接收到该配置指令。
在第二种可能的实施方式中,数据处理装置400可以对当前接入数据处理系统100的主机的组网方式进行检测,该主机的组网方式可以平行组网或者全交叉组网。其中,平行组网,是指网络中的不同主机之间可以不建立交叉连接,如不同主机之间可以不直接建立数据通信的通道;并且,当采用平行组网方式的主机接入数据处理系统100时,子系统101与子系统102通常基于AP模式运行以为主机提供相应的服务。全交叉组网,是指网络中的不同主机之间可以建立连接,从而不同主机之间可以通过该通信连接进行数据通信;并且,当采用全交叉组网方式的主机接入数据处理系统100时,子系统101与子系统102通常基于AA模式运行以为主机提供相应的服务。实际应用时,主机的组网方式也可以是其它方式,本实施例对此并不进行限定。
如此,接收单元401可以通过对主机的组网方式进行检测,判断当前接入数据处理系统100的主机的组网方式是否与子系统101以及子系统102之间的工作模式相匹配,如果不匹配,则接收单元401可以生成配置指令,以触发对子系统101与子系统102之间的工作模式 的切换。如果匹配,则数据处理装置400无需切换子系统101与子系统102之间的工作模式。
上述两种接收配置指令的实现方式仅作为一些示例性说明,在其他实施例中,接收单元也可以是基于其它方式获取配置指令,本实施例对此并不进行限定。
接收单元401在接收到配置指令后,可以将该配置指令发送给处理单元402,以便处理单元402执行切换子系统101与子系统102之间的工作模式的操作。
下面结合附图对处理单元402切换子系统101与子系统102之间的工作模式的具体实现过程进行示例性说明。
参见图6,示出了子系统101与子系统102的工作模式由AP模式切换为AA模式的示意图。
如图6所示,当子系统101与子系统102的工作模式切换为AA模式时,子系统102可以支持对主机202向子系统102请求业务。
具体实现时,处理单元402可以指示站点B中的子系统102,将控制器2中的1′号端口以及控制器3中的3′号端口分配给主机202,以便主机202可以通过该1′号端口或3′号端口访问子系统102中的文件系统。
并且,处理单元402可以指示子系统101将部分分片中的数据迁移至子系统102中,并分别在子系统101以及子系统102中对分片视图进行更新。比如,假设子系统101将分片2048~分片4095中的数据迁移至子系统102,则子系统101以及子系统102中更新后的分片视图如图6所示。
由于主机201以及主机202均可以向站点请求业务,因此,处理单元402可以指示子系统102为主机202分配访问子目录下文件的IP地址。进一步地,子系统102还可以指定主机202的归属站点为站点B,这样,主机202在访问子目录下的文件时(如访问子目录d6和d7下的文件等),可以优先对其归属的站点B中的子系统102进行访问,并可以根据子系统102上的分片视图(如分片2048~分片4095),在子系统102中获得其所需的数据。如此,可以避免主机202访问子系统101而导致文件在不同站点之间进行数据传输。另外,子系统101可以指定主机201的归属站点为站点A,以便主机201优先对站点A中的子系统101进行访问。
如此,可以实现子系统101与子系统102的工作模式由AP模式至AA模式的切换。
参见图7,示出了子系统101与子系统102的工作模式由AA模式切换为AP模式的示意图。
如图7所示,当子系统101与子系统102的工作模式切换为AP模式时,子系统102不支持主机202向子系统102请求业务。
具体实现时,处理单元402可以指示站点B中的子系统102,可以将0′号端口以及2′号端口分配给主机201,以便当子系统101故障时,主机201可以通过该0′号端口或2′号端口访问子系统102中的文件系统。由于在AP模式下,数据处理系统100仅支持主机201向站点请求业务,因此,子系统102可以禁用之前为主机202分配的、用于访问子目录下文件的IP地址。
并且,由于子系统101以及子系统102在基于AA模式运行时,子系统101以及子系统102中的分片均为部分分片,因此,在切换至AP模式时,处理单元402可以指示子系统101将所有分片中的数据备份至子系统102中,子系统102可以将所有分片中的数据备份至子系统101中,并分别在子系统101以及子系统102中对分片视图进行更新。比如,假设在AA模式下,子系统101以及子系统102中的分片分别为分片0~分片2047、分片2048~分片4095,则在切换至AP模式时,子系统101以及子系统102在相互备份分片数据后,子系统101以及 子系统102中更新后的分片视图如图7所示。
如此,可以实现对数据处理系统100中的不同子系统的工作模式在AP模式与AA模式之间进行灵活切换,以满足实际应用场景的需求。
值得注意的是,图1所示的数据处理系统100的系统架构仅作为一种示例,并不用于限定其具体实现局限于该示例。比如,在图8所示的数据处理系统100中,除了包括子系统101以及子系统102之间,还可以包括子系统103以及子系统104,并且,子系统101与子系统103均属于站点A,子系统102与子系统104均属于站点B。在图8所示的数据处理系统中,可以利用子系统101至子系统104支持同时提供AP模式以及AA模式对应的服务。具体地,子系统101与子系统102可以基于AP模式运行,用于为主机集群200中的主机201以及主机202提供AP模式对应的服务;子系统103与子系统104可以基于AA模式运行,用于为主机集群300中的主机301以及主机302提供AA模式对应的服务。
又比如,在图9所示的数据处理系统100中,与图8所示的数据处理系统100类似,同样包括子系统101至子系统104,并且同时支持提供AP模式与AA模式分别对应的服务。其区别在于,图9所示的数据处理系统100中,子系统101至子系统104分别归属于不同的站点,具体的,子系统101归属于站点A、子系统102归属于站点B、子系统103归属于站点C、子系统104归属于站点D。本申请对于数据处理系统的具体架构并不进行限定。
这样,在图8或者图9所示的数据处理系统100中,对于部分要求AP模式的租户,数据处理系统100可以利用子系统101以及子系统102处理该租户的主机所请求的业务;对于另一部分要求AA模式的租户,数据处理系统100可以利用子系统103以及子系统104处理该租户的主机所请求的业务。由于数据处理系统100中的不同子系统分别运行AP模式以及AA模式,这使得数据处理系统100可以根据租户的需求选择相应工作模式的子系统为该租户的主机提供业务服务,以此可以满足实际应用场景中不同主机的业务需求,提高数据处理系统100提供业务服务的普适性。
需要说明的是,上述图2至图9所示的数据处理系统100,是以为租户提供数据存储服务并且该数据存储服务具体为文件存储服务为例进行说明,即,数据处理系统100所存储的数据可以是文件的形式进行保存,而在其它可能的实施方式中,数据处理系统100所提供数据存储服务,也可以是块存储服务或者对象存储服务。其中,块存储服务,是指数据处理系统100在存储数据时,将该数据按照固定大小的尺寸进行分块,每个分块的数据量例如可以是512字节或者4千字节(KB)等。对象存储服务,是指数据处理系统100将数据作为对象(object)进行存储,此时,对象可以是数据处理系统100中数据存储的基本单位,每个对象可以是数据和数据属性的综合体,该数据属性可以根据应用的需求进行设置,包括数据分布、服务质量等。实际应用时,数据处理系统100也可以是提供其它类型的服务,如数据计算服务,如模型训练服务、数据统计服务等,本实施例对此并不进行限定。
并且,上述图1至图9所示的数据处理系统100,可以适用于集中式存储的应用场景或者分布式存储的应用场景。
其中,在集中式存储应用场景中,每个子系统可以由一台或多台计算节点组成中心节点,并且整个子系统的所有数据处理业务可以集中部署在这个中心节点上。此时,计算节点与存储设备之间可以采用盘控分离架构,即计算节点与存储设备独立部署;或者,计算节点与存储设备之间可以采用盘控一体架构,即计算节点可以具有槽位,并通过该槽位将存储设备放置在该计算节点中,与该计算节点集成部署。
在分布式存储应用场景中,每个子系统中的数据可以分散存储在多个独立的存储节点上。 此时,计算节点可以与存储设备集成部署,使得该计算节点同时具有计算能力以及存储能力,并且在该计算节点上可以创建虚拟机,或者也可以不创建虚拟机。或者,计算节点与存储设备之间可以采用存算分离架构,即计算节点与存储设备独立部署并通过网络进行通信。另外,存储设备中可以包括一种或者多种不同的存储介质,本实施例对此并不进行限定。
另外,当租户的主机接入上述数据处理系统100时,可以由管理员根据租户的需求,配置该租户的主机接入AP模式运行的子系统,或者接入AA模式运行的子系统。而在另一种实现方式中,数据处理系统100也可以根据主机的组网方式自动为主机接入适合的子系统。下面结合附图对该自动化过程进行详细说明。
如图10所示,为本申请实施例中一种数据处理方法的流程示意图,该方法可以应用于如图1、图8、或图9所示的数据处理系统100中。实际应用时,该方法也可以应用于其它可适用的数据处理系统中。为便于理解与描述,下面以应用于图8所示的数据处理系统100为例进行示例性说明,该方法具体可以包括:
S901:接收单元401获取接入数据处理系统100的主机的组网形式。
其中,主机的组网形式,可以是平行组网或者全交叉组网等,或者可以是其他类型的组网方式。其中,平行组网,是指网络中的不同主机之间可以不建立交叉连接,如不同主机之间可以不直接建立数据通信的通道。全交叉组网,是指网络中的不同主机之间可以建立连接,从而不同主机之间可以通过该通信连接进行数据通信。
S902:当该主机的组网形式为平行组网时,处理单元402利用数据处理系统100中的子系统101以及子系统102处理该主机所请求的业务,子系统101与子系统102之间的工作模式为AP模式。
S903:当该主机的组网形式为全交叉组网时,处理单元402利用数据处理系统100中的子系统103以及子系统104处理该主机所请求的业务,子系统103与子系统104之间的工作模式为AA模式。
本实施例中,可以根据实际应用的场景需求,将组网形式为平行组网的主机,接入基于AP模式运行的子系统101,并可以为该主机自动配置子系统102上的访问端口以及IP地址,以便在子系统101发生故障后,该主机可以通过访问子系统102实现业务的不中断处理。
而对于组网形式为全交叉组网的主机,可以将该主机接入基于AA模式运行的子系统103,将作为该主机的热备的另一主机接入子系统104中,并配置两个主机具有相同的IP地址、各个主机分别归属的站点,以便子系统103以及子系统104分别各自接入的主机所请求的业务。
在进一步可能的实施方式中,当主机对于接入的子系统的工作模式的要求发生变化时,可以对该主机所接入的子系统的工作模式进行自动切换,如当主机要求子系统101的工作模式由AP模式转换为AA模式时,子系统101可以基于上述图6所示的方式,自动将子系统101余子系统102的工作模式切换为AA模式,以满足主机的实际应用需求,其具体实现过程,可参见前述实施例的相关之处描述,在此不做赘述。其中,可以由接入主机的子系统自动执行工作模式的切换过程,或者可以是在仲裁服务器或者其他设备的控制下实现将子系统的工作模式进行自动切换,本实施例对此并不进行限定。
上述结合附图所描述的本申请实施例中的各种非限定性实施方式,仅作为示例性说明。显然,所描述的实施例是本申请一部分实施例,而不是全部的实施例。基于本申请中的实施例,基于上述内容所获得的所有其它实施例,都属于本申请保护的范围。
上文中结合图1至图10,详细描述了本申请所提供的数据处理系统以及数据处理方法, 下面将结合图11和图12,描述根据本申请所提供的数据处理装置以及实现该数据处理装置功能的计算设备。
与上述方法同样的发明构思,本申请实施例还提供一种数据处理装置。如图11所示,数据处理装置1100应用于数据处理系统,如上述图1至图10所示的数据处理系统100等,该数据处理系统包括第一子系统以及第二子系统,如上述子系统101和子系统102等,图11所示的数据处理装置100包括:
接收单元1101,用于接收配置指令;
处理单元1102,用于根据配置指令将第一子系统和第二子系统之间由第一模式切换为第二模式;其中,第一模式为第一子系统与第二子系统之间工作在活动-消极AP模式和活动-活动AA模式中的一种模式;第二模式为第一子系统与第二子系统之间工作在AP模式和AA模式中的另一种模式。
在一种可能的实施方式中,第一子系统与第二子系统之间基于第一模式为目标租户的主机提供数据处理服务,并且,在将第一子系统与第二子系统之间的工作模式切换为第二模式后,第一子系统与第二子系统之间基于第二模式为目标租户的主机提供数据处理服务。
在一种可能的实施方式中,接收单元1101,具体用于检测到接入数据处理系统的主机的组网模式,由第一组网模式切换为第二组网模式。
在一种可能的实施方式中,第一组网模式为平行组网模式和全交叉组网模式中的一种组网模式,第二组网模式为平行组网模式和全交叉组网模式中的另一种组网模式。
在一种可能的实施方式中,数据处理系统还包括第三子系统以及第四子系统,在第一子系统与第二子系统之间工作在第二模式时,第三子系统与第四子系统之间工作在第一模式。
在一种可能的实施方式中,第一子系统与第三子系统属于第一站点,第二子系统与第四子系统属于第二站点。
在一种可能的实施方式中,第一子系统与第二子系统用于为目标租户提供数据存储服务,第一子系统包括至少一个存储阵列。
在一种可能的实施方式中,第一子系统与第二子系统通过IP协议、RDMA协议、FC中的至少一种进行通信。
在一种可能的实施方式中,第一子系统与第二子系统提供的数据存储服务为文件存储服务。
在一种可能的实施方式中,第一子系统与第二子系统提供的数据存储服务为块存储服务。
在一种可能的实施方式中,第一子系统与第二子系统提供的数据存储服务为对象存储服务。
图11所示的数据处理装置1100,对应于前述实施例中数据处理装置400,因此,数据处理装置1100的具体实现及其所具有的技术效果,可参见前述实施例的相关之处描述,在此不做赘述。
本申请实施例还提供一种计算设备,如图12所示,计算设备1200中可以包括通信接口1210、处理器1220。可选的,计算设备1200中还可以包括存储器1230。其中,存储器1230可以设置于计算设备1200内部,还可以设置于计算设备1200外部。示例性地,上述图1至图10所示实施例中数据处理装置400执行的各个动作均可以由处理器1220实现。处理器1220可以通过通信接口1210获取配置指令,并用于实现上述图9中所述的方法。在实现过程中,处理流程的各步骤可以通过处理器1220中的硬件的集成逻辑电路或者软件形式的指令完成图9所述的方法。为了简洁,在此不再赘述。处理器1220用于实现上述方法所执行的程序代 码可以存储在存储器1230中。存储器1230和处理器1220连接,如耦合连接等。
本申请实施例的一些特征可以由处理器1220执行存储器1230中的程序指令或者软件代码来完成/支持。存储器1230上在加载的软件组件可以从功能或者逻辑上进行概括,例如,图11所示的处理单元1102。而接收单元1101的功能可以由通信接口1210实现。
本申请实施例中涉及到的任一通信接口可以是电路、总线、收发器或者其它任意可以用于进行信息交互的装置。比如计算设备1200中的通信接口1210,示例性地,该其它装置可以是与该计算设备1200相连的设备等。
本申请实施例中涉及的处理器可以是通用处理器、数字信号处理器、专用集成电路、现场可编程门阵列或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件,可以实现或者执行本申请实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者任何常规的处理器等。结合本申请实施例所公开的方法的步骤可以直接体现为硬件处理器执行完成,或者用处理器中的硬件及软件单元组合执行完成。
本申请实施例中的耦合是装置、单元或单元之间的间接耦合或通信连接,可以是电性,机械或其它的形式,用于装置、单元或单元之间的信息交互。
处理器可能和存储器协同操作。存储器可以是非易失性存储器,比如硬盘或固态硬盘等,还可以是易失性存储器,例如随机存取存储器。存储器是能够用于携带或存储具有指令或数据结构形式的期望的程序代码并能够由计算机存取的任何其他介质,但不限于此。
本申请实施例中不限定上述通信接口、处理器以及存储器之间的具体连接介质。比如存储器、处理器以及通信接口之间可以通过总线连接。所述总线可以分为地址总线、数据总线、控制总线等。
基于以上实施例,本申请实施例还提供了一种计算机存储介质,该存储介质中存储软件程序,该软件程序在被一个或多个处理器读取并执行时可实现上述任意一个或多个实施例提供的数据处理装置400执行的方法。所述计算机存储介质可以包括:U盘、移动硬盘、只读存储器、随机存取存储器、磁碟或者光盘等各种可以存储程序代码的介质。
基于以上实施例,本申请实施例还提供了一种芯片,该芯片包括处理器,用于实现上述实施例所涉及的数据处理装置400的功能,例如用于实现图9中所执行的方法。可选地,所述芯片还包括存储器,所述存储器,用于处理器所执行必要的程序指令和数据。该芯片,可以由芯片构成,也可以包含芯片和其他分立器件。
本领域内的技术人员应明白,本申请的实施例可提供为方法、系统、或计算机程序产品。因此,本申请可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且,本申请可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本申请的说明书和权利要求书及上述附图中的术语“第一”、“第二”、“第三”等是用于区别类似的对象,而不必用于描述特定的顺序或先后次序。应该理解这样使用的术语在适当情况下可以互换,这仅仅是描述本申请的实施例中对相同属性的对象在描述时所采用的区分方式。
本申请是参照根据本申请实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器,使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程 图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中,使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品,该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上,使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理,从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
本申请的说明书和权利要求书及上述附图中的术语“第一”、“第二”等是用于区别类似的对象,而不必用于描述特定的顺序或先后次序。应该理解这样使用的术语在适当情况下可以互换,这仅仅是描述本申请的实施例中对相同属性的对象在描述时所采用的区分方式。
显然,本领域的技术人员可以对本申请实施例进行各种改动和变型而不脱离本申请实施例的范围。这样,倘若本申请实施例的这些修改和变型属于本申请权利要求及其等同技术的范围之内,则本申请也意图包含这些改动和变型在内。

Claims (36)

  1. 一种数据处理系统,其特征在于,包含第一子系统和第二子系统;所述数据处理系统包含接收单元和处理单元;其中,
    所述接收单元,用于接收配置指令;
    所述处理单元,用于根据所述配置指令将所述第一子系统和所述第二子系统之间由第一模式切换为第二模式;其中,所述第一模式为所述第一子系统与所述第二子系统之间工作在活动-消极AP模式和活动-活动AA模式中的一种模式;所述第二模式为所述第一子系统与所述第二子系统之间工作在所述AP模式和所述AA模式中的另一种模式。
  2. 根据权利要求1所述的数据处理系统,其特征在于,所述第一子系统与所述第二子系统之间基于所述第一模式为目标租户的主机提供数据处理服务,并且,在将所述第一子系统与所述第二子系统之间的工作模式切换为所述第二模式后,所述第一子系统与所述第二子系统之间基于所述第二模式为所述目标租户的主机提供数据处理服务。
  3. 根据权利要求1或2所述的数据处理系统,其特征在于,所述接收单元,具体用于检测到接入所述数据处理系统的主机的组网模式,由第一组网模式切换为第二组网模式。
  4. 根据权利要求3所述的数据处理系统,其特征在于,所述第一组网模式为平行组网模式和全交叉组网模式中的一种组网模式,所述第二组网模式为所述平行组网模式和所述全交叉组网模式中的另一种组网模式。
  5. 根据权利要求1至4任一项所述的数据处理系统,其特征在于,所述数据处理系统还包括第三子系统以及第四子系统,在所述第一子系统与所述第二子系统之间工作在所述第二模式时,所述第三子系统与所述第四子系统之间工作在所述第一模式。
  6. 根据权利要求5所述的数据处理系统,其特征在于,所述第一子系统与所述第三子系统属于第一站点,所述第二子系统与所述第四子系统属于第二站点。
  7. 根据权利要求1至6任一项所述的数据处理系统,其特征在于,所述第一子系统与所述第二子系统用于为目标租户提供数据存储服务,所述第一子系统包括至少一个存储阵列。
  8. 根据权利要求1至7任一项所述的数据处理系统,其特征在于,所述第一子系统与所述第二子系统通过网际互联协议IP、远程直接数据存取RDMA协议、光纤通道FC中的至少一种进行通信。
  9. 根据权利要求1至8任一项所述的数据处理系统,其特征在于,所述第一子系统与所述第二子系统提供的数据存储服务为文件存储服务。
  10. 根据权利要求1至8任一项所述的数据处理系统,其特征在于,所述第一子系统与所述第二子系统提供的数据存储服务为块存储服务。
  11. 根据权利要求1至8任一项所述的数据处理系统,其特征在于,所述第一子系统与所述第二子系统提供的数据存储服务为对象存储服务。
  12. 一种数据处理方法,其特征在于,所述方法应用于数据处理系统,所述数据处理系统包含第一子系统、第二子系统、接收单元和处理单元,所述方法包括:
    所述接收单元接收配置指令,并将所述配置指令提供给所述处理单元;
    所述处理单元根据所述配置指令将所述第一子系统和所述第二子系统之间由第一模式切换为第二模式;其中,所述第一模式为所述第一子系统与所述第二子系统之间工作在活动-消极AP模式和活动-活动AA模式中的一种模式;所述第二模式为所述第一子系统与所述第二子系统之间工作在所述AP模式和所述AA模式中的另一种模式。
  13. 根据权利要求12所述的方法,其特征在于,所述第一子系统与所述第二子系统之间 基于所述第一模式为目标租户的主机提供数据处理服务,并且,在将所述第一子系统与所述第二子系统之间的工作模式切换为所述第二模式后,所述第一子系统与所述第二子系统之间基于所述第二模式为所述目标租户的主机提供数据处理服务。
  14. 根据权利要求12或13所述的方法,其特征在于,所述接收单元接收配置指令,包括:
    所述接收单元检测到接入所述数据处理系统的主机的组网模式,由第一组网模式切换为第二组网模式。
  15. 根据权利要求14所述的方法,其特征在于,所述第一组网模式为平行组网模式和全交叉组网模式中的一种组网模式,所述第二组网模式为所述平行组网模式和所述全交叉组网模式中的另一种组网模式。
  16. 根据权利要求12至15任一项所述的方法,其特征在于,所述数据处理系统还包括第三子系统以及第四子系统,在所述第一子系统与所述第二子系统之间工作在所述第二模式时,所述第三子系统与所述第四子系统之间工作在所述第一模式。
  17. 根据权利要求16所述的方法,其特征在于,所述第一子系统与所述第三子系统属于第一站点,所述第二子系统与所述第四子系统属于第二站点。
  18. 根据权利要求12至17任一项所述的方法,其特征在于,所述第一子系统与所述第二子系统用于为目标租户提供数据存储服务,所述第一子系统包括至少一个存储阵列。
  19. 根据权利要求12至18任一项所述的方法,其特征在于,所述第一子系统与所述第二子系统通过网际互联协议IP、远程直接数据存取RDMA协议、光纤通道FC中的至少一种进行通信。
  20. 根据权利要求12至19任一项所述的方法,其特征在于,所述第一子系统与所述第二子系统提供的数据存储服务为文件存储服务。
  21. 根据权利要求12至19任一项所述的方法,其特征在于,所述第一子系统与所述第二子系统提供的数据存储服务为块存储服务。
  22. 根据权利要求12至19任一项所述的方法,其特征在于,所述第一子系统与所述第二子系统提供的数据存储服务为对象存储服务。
  23. 一种数据处理装置,其特征在于,所述数据处理装置应用于数据处理系统,所述数据处理系统包括第一子系统以及第二子系统,所述数据处理装置包括:
    所述接收单元,用于接收配置指令;
    所述处理单元,用于根据所述配置指令将所述第一子系统和所述第二子系统之间由第一模式切换为第二模式;其中,所述第一模式为所述第一子系统与所述第二子系统之间工作在活动-消极AP模式和活动-活动AA模式中的一种模式;所述第二模式为所述第一子系统与所述第二子系统之间工作在所述AP模式和所述AA模式中的另一种模式。
  24. 根据权利要求23所述的数据处理装置,其特征在于,所述第一子系统与所述第二子系统之间基于所述第一模式为目标租户的主机提供数据处理服务,并且,在将所述第一子系统与所述第二子系统之间的工作模式切换为所述第二模式后,所述第一子系统与所述第二子系统之间基于所述第二模式为所述目标租户的主机提供数据处理服务。
  25. 根据权利要求23或24所述的数据处理装置,其特征在于,所述接收单元,具体用于检测到接入所述数据处理系统的主机的组网模式,由第一组网模式切换为第二组网模式。
  26. 根据权利要求25所述的数据处理装置,其特征在于,所述第一组网模式为平行组网模式和全交叉组网模式中的一种组网模式,所述第二组网模式为所述平行组网模式和所述全交叉组网模式中的另一种组网模式。
  27. 根据权利要求23至26任一项所述的数据处理装置,其特征在于,所述数据处理系统还包括第三子系统以及第四子系统,在所述第一子系统与所述第二子系统之间工作在所述第二模式时,所述第三子系统与所述第四子系统之间工作在所述第一模式。
  28. 根据权利要求27所述的数据处理装置,其特征在于,所述第一子系统与所述第三子系统属于第一站点,所述第二子系统与所述第四子系统属于第二站点。
  29. 根据权利要求23至28任一项所述的数据处理装置,其特征在于,所述第一子系统与所述第二子系统用于为目标租户提供数据存储服务,所述第一子系统包括至少一个存储阵列。
  30. 根据权利要求23至29任一项所述的数据处理装置,其特征在于,所述第一子系统与所述第二子系统通过网际互联协议IP、远程直接数据存取RDMA协议、光纤通道FC中的至少一种进行通信。
  31. 根据权利要求23至30任一项所述的数据处理装置,其特征在于,所述第一子系统与所述第二子系统提供的数据存储服务为文件存储服务。
  32. 根据权利要求23至30任一项所述的数据处理装置,其特征在于,所述第一子系统与所述第二子系统提供的数据存储服务为块存储服务。
  33. 根据权利要求23至30任一项所述的数据处理装置,其特征在于,所述第一子系统与所述第二子系统提供的数据存储服务为对象存储服务。
  34. 一种计算设备,其特征在于,所述计算设备包括处理器和存储器;
    所述处理器用于执行所述存储器中存储的指令,以使得所述计算设备执行权利要求12至22任一项所述的方法。
  35. 一种计算机可读存储介质,其特征在于,所述计算机可读存储介质中存储有指令,当其在计算设备上运行时,使得所述计算设备执行如权利要求12至22任一项所述的方法。
  36. 一种包含指令的计算机程序产品,其特征在于,当其在计算设备上运行时,使得所述计算设备执行如权利要求12至22任一项所述的方法。
PCT/CN2023/087566 2022-06-22 2023-04-11 数据处理系统、数据处理方法、装置及相关设备 WO2023246241A1 (zh)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
CN202210715668 2022-06-22
CN202210715668.9 2022-06-22
CN202211170871.9 2022-09-23
CN202211170871.9A CN117271205A (zh) 2022-06-22 2022-09-23 数据处理系统、数据处理方法、装置及相关设备

Publications (1)

Publication Number Publication Date
WO2023246241A1 true WO2023246241A1 (zh) 2023-12-28

Family

ID=89199638

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/087566 WO2023246241A1 (zh) 2022-06-22 2023-04-11 数据处理系统、数据处理方法、装置及相关设备

Country Status (2)

Country Link
CN (1) CN117271205A (zh)
WO (1) WO2023246241A1 (zh)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150378767A1 (en) * 2014-06-28 2015-12-31 Vmware, Inc. Using active/active asynchronous replicated storage for live migration
US10009215B1 (en) * 2012-11-30 2018-06-26 EMC IP Holding Company LLC Active/passive mode enabler for active/active block IO distributed disk(s)
US20200233582A1 (en) * 2019-01-23 2020-07-23 EMC IP Holding Company LLC Seamless transitioning among replication modes in a storage system

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10009215B1 (en) * 2012-11-30 2018-06-26 EMC IP Holding Company LLC Active/passive mode enabler for active/active block IO distributed disk(s)
US20150378767A1 (en) * 2014-06-28 2015-12-31 Vmware, Inc. Using active/active asynchronous replicated storage for live migration
US20200233582A1 (en) * 2019-01-23 2020-07-23 EMC IP Holding Company LLC Seamless transitioning among replication modes in a storage system

Also Published As

Publication number Publication date
CN117271205A (zh) 2023-12-22

Similar Documents

Publication Publication Date Title
US20200358848A1 (en) Methods, systems, and media for providing distributed database access during a network split
TWI710915B (zh) 一種基於互聯網資料中心的資源處理方法、相關裝置以及通信系統
US9990262B2 (en) Dynamic mirroring
JP6317856B2 (ja) クラスタ間冗長構成におけるスムーズな制御部交代
US9842033B2 (en) Storage cluster failure detection
US8838850B2 (en) Cluster control protocol
US20160077996A1 (en) Fibre Channel Storage Array Having Standby Controller With ALUA Standby Mode for Forwarding SCSI Commands
US10320905B2 (en) Highly available network filer super cluster
JP2005267327A (ja) ストレージシステム
US10534541B2 (en) Asynchronous discovery of initiators and targets in a storage fabric
CN108319623A (zh) 一种数据重分布方法、装置及数据库集群
JP2009237826A (ja) ストレージシステム及びそのボリューム管理方法
CN113010496B (zh) 一种数据迁移方法、装置、设备和存储介质
EP3495939B1 (en) Method and device for storing data in distributed block storage system, and computer readable storage medium
WO2018157605A1 (zh) 一种集群文件系统中消息传输的方法及装置
CN113849136B (zh) 一种基于国产平台的自动化fc块存储处理方法和系统
US8838768B2 (en) Computer system and disk sharing method used thereby
US10782889B2 (en) Fibre channel scale-out with physical path discovery and volume move
JP2020191100A (ja) システム及びその制御方法並びにプログラム
US10853203B2 (en) Storage aggregate restoration
CN112104729A (zh) 一种存储系统及其缓存方法
WO2023246241A1 (zh) 数据处理系统、数据处理方法、装置及相关设备
EP4357949A1 (en) Authentication method and apparatus, and storage system
US11977510B2 (en) Configuring a file server
US10768834B2 (en) Methods for managing group objects with different service level objectives for an application and devices thereof

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23825896

Country of ref document: EP

Kind code of ref document: A1