WO2017114262A1 - 一种管理分布式存储系统的方法及分布式存储系统 - Google Patents

一种管理分布式存储系统的方法及分布式存储系统 Download PDF

Info

Publication number
WO2017114262A1
WO2017114262A1 PCT/CN2016/111328 CN2016111328W WO2017114262A1 WO 2017114262 A1 WO2017114262 A1 WO 2017114262A1 CN 2016111328 W CN2016111328 W CN 2016111328W WO 2017114262 A1 WO2017114262 A1 WO 2017114262A1
Authority
WO
WIPO (PCT)
Prior art keywords
storage device
data
write
request
identifier
Prior art date
Application number
PCT/CN2016/111328
Other languages
English (en)
French (fr)
Inventor
陈业夫
姜军
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 WO2017114262A1 publication Critical patent/WO2017114262A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/16—Error detection or correction of the data by redundancy in hardware
    • G06F11/20—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/16—Error detection or correction of the data by redundancy in hardware
    • G06F11/20—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
    • G06F11/2053—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where persistent mass storage functionality or persistent mass storage control functionality is redundant
    • G06F11/2089—Redundant storage control functionality
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/16—Error detection or correction of the data by redundancy in hardware
    • G06F11/20—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
    • G06F11/2053—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where persistent mass storage functionality or persistent mass storage control functionality is redundant
    • G06F11/2094—Redundant storage or storage space

Definitions

  • the present invention relates to the field of computer technologies, and in particular, to a method for managing a distributed storage system and a distributed storage system.
  • a lock server In a distributed storage system, a lock server is responsible for managing mutually exclusive access of multiple storage devices to the same resource. When the storage device needs to perform read/write operations on the resource, the lock server needs to apply for the corresponding permission. Only after the permission is obtained, the storage device can perform the corresponding operation on the resource.
  • the storage device 1 may fail before the execution of the write operation succeeds.
  • the lock server reclaims the grant storage device 1 Permission to grant storage device 2 permission to perform the operations it is requesting to perform. This is because if the lock server occupies the fault that the faulty storage device operates on the data resource, it will block the normal storage device from gaining access to the data resource and affect the efficiency of the distributed storage system.
  • the embodiment of the present application provides a method for managing a distributed storage system and a distributed storage system, which is used to solve the problem that the write data request cannot be executed normally after the storage device that obtains the permission to execute the write data request fails in the prior art. problem.
  • an embodiment of the present invention provides a method for managing a distributed storage system, where the distributed storage system includes multiple storage devices and a lock server, where each storage device includes a cache and The hard disk, the method includes the following steps: the first storage device receives the write data request, and sends a first write lock request to the lock server, where the first write lock request is used to apply for executing the write data request, and the first write lock request includes the first An identifier of the storage device and an identifier of the second storage device, wherein the identifier of the second storage device can be identified by the lock server as an identifier of the backup storage device that performs the write data request.
  • the lock server grants the first storage device permission to write a data request, and stores the identifier of the backup storage device; after obtaining the permission, the first storage device writes the data to be written into the cache of the first storage device; and, is to be written
  • the inbound data is sent to the second storage device; the second storage device writes the data to be written into the local cache, and when determining that the first storage device is faulty, requests the lock server to execute the permission to write the data request, and the lock server confirms
  • the second storage device is the foregoing standby storage device, the second storage device is granted the right; after obtaining the permission, the second storage device writes the data to be written in the local cache to the hard disk of the second storage device.
  • the write data request in a case where the first storage device that performs the write data request is obtained, the write data request can be continued by the second storage device that is the backup storage device that performs the write data request, and the write data request is successfully succeeded. Execution to improve system reliability.
  • the write data request received by the first storage device is specifically used to request to update the first data to the data to be written
  • the method for managing the distributed storage system further includes the following steps: after the first storage device fails, the third storage device applies to the lock server for permission to perform a write operation of updating the first data to the second data to be written, and the lock server determines that the third storage device is not After the backup storage device, the permission requested by the third storage device is not granted.
  • the lock server preferentially grants the permission to perform the write data request to the foregoing backup storage device, and accordingly rejects the non-reserved storage device requesting the right to execute the foregoing write data request.
  • the conflicting authority is used to preferentially implement the write data request received by the first storage device to be successfully executed, thereby ensuring the reliability of the system.
  • the write data request received by the first storage device is specifically used to request to update the first data to the data to be written, and the method for managing the distributed storage system
  • the method further includes the following steps: after the first storage device fails, the third storage device is locked to the lock
  • the server applies for permission to read the read operation of the first data, and the lock server does not grant the third storage device the right to apply after determining that the third storage device is not the standby storage device.
  • the lock server preferentially grants the permission to perform the write data request to the foregoing backup storage device, and accordingly rejects the non-reserved storage device requesting the right to execute the foregoing write data request.
  • the conflicting authority is used to preferentially implement the write data request received by the first storage device to be successfully executed, thereby ensuring the reliability of the system.
  • the method for managing the distributed storage system includes: the lock server and the first storage device determine, by negotiation, that the content of the preset field in the first write lock request is an identifier of the backup storage device that performs the write data request, and the first storage device sends the first write lock to the lock server.
  • the identifier of the second storage device is added to the preset field of the request, so that the lock server identifies the identifier of the second storage device as the identifier of the standby storage device. .
  • the lock server can identify the identifier of the standby storage device in the first write lock request by negotiating the lock server with the first storage device, so that the backup storage device preferentially obtains the permission to execute the write data request. may.
  • the first storage device is specifically configured to pass the second The mirrored channel between the storage devices sends the data to be written to the second storage device, and the second storage device writes the received data to be written to the cache of the second storage device.
  • sending data to be written through the mirroring channel not only reduces the time required for sending the data to be written, but also improves the security of the data.
  • the embodiment of the present invention provides a distributed storage system, including: at least two storage devices and a lock server, each storage device includes a cache and a hard disk; wherein, the first storage device of the at least two storage devices For receiving a write data request, the write data request includes data to be written, and sending a first write lock request to the lock server, where the first write lock request is used to apply for executing a write data request, and the first write lock request includes the first The identifier of the storage device and the second storage device of the at least two storage devices
  • the second storage device is an alternate storage device that performs a write data request;
  • the lock server is configured to: receive the first write lock request, send a response result of the first write lock request to the first storage device, and the first write lock request The result of the response is used to indicate that the first storage device obtains the permission to execute the write data request, and confirms that the identifier of the second storage device in the first write lock request is the identifier of the standby storage device, and stores the identifier
  • the write data request is used to request to update the first data to the data to be written, and the write data request further includes the identifier of the first data;
  • the system further includes: a third storage device, configured to: after the first storage device fails, receive a second write data request, where the second write data request is used to request to update the first data to the second data to be written, The second write data request includes an identifier of the first data and a second data to be written; a third write lock request is sent to the lock server, and the third write lock request is used to apply for performing the update of the first data to the second data to be written.
  • the third write lock request includes the identifier of the third storage device; the lock server is further configured to: receive the third write lock request, and determine that the identifier of the third storage device carried by the third write lock request is not the backup storage device And sending a response result of the third write lock request to the third storage device, where the response result of the third write lock request is used to indicate that the third storage device is not obtained to update the first data to the first Write permissions for data to be written.
  • the write data request is for The request is to update the first data to the data to be written, and the write data request further includes the identifier of the first data.
  • the distributed storage system further includes: a third storage device, configured to: after the first storage device fails, receive and read a read request of the first data, the read request includes an identifier of the first data; a fourth read lock request is sent to the lock server, and the fourth read lock request is used to apply for performing a read operation of reading the first data, the fourth The read lock request includes the identifier of the third storage device; the lock server is further configured to: receive the fourth read lock request, and determine that the identifier of the third storage device carried by the fourth read lock request is not the identifier of the standby storage device; The storage device sends a response result of the fourth read lock request, and the response result of the fourth read lock request is used to indicate that the third storage device does not obtain the permission to perform the read operation of reading the first data.
  • the preset field of the first write lock request The identifier of the second storage device is included; the lock server is further configured to: before confirming that the identifier of the second storage device is the identifier of the standby storage device, agree with the first storage device, in a preset field in the first write lock request Identifies the identity of the alternate storage device that is executing the write data request.
  • the first storage device passes the second storage
  • the mirrored channel between devices sends the write data to the second storage device.
  • FIG. 1 is a schematic diagram of a distributed storage system according to an embodiment of the present invention.
  • FIG. 2 is a schematic flowchart of a method for managing a distributed storage system according to an embodiment of the present invention
  • FIG. 3 is another schematic flowchart of a method for managing a distributed storage system according to an embodiment of the present invention.
  • FIG. 4 is still another schematic flowchart of a method for managing a distributed storage system according to an embodiment of the present invention.
  • the distributed storage system includes a plurality of storage devices, and each storage device may include an interface, a cache, a processor, and a hard disk.
  • the interface is configured to receive a request sent by the client.
  • the interface may be a transmission control protocol/Internet Protocol (TCP/IP)-based communication interface.
  • TCP/IP transmission control protocol/Internet Protocol
  • the cache may be a separate storage unit, such as a non-Volatile Random Access Memory (NVRAM) in the storage device, or a storage area in the memory of the storage device.
  • NVRAM non-Volatile Random Access Memory
  • the processor may include one or more processing units.
  • the processor may be a central processing unit (CPU), or may be an application specific integrated circuit (ASIC), or may be One or more integrated circuits configured to implement embodiments of the present invention, such as one or more microprocessors (DSPs), or one or more Field Programmable Gate Arrays Abbreviation: FPGA).
  • CPU central processing unit
  • ASIC application specific integrated circuit
  • DSPs microprocessors
  • FPGA Field Programmable Gate Arrays Abbreviation: FPGA
  • the processor is configured to invoke an instruction in the memory to perform the steps included in the method for managing the distributed storage system in the embodiment of the present invention.
  • the hard disk is used to store data to be written, and the data request is successfully executed after the data to be written is written to the hard disk.
  • the storage device further includes a buffer, which may be a separate storage unit. It can also be a storage area in the memory of the storage device, and the buffer unit is used to store a request received by the interface of the storage device.
  • a buffer which may be a separate storage unit. It can also be a storage area in the memory of the storage device, and the buffer unit is used to store a request received by the interface of the storage device.
  • the storage device performs a write data request by using a distributed transaction.
  • the storage device receives a write data request through the interface.
  • the processor executes the instruction, and the storage device performs the following steps: first, save the write data request in the buffer, and then apply to the lock server for permission to write the data request, and after obtaining the permission, store the buffer.
  • the write data request contains the data to be written to the cache, and then the data to be written stored in the cache is written to the hard disk.
  • the distributed storage system further includes a lock server, and the functions thereof include: maintaining the state of the lock authority of each data resource, processing the permission of the application sent by the storage device to read/write the data resource, and according to the state of the lock authority of the data resource. Grant the storage device permission to apply for it, or deny permission to apply.
  • FIG. 2 is a schematic flowchart of a method for managing a distributed storage system according to an embodiment of the present disclosure, where the process includes the following steps:
  • Step 101 The first storage device receives a write data request, where the write data request includes data to be written.
  • Step 102 The first storage device sends a first write lock request to the lock server, where the first write lock request is used to apply for permission to write a data request, where the first write lock request includes the identifier of the first storage device and the second storage device.
  • the second storage device is an alternate storage device that performs a write data request.
  • the first storage device after receiving the write data request, applies to the lock server for permission to execute the write data request, and the first write lock request sent to the lock server includes not only its own identifier but also the second storage device.
  • the manner in which the first storage device selects the second storage device as the backup storage device that performs the write data request includes:
  • the distributed storage system includes a primary storage device for performing load balancing between the storage devices, including allocating spare storage devices.
  • the first storage device After receiving the write data request, the first storage device sends the allocated backup storage to the primary storage device.
  • the request of the device the primary storage device may allocate the second storage device as a backup storage device for executing the write data request according to factors such as the load condition of each storage device or the communication quality between the storage device and the first storage device.
  • the first storage device may broadcast a backup storage device request to other storage devices in the distributed storage system, and the storage device receiving the request selects whether to or not according to factors such as its own load or communication quality with the first storage device.
  • the first storage device returns a confirmation response.
  • the first storage device may confirm that the storage device that returns the acknowledgment response is the backup storage device, or determine that the storage device with the smallest current load in the storage device that returns the acknowledgment response is the standby storage device, or determine the storage device that returns the acknowledgment response.
  • the storage device with the best communication quality with itself is the above-mentioned backup storage device, and so on, which are not exemplified herein.
  • Step 103 The lock server confirms that the identifier of the second storage device is an identifier of the standby storage device, and stores the identifier of the standby storage device.
  • the lock server can resolve the identifier of the second storage device in the first write lock request to the identifier of the backup storage device that performs the write data request. Its implementation can be:
  • the lock server and the first storage device determine, by negotiation, the content of the preset field in the first write lock request as the identifier of the backup storage device that performs the write data request.
  • the negotiation process may be performed when the first storage device joins the distributed storage system, or when the lock server is connected to the distributed storage system, or the first storage device sends the first write lock request to the lock server. Completed before.
  • the identifier of the second storage device is added to the negotiated preset field, and the lock server can identify the identifier of the second storage device as the backup storage device that performs the write data request.
  • the first storage device stores the identifier of the first storage device and the identifier of the second storage device in the storage device identifier field, where the identifier of the first storage device is first.
  • the first label in the storage device identification field is The identification is identified as the identification of the storage device performing the write data request, and the identification following the first identification is identified as the identification of the alternate storage device performing the write data request.
  • Step 104 After the first storage device obtains the permission, the first storage device writes the data to be written into the cache of the first storage device; and the first storage device sends the data to be written to the second storage device.
  • the lock server in response to the first write lock request, grants the first storage device permission to perform a write data request, and after obtaining the permission, the first storage device writes the data to be written into the cache of the first storage device, and Sending data to be written to the second storage device.
  • the first storage device may send the data to be written to the second storage device through a public network that communicates between the storage devices.
  • the first storage device can also send the data to be written to the second storage device through the mirroring channel with the second storage device, and send the data to be written through the mirroring channel, which not only reduces the time required for sending the data to be written, but also can Improve the security of your data.
  • the first storage device may perform an operation of writing data to be written to the local cache (ie, the cache of the first storage device) and transmitting the data to be written to the second storage device in parallel, or may first write the data to be written.
  • the data is written to the local cache, and after the local cache is successfully written, the data to be written is sent to the second storage device.
  • Step 105 The second storage device receives the data to be written sent by the first storage device, and writes the data to be written into the cache of the second storage device.
  • the second storage device receives the data to be written sent by the first storage device, and writes the data to be written to the local cache of the second storage device, so as to continue according to the data to be written in the cache after the first storage device fails. Execute a write data request.
  • Step 106 When it is determined that the first storage device fails, the second storage device sends a second write lock request to the lock server, the second write lock request is used to apply for permission to write a data request, and the second write lock request includes the second storage. The identity of the device.
  • the second storage device when determining that the first storage device is faulty, saves data to be written, The write data request can continue to be executed, but the permission to execute the write data request needs to be obtained, and therefore, the second storage device sends a request to the lock server to execute the write data request.
  • the second storage device determines that the first storage device is faulty, including but not limited to the following implementation manners:
  • the primary storage device in the distributed storage system can determine the state of each secondary storage device in the distributed storage system by heartbeat detection, and after detecting that the first storage device fails, to each of the distributed storage systems A notification is sent from the storage device (including the second storage device) indicating that the first storage device has failed.
  • the storage device that receives the notification may send a request to the lock server to apply for permission to execute the write data request received by the first storage device, where the second storage device is a backup storage device that performs the write data request, and is capable of The server gets this privilege, and the rest of the storage devices won't get that privilege.
  • a variation of the above mode 1 is that, in the storage device that receives the notification, only the second pure device storing the data to be written transmits a right to request execution of the write data request to the lock server.
  • Another variation of the foregoing mode 1 is that after the primary storage device determines that the first storage device is faulty by the heartbeat detection, the primary storage device transmits the notification only to the second storage device that stores the data to be written.
  • the second storage device After the second storage device receives the data to be written, it determines that it is the backup storage device that performs the write data request, so the second storage device actively detects the state of the first storage device (the heartbeat detection mode may be adopted), so that When the first storage device fails, the lock server can be requested to execute the write data request in time to enable the write data request to be completed in time.
  • the heartbeat detection mode may be adopted
  • Step 107 The lock server confirms that the identifier of the second storage device is consistent with the identifier of the stored backup storage device, and sends a response result of the second write lock request to the second storage device, where the response result is used to indicate that the second storage device obtains the execution of the write. Permissions for data requests.
  • the lock server determines whether the storage device identifier carried by the second write lock request is consistent with the previously stored spare storage device identifier, and after determining that the two are consistent, the lock server grants the second The storage device performs the permission to write data requests.
  • Step 108 After the second storage device obtains the permission, the second storage device writes the data to be written. Enter the hard disk of the second storage device.
  • the second storage device writes the data to be written saved in the local cache to the hard disk of the second storage device, and performs a write data request succeeding.
  • the second storage device can write the data to be written into the hard disk of other storage devices in the distributed storage system.
  • the write data request may include information of the target storage location, and the second storage device writes the data to be written to the target hard disk according to the information.
  • the lock server when the first storage device receives the write data request and requests the lock server to perform the write data request, the lock server is configured to save the identifier of the second storage device, and the identifier of the second storage device is used as the write data request.
  • the ID of the alternate storage device After obtaining the permission to execute the write data request, the first storage device stores the data to be written in the local cache, and sends the data to be written to the second storage device.
  • the second storage device stores the data to be written sent by the first storage device in a local cache to apply to the lock server for permission to write a data request after the first storage device fails.
  • the lock server Since the identifier of the alternate storage device that performs the write data request saved by the lock server is consistent with the identifier of the second storage device, the lock server will grant the second storage device permission to write the data request. After obtaining the permission, the second storage device can write the data to be written in the local cache to the hard disk, and the completion of the write data request is successful. Therefore, in the case where the first storage device that performs the write data request is obtained, the write data request can still be continued by the second storage device as the backup storage device that performs the write data request, ensuring that the write data request is successfully executed, improving System reliability.
  • the write data request received by the first storage device is used to request to update the first data to the data to be written, where the write data request further includes an identifier of the first data.
  • the method for managing a distributed storage system further includes the following steps:
  • Step 201 After the first storage device fails, the third storage device sends a third write lock request to the lock server, and the third write lock request is used to apply for performing a write operation of updating the first data to the second data to be written. Permission, the third write lock request includes an identifier of the third storage device;
  • Step 202 The lock server determines that the identifier of the third storage device is not the identifier of the standby storage device.
  • Step 203 The lock server sends a response result of the third write lock request to the third storage device, and third.
  • the response result of the write lock request is used to indicate that the third storage device has not obtained the right to perform a write operation of updating the first data to the second data to be written.
  • the distributed storage system further includes a third storage device, wherein the third storage device is capable of receiving a second write data request, and the second write data request and the write data request received by the first storage device are both requested to the first data.
  • the first data in the embodiment of the present invention refers to data resources, not numerical values.
  • the third storage device After receiving the second write data request, the third storage device requests the lock server to execute the second write data request.
  • write operations for the same data are mutually exclusive, that is, only one storage device at a time can obtain the right to write a data request (data update request) to the first data.
  • the lock server determines that the first storage device has the right to write the data request for the first data, and accordingly refuses to execute the second write data. The requested permission is granted to the third storage device.
  • the lock server determines that the third storage device is not executing the write data. After the requested spare storage device, the permission to execute the second write data request is denied to the third storage device.
  • the lock server preferentially grants the foregoing backup storage device the right to write the first data requested to be updated by the write data request, and accordingly rejects the application of the non-backup storage device.
  • the permission to execute the foregoing write data request conflicts with priority, so that the write data request received by the first storage device is successfully executed, and the reliability of the system is ensured.
  • the write data request received by the first storage device is used to request to update the first data to the data to be written, where the write data request further includes an identifier of the first data.
  • the method for managing a distributed storage system further includes the following steps:
  • Step 301 After the first storage device fails, the third storage device sends a fourth read lock request to the lock server, and the fourth read lock request is used to apply for permission to read the read operation of the first data.
  • the read lock request includes an identifier of the third storage device;
  • Step 302 The lock server determines that the identifier of the third storage device is not the identifier of the standby storage device.
  • Step 303 The lock server sends a response result of the fourth read lock request to the third storage device, and the response result of the fourth read lock request is used to indicate that the third storage device does not obtain the permission to perform the read operation of reading the first data.
  • the third storage device is capable of receiving a read data request for reading the first data, because the permission to perform the write operation is an exclusive right, and the authority to execute the read data request is exclusive to the permission to execute the foregoing write data request.
  • the third storage device After receiving the second write data request, the third storage device requests the lock server to execute the second write data request. If the third storage device sends the fourth read lock request before the first storage device fails, the lock server determines that the first storage device has the right to write the data request for the first data, and accordingly refuses to perform the read data request. The permission is granted to the third storage device.
  • the lock server determines that the third storage device is not executing the write data. After the requested alternate storage device, the permission to perform the read data request is denied to the third storage device.
  • the lock server preferentially grants the foregoing backup storage device the right to write the first data requested to be updated by the write data request, and accordingly rejects the application of the non-backup storage device.
  • the permission to execute the foregoing write data request conflicts with priority, so that the write data request received by the first storage device is successfully executed, and the reliability of the system is ensured.
  • the second storage device may be a collective name of multiple backup storage devices that perform write data requests, that is, the number of the backup storage devices may be two or more.
  • the permission to write the data request may be competed by the plurality of spare storage devices.
  • the primary storage device specifies one of the backup storage devices according to factors such as load balancing to perform step 106, and requests the lock server to execute the permission to write the data request.
  • a plurality of spare storage devices have priority ordering, and the lock server preferentially grants permission to perform data write requests. Give priority to the highest priority spare storage device.
  • an embodiment of the present invention provides a distributed storage system.
  • the system includes: at least two storage devices and a lock server, each of which includes a cache and a hard disk;
  • the first storage device of the at least two storage devices is configured to: receive a write data request, the write data request includes data to be written; send a first write lock request to the lock server, where the first write lock request is used to apply for writing
  • the permission of the data request, the first write lock request includes an identifier of the first storage device and an identifier of the second storage device of the at least two storage devices, and the second storage device is a backup storage device that performs a write data request;
  • the lock server is configured to: receive a first write lock request, and send a response result of the first write lock request to the first storage device, where the response result of the first write lock request is used to instruct the first storage device to obtain the permission to execute the write data request, And confirming that the identifier of the second storage device in the first write lock request is an identifier of the standby storage device, and storing the identifier of the standby storage device;
  • the first storage device is further configured to: after receiving the response result of the first write lock request, write the data to be written into the cache of the first storage device; and send the data to be written to the second storage device;
  • the second storage device is configured to: receive data to be written sent by the first storage device, store the data to be written in a cache of the second storage device; and send a message to the lock server when determining that the first storage device is faulty a second write lock request, the second write lock request is used to apply for permission to write a data request, and the second write lock request includes an identifier of the second storage device;
  • the lock server is further configured to: receive a second write lock request; and send a second write lock request to the second storage device when confirming that the identifier of the second storage device carried by the second write lock request is consistent with the identifier of the stored backup storage device
  • the response result is used to indicate that the second storage device obtains the permission
  • the second storage device is further configured to: receive the response result of the second write lock request, and write the data to be written into the hard disk of the second storage device.
  • the write data request received by the first storage device is used to request to update the first data to the data to be written, and the write data request further includes the identifier of the first data;
  • the storage system also includes:
  • a third storage device configured to: after the first storage device fails, receive a second write data request, where the second write data request is used to request to update the first data to the second data to be written, and the second write data request The identifier of the first data and the second data to be written are included; the third write lock request is sent to the lock server, and the third write lock request is used to apply for permission to perform the write operation of updating the first data to the second data to be written.
  • the third write lock request includes an identifier of the third storage device;
  • the lock server is further configured to: receive a third write lock request, determine that the identifier of the third storage device carried by the third write lock request is not the identifier of the standby storage device; and send a response result of the third write lock request to the third storage device, The response result of the triple write lock request is used to instruct the third storage device not to obtain the right to perform a write operation of updating the first data to the second data to be written.
  • the write data request is used to request to update the first data to the data to be written, and the write data request further includes the identifier of the first data.
  • the system further includes:
  • a third storage device configured to: after the first storage device fails, receive a read request for reading the first data, the read request includes an identifier of the first data; and send a fourth read lock request to the lock server, the fourth read lock Requesting permission to apply for performing a read operation of reading the first data, the fourth read lock request including an identifier of the third storage device;
  • the lock server is further configured to: receive a fourth read lock request, determine that the identifier of the third storage device carried by the fourth read lock request is not the identifier of the standby storage device; and send a response result of the fourth read lock request to the third storage device, The response result of the four read lock request is used to indicate that the third storage device has not obtained the permission to perform the read operation of reading the first data.
  • the preset field of the first write lock request includes an identifier of the second storage device
  • the lock server is further configured to: before confirming that the identifier of the second storage device is the identifier of the standby storage device, agree with the first storage device that the identifier in the preset field in the first write lock request is an alternate storage that performs a write data request. The identity of the device.
  • the first storage device is configured to: send data to be written to the second Storage devices, including:
  • the tape write data is transmitted to the second storage device through a mirrored channel with the second storage device.
  • the distributed storage system in this embodiment and the foregoing method for managing the distributed storage system are based on two aspects under the same inventive concept.
  • the implementation process of the method has been described in detail above, so those skilled in the art can
  • the foregoing description clearly understands the structure and implementation process of the distributed storage system in this embodiment. For the sake of brevity of the description, no further description is provided herein.
  • the lock server When the first storage device receives the write data request and requests the lock server to execute the write data request, the lock server is instructed to save the identifier of the second storage device, and the identifier of the second storage device is used as the identifier of the backup storage device that performs the write data request. .
  • the first storage device After obtaining the permission to execute the write data request, the first storage device stores the data to be written in the local cache, and sends the data to be written to the second storage device.
  • the second storage device stores the data to be written sent by the first storage device in a local cache to apply to the lock server for permission to write a data request after the first storage device fails.
  • the lock server Since the identifier of the alternate storage device that performs the write data request saved by the lock server is consistent with the identifier of the second storage device, the lock server will grant the second storage device permission to write the data request. After obtaining the permission, the second storage device can write the data to be written in the local cache to the hard disk, and the completion of the write data request is successful. Therefore, in the case where the first storage device that performs the write data request is obtained, the write data request can still be continued by the second storage device as the backup storage device that performs the write data request, ensuring that the write data request is successfully executed, improving System reliability.
  • embodiments of the present invention can be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or a combination of software and hardware. Moreover, the invention can 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.) including computer usable program code.
  • computer-usable storage media including but not limited to disk storage, CD-ROM, optical storage, etc.
  • the present invention is directed to a method, apparatus (system), and computer program according to embodiments of the present application.
  • the flow chart and/or block diagram of the product is described. It will be understood that each flow and/or block of the flowchart illustrations and/or FIG.
  • These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing device to produce a machine for the execution of instructions for execution by a processor of a computer or other programmable data processing device.
  • the computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device.
  • the apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.

Abstract

一种管理分布式存储系统的方法及分布式存储系统，用于解决获得执行写数据请求的权限的存储设备发生故障后，该写数据请求无法被正常执行的问题。该方法包括：第一存储设备接收写数据请求，向锁服务器申请执行写数据请求的权限时，指示锁服务器保存第二存储设备的标识，将第二存储设备的标识作为执行写数据请求的备用存储设备的标识。第一存储设备在获得执行写数据请求的权限之后，将待写入数据存入本地缓存，并将待写入数据发送给第二存储设备。第二存储设备将第一存储设备发送的待写入数据存入本地缓存，以在第一存储设备发生故障后，向锁服务器申请执行写数据请求的权限。并在获得该权限之后，将本地缓存中的待写入数据写入硬盘。

Description

一种管理分布式存储系统的方法及分布式存储系统 技术领域
本发明涉及计算机技术领域，特别涉及一种管理分布式存储系统的方法及分布式存储系统。
背景技术
在分布式存储系统中，锁服务器负责管理多个存储设备对同一个资源的互斥访问。当存储设备需要对资源进行读/写操作时，需要向锁服务器申请相应的权限，只有在获得权限之后，存储设备才能执行对资源的对应的操作。
实际情况中，存储设备1在获得执行对数据资源的写操作的权限后，可能在写操作执行成功之前发生故障。这种情况下，若有存储设备2申请对该数据资源进行读操作的权限，或者申请对该数据进行另一写操作(请求写入的数据不同)的权限，锁服务器会收回授予存储设备1的权限，授予存储设备2执行其申请执行的操作的权限。这是因为锁服务器若让故障存储设备占用对数据资源进行操作的权限，将阻塞正常存储设备获得对该数据资源进行操作的权限，影响分布式存储系统的效率。
但是，收回故障存储设备拥有的对数据资源进行写操作的权限之后，将导致故障存储设备负责的该写操作无法完成，影响分布式系统的可靠性。
发明内容
本申请实施例提供一种管理分布式存储系统的方法及分布式存储系统，用于解决现有技术中获得执行写数据请求的权限的存储设备发生故障后，该写数据请求无法被正常执行的问题。
第一方面，本发明实施例提供一种管理分布式存储系统的方法，所述分布式存储系统包括多个存储设备和锁服务器，其中，每个存储设备均包括缓存和 硬盘，所述方法包括如下步骤：第一存储设备接收写数据请求，向锁服务器发送第一写锁请求，第一写锁请求用于申请执行写数据请求的权限，第一写锁请求包括第一存储设备的标识和第二存储设备的标识，其中，第二存储设备的标识能够被锁服务器识别为执行写数据请求的备用存储设备的标识。锁服务器授予第一存储设备执行写数据请求的权限，并存储备用存储设备的标识；第一存储设备获得该权限之后，将待写入数据写入第一存储设备的缓存；并且，将待写入数据发送给第二存储设备；第二存储设备将该待写入数据写入本地缓存，并在确定第一存储设备发生故障时，向锁服务器申请执行写数据请求的权限，锁服务器在确认第二存储设备为前述备用存储设备后，授予第二存储设备该权限；第二存储设备在获得该权限之后，将本地缓存中的待写入数据写入第二存储设备的硬盘中。
上述实现方式中，在获得执行写数据请求的第一存储设备故障的情况下，仍然能够通过作为执行写数据请求的备用存储设备的第二存储设备继续执行写数据请求，保证写数据请求被成功执行，提高系统的可靠性。
结合第一方面，在第一方面的第一种可能的实现方式中，第一存储设备接收的写数据请求具体用于请求将第一数据更新为待写入数据，管理分布式存储系统的方法还包括如下步骤：在第一存储设备发生故障后，第三存储设备向锁服务器申请执行将第一数据更新为第二待写入数据的写操作的权限，锁服务器在确定第三存储设备不是备用存储设备之后，不授予第三存储设备所申请的权限。上述可能的实现方式中，锁服务器在第一存储设备故障后，优先将执行写数据请求的权限授予前述备用存储设备，并据此拒绝非备用存储设备申请的与执行前述写数据请求的权限相冲突的权限，以优先实现第一存储设备接收的写数据请求被成功执行，保证系统的可靠性。
结合第一方面，在第一方面的第二种可能的实现方式中，第一存储设备接收的写数据请求具体用于请求将第一数据更新为待写入数据，管理分布式存储系统的方法还包括如下步骤：在第一存储设备发生故障后，第三存储设备向锁 服务器申请执行读取第一数据的读操作的权限，锁服务器在确定第三存储设备不是备用存储设备之后不授予第三存储设备所申请的权限。上述可能的实现方式中，锁服务器在第一存储设备故障后，优先将执行写数据请求的权限授予前述备用存储设备，并据此拒绝非备用存储设备申请的与执行前述写数据请求的权限相冲突的权限，以优先实现第一存储设备接收的写数据请求被成功执行，保证系统的可靠性。
结合第一方面以及第一方面的第一至第二种可能的实现方式中的任一实现方式，在本发明实施例的第三种可能的实现方式中，管理分布式存储系统的方法方法还包括：锁服务器与第一存储设备通过协商确定第一写锁请求中的预设字段的内容为执行写数据请求的备用存储设备的标识，第一存储设备在发送至锁服务器的第一写锁请求的预设字段中添加第二存储设备的标识，以使锁服务器将第二存储设备的标识识别为备用存储设备的标识。。上述可能的实现方式中，通过锁服务器与第一存储设备的协商，使得锁服务器能够识别第一写锁请求中的备用存储设备的标识，进而使得备用存储设备优先获得执行写数据请求的权限成为可能。
结合第一方面以及第一方面的第一至第二种可能的实现方式中的任一实现方式，在本发明实施例的第四种可能的实现方式中，第一存储设备具体通过与第二存储设备间的镜像通道将待写入数据发送给第二存储设备，第二存储设备将接收的待写入数据写入第二存储设备的缓存。上述可能的实现方式中，通过镜像通道发送待写入数据，不仅减少发送待写入数据的耗时，也能提高数据的安全。
第二方面，本发明实施例提供一种分布式存储系统，包括：至少两个存储设备以及锁服务器，每个存储设备均包括缓存和硬盘；其中，至少两个存储设备中的第一存储设备用于：接收写数据请求，写数据请求包括待写入数据；向锁服务器发送第一写锁请求，第一写锁请求用于申请执行写数据请求的权限，第一写锁请求包括第一存储设备的标识和至少两个存储设备中的第二存储设 备的标识，第二存储设备是执行写数据请求的备用存储设备；锁服务器用于：接收第一写锁请求，向第一存储设备发送第一写锁请求的响应结果，第一写锁请求的响应结果用于指示第一存储设备获得执行写数据请求的权限，并确认第一写锁请求中的第二存储设备的标识是备用存储设备的标识，并存储备用存储设备的标识；第一存储设备还用于：在接收第一写锁请求的响应结果之后，将待写入数据写入第一存储设备的缓存；并且，将待写入数据发送给第二存储设备；第二存储设备用于：接收第一存储设备发送的待写入数据，将待写入数据存储在第二存储设备的缓存中；以及在确定第一存储设备发生故障时，向锁服务器发送第二写锁请求，第二写锁请求用于申请执行写数据请求的权限，第二写锁请求包括第二存储设备的标识；锁服务器还用于：接收第二写锁请求；在确认第二写锁请求携带的第二存储设备的标识与存储的备用存储设备的标识一致时，向第二存储设备发送第二写锁请求的响应结果，响应结果用于指示第二存储设备获得权限；第二存储设备还用于：接收第二写锁请求的响应结果，将待写入数据写入第二存储设备的硬盘中。
结合第二方面，在第二方面的第一种可能的实现方式中，写数据请求用于请求将第一数据更新为待写入数据，写数据请求还包括第一数据的标识；分布式存储系统还包括：第三存储设备，用于：在第一存储设备发生故障后，接收第二写数据请求，第二写数据请求用于请求将第一数据更新为第二待写入数据，第二写数据请求包括第一数据的标识以及第二待写入数据；向锁服务器发送第三写锁请求，第三写锁请求用于申请执行将第一数据更新为第二待写入数据的写操作的权限，第三写锁请求包括第三存储设备的标识；锁服务器还用于：接收第三写锁请求，确定第三写锁请求携带的第三存储设备的标识不是备用存储设备的标识；向第三存储设备发送第三写锁请求的响应结果，第三写锁请求的响应结果用于指示第三存储设备未获得执行将第一数据更新为第二待写入数据的写操作的权限。
结合第二方面，在第二方面的第二种可能的实现方式中，写数据请求用于 请求将第一数据更新为待写入数据，写数据请求还包括第一数据的标识；分布式存储系统还包括：第三存储设备，用于：在第一存储设备发生故障后，接收读取第一数据的读请求，读请求包括第一数据的标识；向锁服务器发送第四读锁请求，第四读锁请求用于申请执行读取第一数据的读操作的权限，所述第四读锁请求包括所述第三存储设备的标识；锁服务器还用于：接收第四读锁请求，确定第四读锁请求携带的第三存储设备的标识不是备用存储设备的标识；向第三存储设备发送第四读锁请求的响应结果，第四读锁请求的响应结果用于指示第三存储设备未获得执行读取第一数据的读操作的权限。
结合第二方面以及第二方面的第一至第二种可能的实现方式中的任一实现方式，在本发明实施例的第三种可能的实现方式中，第一写锁请求的预设字段中包含第二存储设备的标识；锁服务器还用于：在确认第二存储设备的标识是备用存储设备的标识之前，与第一存储设备约定，第一写锁请求中的预设字段中的标识为执行写数据请求的备用存储设备的标识。
结合第二方面以及第二方面的第一至第二种可能的实现方式中的任一实现方式，在本发明实施例的第四种可能的实现方式中，第一存储设备通过与第二存储设备间的镜像通道将带写入数据发送给第二存储设备。
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要介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域的普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。
图1为本发明实施例中分布式存储系统的示意图；
图2为本发明实施例中管理分布式存储系统的方法的流程示意图；
图3为本发明实施例中管理分布式存储系统的方法的另一流程示意图；
图4为本发明实施例中管理分布式存储系统的方法的又一流程示意图。
具体实施方式
下面通过附图以及具体实施例对本发明技术方案做详细的说明，应当理解本申请实施例以及实施例中的具体特征是对本发明技术方案的详细的说明，而不是对本发明技术方案的限定，在不冲突的情况下，本申请实施例以及实施例中的技术特征可以相互组合。
如图1所示，分布式存储系统包括若干个存储设备，每个存储设备可以包括接口，缓存(cache)，处理器以及硬盘。
其中，接口用于接收客户端发送的请求，例如，接口可以为基于传输控制协议/因特网互联协议(Transmission Control Protocol/Internet Protocol；简称：TCP/IP)的通信接口。
缓存可以为一单独的存储单元，如将存储设备中的非易失性随机访问存储器(Non-Volatile Random Access Memory；简称：NVRAM)，也可以为存储设备的内存中的一个存储区域，用于在存储设备获得执行写数据请求的权限后，对待写入数据进行缓存。
处理器可以包括一个或多个处理单元，例如，处理器可以为中央处理器(Central Processing Unit；简称：CPU)，也可以是特定集成电路(Application Specific Intergrated Circuit；简称：ASIC)，或者是被配置成实施本发明实施例的一个或多个集成电路，例如：一个或多个微处理器(digital singnal processor；简称：DSP)，或，一个或者多个现场可编程门阵列(Field Programmable Gate Array；简称：FPGA)。
在本发明实施例中，处理器用于调用内存中的指令以执行本发明实施例中管理分布式存储系统的方法包括的步骤。
硬盘用于存储待写入数据，在待写入数据写入硬盘之后，写数据请求才被成功执行。
另外，存储设备还包括缓冲区(buffer)，该缓冲区可以为单独的存储单元， 也可以为存储设备的内存中的一个存储区域，缓冲区单元用于存储存储设备的接口接收的请求。
本发明实施例中，存储设备采用分布式事务执行写数据请求。首先，存储设备通过接口接收写数据请求。然后，通过处理器执行指令，存储设备执行如下步骤：首先，将写数据请求保存在缓冲区之中，然后，向锁服务器申请执行写数据请求的权限，在获得该权限之后，将缓冲区存储的写数据请求包含的待写入数据写入缓存，然后，再将缓存中存储的待写入数据写入硬盘。
分布式存储系统还包括锁服务器，其作用包括：维护每个数据资源的锁权限的状态，处理存储设备发送的申请对数据资源进行读/写操作的权限，并根据数据资源的锁权限的状态向存储设备授予其申请的权限，或者，拒绝其申请的权限。
图2为本发明实施例提供的管理分布式存储系统的方法的流程示意图，该流程包括如下步骤：
步骤101：第一存储设备接收写数据请求，该写数据请求包括待写入数据。
步骤102：第一存储设备向锁服务器发送第一写锁请求，第一写锁请求用于申请执行写数据请求的权限，第一写锁请求包括第一存储设备的标识和第二存储设备的标识，第二存储设备是执行写数据请求的备用存储设备。
具体的，第一存储设备接收写数据请求之后，向锁服务器申请执行该写数据请求的权限，在发送给锁服务器的第一写锁请求中不仅包含自身的标识，还包括第二存储设备的标识，用于表示第二存储设备为执行写数据请求的备用存储设备。
其中，第一存储设备选择第二存储设备作为执行写数据请求的备用存储设备的方式包括：
其一，分布式存储系统中包含主存储设备，用于进行存储设备间的负载均衡，包括分配备用存储设备。
第一存储设备在接收到写数据请求之后，向主存储设备发送分配备用存储 设备的请求，主存储设备可以根据各存储设备的负载情况或者个存储设备与第一存储设备间的通信质量等因素，为其分配第二存储设备作为执行写数据请求的备用存储设备。
其二，第一存储设备可以向分布式存储系统中的其他存储设备广播备用存储设备请求，接收该请求的存储设备根据自身的负载或与第一存储设备间的通信质量等因素，选择是否向第一存储设备返回确认响应。
第一存储设备可以确认最早返回确认响应的存储设备为上述备用存储设备，或者，确定返回确认响应的存储设备中当前负载最小的存储设备为上述备用存储设备，或者，确定返回确认响应的存储设备中与自己通信质量最好的存储设备为上述备用存储设备，等等，在此不一一举例。
步骤103：锁服务器确认第二存储设备的标识是备用存储设备的标识，并存储备用存储设备的标识。
具体的，锁服务器接收第一写锁请求之后，能够将第一写锁请求中的第二存储设备的标识解析为执行该写数据请求的备用存储设备的标识。其实现方式可以为：
其一，锁服务器与第一存储设备通过协商确定第一写锁请求中的预设字段的内容为执行写数据请求的备用存储设备的标识。该协商的过程可以在第一存储设备加入分布式存储系统时完成的，也可以是锁服务器与分布式存储系统连接时进行的，还可以是第一存储设备向锁服务器发送第一写锁请求之前完成的。
第一存储设备在生成第一写锁请求时，在协商出的预设字段中加入第二存储设备的标识，锁服务器即可将第二存储设备的标识识别为执行写数据请求的备用存储设备的标识。
其二，第一存储设备在生成第一写锁请求时，将第一存储设备的标识以及第二存储设备的标识顺序存放在存储设备标识字段中，其中，第一存储设备的标识在前。锁服务器接收第一写锁请求后，将存储设备标识字段中的第一个标 识识别为执行写数据请求的存储设备的标识，而将第一个标识之后的标识识别为执行写数据请求的备用存储设备的标识。
步骤104：在第一存储设备获得权限之后，第一存储设备将待写入数据写入第一存储设备的缓存；并且，第一存储设备将待写入数据发送给第二存储设备。
具体的，锁服务器响应第一写锁请求，授予第一存储设备执行写数据请求的权限，而第一存储设备在获得该权限之后，将待写入数据写入第一存储设备的缓存，并向第二存储设备发送待写入数据。
其中，第一存储设备可以通过存储设备间通信的公用网络将待写入数据发送给第二存储设备。
第一存储设备也可以通过与第二存储设备间的镜像通道将待写入数据发送给第二存储设备，通过镜像通道发送待写入数据，不仅减少发送待写入数据的耗时，也能提高数据的安全。
另外，第一存储设备可以并行执行将待写入数据写入本地缓存(即，第一存储设备的缓存)以及将待写入数据发送给第二存储设备的操作，也可以先将待写入数据写入本地缓存，待写入本地缓存成功之后，再向第二存储设备发送待写入数据。
步骤105：第二存储设备接收第一存储设备发送的待写入数据，将待写入数据写入第二存储设备的缓存。
具体的，第二存储设备接收第一存储设备发送的待写入数据，将其写入第二存储设备的本地缓存，以在第一存储设备故障后，能够根据缓存中的待写入数据继续执行写数据请求。
步骤106：确定第一存储设备发生故障时，第二存储设备向锁服务器发送第二写锁请求，第二写锁请求用于申请执行写数据请求的权限，第二写锁请求包括第二存储设备的标识。
具体的，在确定第一存储设备故障时，第二存储设备保存有待写入数据， 能够继续执行写数据请求，但是需要获得执行写数据请求的权限，因此，第二存储设备向锁服务器发送申请执行该写数据请求的权限。
其中，第二存储设备确定第一存储设备故障，包括但不限于以下实现方式：
方式1，分布式存储系统中的主存储设备能够通过心跳检测确定分布式存储系统中每个从存储设备的状态，在检测到第一存储设备发生故障后，向分布式存储系统中的每个从存储设备(包括第二存储设备)发送通知，指明第一存储设备发生故障。
而接收到该通知的存储设备均可以向锁服务器发送申请执行第一存储设备接收的写数据请求的权限的请求，其中的第二存储设备为执行该写数据请求的备用存储设备，能够从锁服务器获得该权限，而其余的存储设备将不能获得该权限。
上述方式1的一种变形为：而接收到该通知的存储设备中，只有存储有待写入数据的第二纯出设备会向锁服务器发送申请执行该写数据请求的权限。
上述方式1的另一种变形为：主存储设备在通过心跳检测确定第一存储设备故障后，只向存储有待写入数据的第二存储设备发送上述通知。
方式2，第二存储设备接收待写入数据之后，确定自身作为执行写数据请求的备用存储设备，故第二存储设备主动检测第一存储设备的状态(可以采用心跳检测的方式)，以便在第一存储设备故障时，能够及时向锁服务器申请执行写数据请求的权限，以使写数据请求能够及时完成。
步骤107：锁服务器确认第二存储设备的标识与存储的备用存储设备的标识一致时，向第二存储设备发送第二写锁请求的响应结果，响应结果用于指示第二存储设备获得执行写数据请求的权限。
具体的，锁服务器在接收到上述第二写锁请求之后，判断第二写锁请求携带的存储设备标识与之前存储的备用存储设备标识是否一致，在确定二者一致之后，锁服务器授予第二存储设备执行写数据请求的权限。
步骤108：在第二存储设备获得权限之后，第二存储设备将待写入数据写 入第二存储设备的硬盘中。
具体的，第二存储设备在获得该权限之后，将本地缓存中保存的待写入数据写入第二存储设备的硬盘，执行写数据请求成功。
一种可能的变形方案为：第二存储设备可以将待写入数据写入分布式存储系统中其他存储设备的硬盘中。例如，写数据请求中可能包含有目标存储位置的信息，第二存储设备根据该信息将待写入数据写入目标硬盘。
上述技术方案中，第一存储设备接收写数据请求，向锁服务器申请执行写数据请求的权限时，指示锁服务器保存第二存储设备的标识，将第二存储设备的标识作为执行写数据请求的备用存储设备的标识。第一存储设备在获得执行写数据请求的权限之后，将待写入数据存入本地缓存，并将待写入数据发送给第二存储设备。第二存储设备将第一存储设备发送的待写入数据存入本地缓存，以在第一存储设备发生故障后，向锁服务器申请执行写数据请求的权限。由于锁服务器保存的执行写数据请求的备用存储设备的标识与第二存储设备的标识一致，锁服务器将授予第二存储设备执行写数据请求的权限。第二存储设备在获得该权限之后，即可将本地缓存中的待写入数据写入硬盘，完成写数据请求执行成功。因此，在获得执行写数据请求的第一存储设备故障的情况下，仍然能够通过作为执行写数据请求的备用存储设备的第二存储设备继续执行写数据请求，保证写数据请求被成功执行，提高系统的可靠性。
可选的，本发明实施例中，第一存储设备接收的写数据请求用于请求将第一数据更新为待写入数据，该写数据请求还包括第一数据的标识。参见图3，管理分布式存储系统的方法还包括如下步骤：
步骤201：在第一存储设备发生故障后，第三存储设备向锁服务器发送第三写锁请求，第三写锁请求用于申请执行将第一数据更新为第二待写入数据的写操作的权限，第三写锁请求包括第三存储设备的标识；
步骤202：锁服务器确定第三存储设备的标识不是备用存储设备的标识；
步骤203：锁服务器向第三存储设备发送第三写锁请求的响应结果，第三 写锁请求的响应结果用于指示第三存储设备未获得执行将第一数据更新为第二待写入数据的写操作的权限。
具体的，分布式存储系统还包括第三存储设备，第三存储设备能够接收第二写数据请求，该第二写数据请求与前述第一存储设备接收的写数据请求均对请求对第一数据进行更新，需要说明的是，本发明实施例中的第一数据，指的是数据资源，而不是数值。
第三存储设备在接收到该第二写数据请求之后，要向锁服务器申请执行第二写数据请求的权限。但是，针对同一数据的写操作具有互斥性，即，同一时刻只能有一个存储设备获得对第一数据进行写数据请求(数据更新请求)的权限。
如果第三存储设备在第一存储设备故障前发送上述第三写锁请求，锁服务器确定第一存储设备正拥有对第一数据进行写数据请求的权限，并据此拒绝将执行第二写数据请求的权限授予第三存储设备。
如果第三存储设备在第一存储设备故障后发送上述第三写锁请求，由于锁服务器需要优先将执行前述写数据请求的权限授予备用存储设备，锁服务器在确定第三存储设备不是执行写数据请求的备用存储设备之后，拒绝将执行第二写数据请求的权限授予第三存储设备。
上述技术方案中，锁服务器在第一存储设备故障后，优先将对写数据请求所请求更新的第一数据进行写操作的权限授予前述备用存储设备，并据此拒绝非备用存储设备申请的与执行前述写数据请求的权限相冲突的权限，以优先实现第一存储设备接收的写数据请求被成功执行，保证系统的可靠性。
可选的，本发明实施例中，第一存储设备接收的写数据请求用于请求将第一数据更新为待写入数据，该写数据请求还包括第一数据的标识。参见图4，管理分布式存储系统的方法还包括如下步骤：
步骤301：在第一存储设备发生故障后，第三存储设备向锁服务器发送第四读锁请求，第四读锁请求用于申请执行读取第一数据的读操作的权限，第四 读锁请求包括第三存储设备的标识；
步骤302：锁服务器确定第三存储设备的标识不是备用存储设备的标识；
步骤303：锁服务器向第三存储设备发送第四读锁请求的响应结果，第四读锁请求的响应结果用于指示第三存储设备未获得执行读取第一数据的读操作的权限。
具体的，第三存储设备能够接收读取第一数据的读数据请求，因为执行写操作的权限为独占权限，执行该读数据请求的权限与执行前述写数据请求的权限相斥。
第三存储设备在接收到该第二写数据请求之后，要向锁服务器申请执行第二写数据请求的权限。如果第三存储设备在第一存储设备故障前发送上述第四读锁请求，锁服务器确定第一存储设备正拥有对第一数据进行写数据请求的权限，并据此拒绝将执行读数据请求的权限授予第三存储设备。
如果第三存储设备在第一存储设备故障后发送上述第四读锁请求，由于锁服务器需要优先将执行前述写数据请求的权限授予备用存储设备，锁服务器在确定第三存储设备不是执行写数据请求的备用存储设备之后，拒绝将执行读数据请求的权限授予第三存储设备。
上述技术方案中，锁服务器在第一存储设备故障后，优先将对写数据请求所请求更新的第一数据进行写操作的权限授予前述备用存储设备，并据此拒绝非备用存储设备申请的与执行前述写数据请求的权限相冲突的权限，以优先实现第一存储设备接收的写数据请求被成功执行，保证系统的可靠性。
可选的，本发明实施例中，第二存储设备可以为多个执行写数据请求的备用存储设备的统称，即，备用存储设备的数量可以为2个或以上。
这种情况下，在第一存储设备故障后，可以由多个备用存储设备竞争执行写数据请求的权限。或者，由主存储设备根据负载均衡等因素指定其中的一个备用存储设备执行步骤106，向锁服务器申请执行写数据请求的权限。又或者，多个备用存储设备具有优先级排序，锁服务器优先将执行写数据请求的权限授 予优先级最高的备用存储设备。
基于相同的发明构思，本发明实施例提供一种分布式存储系统，继续参照图1，该系统包括：至少两个存储设备以及锁服务器，每个存储设备均包括缓存和硬盘；
其中，至少两个存储设备中的第一存储设备用于：接收写数据请求，写数据请求包括待写入数据；向锁服务器发送第一写锁请求，第一写锁请求用于申请执行写数据请求的权限，第一写锁请求包括第一存储设备的标识和至少两个存储设备中的第二存储设备的标识，第二存储设备是执行写数据请求的备用存储设备；
锁服务器用于：接收第一写锁请求，向第一存储设备发送第一写锁请求的响应结果，第一写锁请求的响应结果用于指示第一存储设备获得执行写数据请求的权限，并确认第一写锁请求中的第二存储设备的标识是备用存储设备的标识，并存储备用存储设备的标识；
第一存储设备还用于：在接收第一写锁请求的响应结果之后，将待写入数据写入第一存储设备的缓存；并且，将待写入数据发送给第二存储设备；
第二存储设备用于：接收第一存储设备发送的待写入数据，将待写入数据存储在第二存储设备的缓存中；以及在确定第一存储设备发生故障时，向锁服务器发送第二写锁请求，第二写锁请求用于申请执行写数据请求的权限，第二写锁请求包括第二存储设备的标识；
锁服务器还用于：接收第二写锁请求；在确认第二写锁请求携带的第二存储设备的标识与存储的备用存储设备的标识一致时，向第二存储设备发送第二写锁请求的响应结果，响应结果用于指示第二存储设备获得权限；
第二存储设备还用于：接收第二写锁请求的响应结果，将待写入数据写入第二存储设备的硬盘中。
可选的，本发明实施例中，第一存储设备接收的所述写数据请求用于请求将第一数据更新为待写入数据，写数据请求还包括第一数据的标识；分布式存 储系统还包括：
第三存储设备，用于：在第一存储设备发生故障后，接收第二写数据请求，第二写数据请求用于请求将第一数据更新为第二待写入数据，第二写数据请求包括第一数据的标识以及第二待写入数据；向锁服务器发送第三写锁请求，第三写锁请求用于申请执行将第一数据更新为第二待写入数据的写操作的权限，第三写锁请求包括第三存储设备的标识；
锁服务器还用于：接收第三写锁请求，确定第三写锁请求携带的第三存储设备的标识不是备用存储设备的标识；向第三存储设备发送第三写锁请求的响应结果，第三写锁请求的响应结果用于指示第三存储设备未获得执行将第一数据更新为第二待写入数据的写操作的权限。
可选的，本发明实施例中，写数据请求用于请求将第一数据更新为待写入数据，写数据请求还包括第一数据的标识；系统还包括：
第三存储设备，用于：在第一存储设备发生故障后，接收读取第一数据的读请求，读请求包括第一数据的标识；向锁服务器发送第四读锁请求，第四读锁请求用于申请执行读取第一数据的读操作的权限，第四读锁请求包括第三存储设备的标识；
锁服务器还用于：接收第四读锁请求，确定第四读锁请求携带的第三存储设备的标识不是备用存储设备的标识；向第三存储设备发送第四读锁请求的响应结果，第四读锁请求的响应结果用于指示第三存储设备未获得执行读取第一数据的读操作的权限。
可选的，本发明实施例中，第一写锁请求的预设字段中包含第二存储设备的标识；
锁服务器还用于：在确认第二存储设备的标识是备用存储设备的标识之前，与第一存储设备约定，第一写锁请求中的预设字段中的标识为执行写数据请求的备用存储设备的标识。
可选的，本发明实施例中，第一存储设备用于：将待写入数据发送给第二 存储设备，包括：
通过与第二存储设备间的镜像通道将带写入数据发送给第二存储设备。
本实施例中的分布式存储系统与前述管理分布式存储系统的方法是基于同一发明构思下的两个方面，在前面已经对方法的实施过程作了详细的描述，所以本领域技术人员可根据前述描述清楚地了解本实施例中的分布式存储系统的结构及实施过程，为了说明书的简洁，在此就不再赘述了。
本申请实施例中提供的一个或多个技术方案，至少具有如下技术效果或优点：
第一存储设备接收写数据请求，向锁服务器申请执行写数据请求的权限时，指示锁服务器保存第二存储设备的标识，将第二存储设备的标识作为执行写数据请求的备用存储设备的标识。第一存储设备在获得执行写数据请求的权限之后，将待写入数据存入本地缓存，并将待写入数据发送给第二存储设备。第二存储设备将第一存储设备发送的待写入数据存入本地缓存，以在第一存储设备发生故障后，向锁服务器申请执行写数据请求的权限。由于锁服务器保存的执行写数据请求的备用存储设备的标识与第二存储设备的标识一致，锁服务器将授予第二存储设备执行写数据请求的权限。第二存储设备在获得该权限之后，即可将本地缓存中的待写入数据写入硬盘，完成写数据请求执行成功。因此，在获得执行写数据请求的第一存储设备故障的情况下，仍然能够通过作为执行写数据请求的备用存储设备的第二存储设备继续执行写数据请求，保证写数据请求被成功执行，提高系统的可靠性。
本领域内的技术人员应明白，本发明的实施例可提供为方法、系统、或计算机程序产品。因此，本发明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，本发明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。
本发明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。
尽管已描述了本发明的优选实施例，但本领域内的技术人员一旦得知了基本创造性概念，则可对这些实施例作出另外的变更和修改。所以，所附权利要求意欲解释为包括优选实施例以及落入本发明范围的所有变更和修改。
显然，本领域的技术人员可以对本发明进行各种改动和变型而不脱离本发明的精神和范围。这样，倘若本发明的这些修改和变型属于本发明权利要求及其等同技术的范围之内，则本发明也意图包含这些改动和变型在内。

Claims (10)

  1. 一种管理分布式存储系统的方法，其特征在于，所述分布式存储系统包括至少两个存储设备和锁服务器，其中，每个存储设备均包括缓存和硬盘，所述方法包括：
    第一存储设备接收写数据请求，所述写数据请求包括待写入数据；
    所述第一存储设备向所述锁服务器发送第一写锁请求，所述第一写锁请求用于申请执行所述写数据请求的权限，所述第一写锁请求包括所述第一存储设备的标识和第二存储设备的标识，所述第二存储设备是执行所述写数据请求的备用存储设备；
    所述锁服务器确认所述第二存储设备的标识是所述备用存储设备的标识，并存储所述备用存储设备的标识；
    在所述第一存储设备获得所述权限之后，所述第一存储设备将所述待写入数据写入所述第一存储设备的缓存；并且，所述第一存储设备将所述待写入数据发送给所述第二存储设备；
    确定所述第一存储设备发生故障时，所述第二存储设备向所述锁服务器发送第二写锁请求，所述第二写锁请求用于申请执行所述写数据请求的权限，所述第二写锁请求包括所述第二存储设备的标识；
    所述锁服务器确认所述第二存储设备的标识与存储的所述备用存储设备的标识一致时，向所述第二存储设备发送所述第二写锁请求的响应结果，所述响应结果用于指示所述第二存储设备获得执行所述写数据请求的权限；
    在所述第二存储设备获得所述权限之后，所述第二存储设备将所述待写入数据写入所述第二存储设备的硬盘中。
  2. 根据权利要求1所述的方法，其特征在于，所述写数据请求用于请求将第一数据更新为所述待写入数据，所述写数据请求还包括所述第一数据的标识；所述方法还包括：
    在所述第一存储设备发生故障后，第三存储设备向所述锁服务器发送第三 写锁请求，所述第三写锁请求用于申请执行将所述第一数据更新为第二待写入数据的写操作的权限，所述第三写锁请求包括所述第三存储设备的标识；
    所述锁服务器确定所述第三存储设备的标识不是所述备用存储设备的标识；
    所述锁服务器向所述第三存储设备发送所述第三写锁请求的响应结果，所述第三写锁请求的响应结果用于指示所述第三存储设备未获得执行将所述第一数据更新为所述第二待写入数据的写操作的权限。
  3. 根据权利要求1所述的方法，其特征在于，所述写数据请求用于请求将第一数据更新为所述待写入数据，所述写数据请求还包括所述第一数据的标识；所述方法还包括：
    在所述第一存储设备发生故障后，第三存储设备向所述锁服务器发送第四读锁请求，所述第四读锁请求用于申请执行读取所述第一数据的读操作的权限，所述第四读锁请求包括所述第三存储设备的标识；
    所述锁服务器确定所述第三存储设备的标识不是所述备用存储设备的标识；
    所述锁服务器向所述第三存储设备发送所述第四读锁请求的响应结果，所述第四读锁请求的响应结果用于指示所述第三存储设备未获得执行读取所述第一数据的读操作的权限。
  4. 根据权利要求1-3任一项所述的方法，其特征在于，所述第二存储设备的标识携带在所述第一写锁请求的预设字段中；
    所述方法还包括：
    所述锁服务器与所述第一存储设备通过协商确定所述第一写锁请求中的所述预设字段的内容为执行所述写数据请求的备用存储设备的标识。
  5. 根据权利要求1-3任一项所述的方法，其特征在于，所述第一存储设备将所述待写入数据发送给所述第二存储设备，包括：
    所述第一存储设备通过与所述第二存储设备间的镜像通道将所述待写入 数据发送给所述第二存储设备；
    所述方法还包括：所述第二存储设备将接收的所述待写入数据写入所述第二存储设备的缓存。
  6. 一种分布式存储系统，其特征在于，所述系统包括：至少两个存储设备以及锁服务器，每个存储设备均包括缓存和硬盘；
    其中，所述至少两个存储设备中的第一存储设备用于：接收写数据请求，所述写数据请求包括待写入数据；向所述锁服务器发送第一写锁请求，所述第一写锁请求用于申请执行所述写数据请求的权限，所述第一写锁请求包括所述第一存储设备的标识和所述至少两个存储设备中的第二存储设备的标识，所述第二存储设备是执行所述写数据请求的备用存储设备；
    所述锁服务器用于：接收所述第一写锁请求，向所述第一存储设备发送所述第一写锁请求的响应结果，所述第一写锁请求的响应结果用于指示所述第一存储设备获得执行所述写数据请求的权限，并确认所述第一写锁请求中的所述第二存储设备的标识是所述备用存储设备的标识，并存储所述备用存储设备的标识；
    所述第一存储设备还用于：在接收所述第一写锁请求的响应结果之后，将所述待写入数据写入所述第一存储设备的缓存；并且，将所述待写入数据发送给所述第二存储设备；
    所述第二存储设备用于：接收所述第一存储设备发送的所述待写入数据，将所述待写入数据存储在所述第二存储设备的缓存中；以及在确定所述第一存储设备发生故障时，向所述锁服务器发送第二写锁请求，所述第二写锁请求用于申请执行所述写数据请求的权限，所述第二写锁请求包括所述第二存储设备的标识；
    所述锁服务器还用于：接收所述第二写锁请求；在确认所述第二写锁请求携带的所述第二存储设备的标识与存储的所述备用存储设备的标识一致时，向所述第二存储设备发送所述第二写锁请求的响应结果，所述响应结果用于指示 所述第二存储设备获得权限；
    所述第二存储设备还用于：接收所述第二写锁请求的响应结果，将所述待写入数据写入所述第二存储设备的硬盘中。
  7. 根据权利要求6所述的系统，其特征在于，所述写数据请求用于请求将第一数据更新为所述待写入数据，所述写数据请求还包括所述第一数据的标识；所述系统还包括：
    第三存储设备，用于：在所述第一存储设备发生故障后，接收第二写数据请求，所述第二写数据请求用于请求将所述第一数据更新为第二待写入数据，所述第二写数据请求包括所述第一数据的标识以及所述第二待写入数据；向所述锁服务器发送第三写锁请求，所述第三写锁请求用于申请执行将所述第一数据更新为第二待写入数据的写操作的权限，所述第三写锁请求包括所述第三存储设备的标识；
    所述锁服务器还用于：接收所述第三写锁请求，确定所述第三写锁请求携带的所述第三存储设备的标识不是所述备用存储设备的标识；向所述第三存储设备发送所述第三写锁请求的响应结果，所述第三写锁请求的响应结果用于指示所述第三存储设备未获得执行将所述第一数据更新为所述第二待写入数据的写操作的权限。
  8. 根据权利要求6所述的系统，其特征在于，所述写数据请求用于请求将第一数据更新为所述待写入数据，所述写数据请求还包括所述第一数据的标识；所述系统还包括：
    第三存储设备，用于：在所述第一存储设备发生故障后，接收读取所述第一数据的读请求，所述读请求包括所述第一数据的标识；向所述锁服务器发送第四读锁请求，所述第四读锁请求用于申请执行读取所述第一数据的读操作的权限，所述第四读锁请求包括所述第三存储设备的标识；
    所述锁服务器还用于：接收所述第四读锁请求，确定所述第四读锁请求携带的所述第三存储设备的标识不是所述备用存储设备的标识；向所述第三存储 设备发送所述第四读锁请求的响应结果，所述第四读锁请求的响应结果用于指示所述第三存储设备未获得执行读取所述第一数据的读操作的权限。
  9. 根据权利要求6-8任一项所述的系统，其特征在于，所述第一写锁请求的预设字段中包含所述第二存储设备的标识；
    所述锁服务器还用于：在确认所述第二存储设备的标识是所述备用存储设备的标识之前，与所述第一存储设备约定，所述第一写锁请求中的所述预设字段中的标识为执行所述写数据请求的备用存储设备的标识。
  10. 根据权利要求6-8任一项所述的系统，其特征在于，所述第一存储设备用于：将所述待写入数据发送给所述第二存储设备，包括：
    通过与所述第二存储设备间的镜像通道将所述带写入数据发送给所述第二存储设备。
PCT/CN2016/111328 2015-12-31 2016-12-21 一种管理分布式存储系统的方法及分布式存储系统 WO2017114262A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201511030713.3 2015-12-31
CN201511030713.3A CN105653406B (zh) 2015-12-31 2015-12-31 一种管理分布式存储系统的方法及分布式存储系统

Publications (1)

Publication Number Publication Date
WO2017114262A1 true WO2017114262A1 (zh) 2017-07-06

Family

ID=56490265

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/111328 WO2017114262A1 (zh) 2015-12-31 2016-12-21 一种管理分布式存储系统的方法及分布式存储系统

Country Status (2)

Country Link
CN (1) CN105653406B (zh)
WO (1) WO2017114262A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111857545A (zh) * 2019-04-28 2020-10-30 深信服科技股份有限公司 一种读写锁冲突判断方法、系统、设备及计算机介质
CN112087417A (zh) * 2020-07-22 2020-12-15 深圳奇迹智慧网络有限公司 终端权限控制方法、装置、计算机设备和存储介质

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105653406B (zh) * 2015-12-31 2019-05-24 华为技术有限公司 一种管理分布式存储系统的方法及分布式存储系统
CN106844108B (zh) * 2016-12-29 2019-05-24 成都华为技术有限公司 一种数据存储方法、服务器以及存储系统
CN109213427B (zh) * 2017-06-30 2021-06-29 伊姆西Ip控股有限责任公司 用于管理存储系统的方法和设备
CN107463670B (zh) * 2017-08-03 2020-06-05 北京奇艺世纪科技有限公司 数据存储、读取方法及装置
CN111654519B (zh) * 2017-09-06 2024-04-30 华为技术有限公司 用于传输数据处理请求的方法和装置
CN110199270B (zh) * 2017-12-26 2022-09-02 华为技术有限公司 存储系统中存储设备的管理方法及装置
CN110879760B (zh) * 2018-09-05 2022-09-02 北京鲸鲨软件科技有限公司 一种统一存储系统及方法、电子设备
CN110321357B (zh) * 2019-06-14 2021-06-29 北京奇艺世纪科技有限公司 一种读写控制方法、装置及服务器
CN112333015B (zh) * 2020-10-28 2022-11-01 重庆紫光华山智安科技有限公司 媒体数据存储方法、装置、系统、电子设备及存储介质
KR102331734B1 (ko) 2021-03-19 2021-12-01 쿠팡 주식회사 분산 시스템의 데이터 처리 방법 및 이를 위한 전자 장치

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101252603A (zh) * 2008-04-11 2008-08-27 清华大学 基于存储区域网络san的集群分布式锁管理方法
CN104113587A (zh) * 2014-06-23 2014-10-22 华中科技大学 一种分布式文件系统客户端元数据缓存优化方法
CN104636088A (zh) * 2015-02-17 2015-05-20 华为技术有限公司 一种将数据写入数据服务器的方法及存储系统
CN105653406A (zh) * 2015-12-31 2016-06-08 华为技术有限公司 一种管理分布式存储系统的方法及分布式存储系统

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6574749B1 (en) * 1999-10-29 2003-06-03 Nortel Networks Limited Reliable distributed shared memory
CN102063338B (zh) * 2010-12-28 2013-03-20 华为技术有限公司 一种请求独占资源的方法及装置
US9141440B2 (en) * 2011-12-29 2015-09-22 Red Hat, Inc. Fault tolerant distributed lock manager
CN105426271B (zh) * 2015-12-22 2018-09-21 华为技术有限公司 对分布式存储系统的锁管理的方法和装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101252603A (zh) * 2008-04-11 2008-08-27 清华大学 基于存储区域网络san的集群分布式锁管理方法
CN104113587A (zh) * 2014-06-23 2014-10-22 华中科技大学 一种分布式文件系统客户端元数据缓存优化方法
CN104636088A (zh) * 2015-02-17 2015-05-20 华为技术有限公司 一种将数据写入数据服务器的方法及存储系统
CN105653406A (zh) * 2015-12-31 2016-06-08 华为技术有限公司 一种管理分布式存储系统的方法及分布式存储系统

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111857545A (zh) * 2019-04-28 2020-10-30 深信服科技股份有限公司 一种读写锁冲突判断方法、系统、设备及计算机介质
CN111857545B (zh) * 2019-04-28 2023-12-29 深信服科技股份有限公司 一种读写锁冲突判断方法、系统、设备及计算机介质
CN112087417A (zh) * 2020-07-22 2020-12-15 深圳奇迹智慧网络有限公司 终端权限控制方法、装置、计算机设备和存储介质
CN112087417B (zh) * 2020-07-22 2022-10-21 深圳奇迹智慧网络有限公司 终端权限控制方法、装置、计算机设备和存储介质

Also Published As

Publication number Publication date
CN105653406B (zh) 2019-05-24
CN105653406A (zh) 2016-06-08

Similar Documents

Publication Publication Date Title
WO2017114262A1 (zh) 一种管理分布式存储系统的方法及分布式存储系统
AU2019246872B2 (en) Tiered connection pooling methods, systems and computer readable storage media
KR102564157B1 (ko) 고속 비휘발성 메모리 컨트롤러의 네임스페이스 예약 처리 방법 및 이를 포함하는 데이터 관리 시스템
KR101941728B1 (ko) 클러스터화된 클라이언트 장애 해결 기법
US20180082053A1 (en) Application token through associated container
US8543773B2 (en) Distributed shared memory
JP4876170B2 (ja) グリッド・システムにおいてセキュリティ強制を追跡するシステムおよび方法
US7707586B2 (en) Operating system independent agent
WO2018068626A1 (zh) 一种磁盘锁的管理方法、装置和系统
JP2007122664A (ja) 情報処理方法および情報処理装置
JP7096736B2 (ja) システム、及びデータ処理方法
JP4407956B2 (ja) 情報処理方法および情報処理装置
WO2018094962A1 (zh) 一种迁移文件权限的方法、装置以及系统
WO2018028219A1 (zh) 设备仲裁的方法和设备
CN106462472B (zh) 使用多个租赁服务器获取资源租赁
US20180139193A1 (en) Systems and methods for managing credentials used to authenticate access in data processing systems
JP5478591B2 (ja) 情報システム及びその認証状態管理方法
US20210377308A1 (en) Reducing memory footprint after tls connection establishment
CN114285845B (zh) 云环境下的联网认证方法、系统、计算机设备及存储介质
CN117834125A (zh) 加密传输方法、装置及系统、电子设备、存储介质
KR20130017246A (ko) 인터넷 공유 자원에 대한 동기화를 위한 방법 및 장치

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: 16881051

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 16881051

Country of ref document: EP

Kind code of ref document: A1