WO2015085826A1

WO2015085826A1 - Method and apparatus for accessing shared resource

Info

Publication number: WO2015085826A1
Application number: PCT/CN2014/088928
Authority: WO
Inventors: 邹蛟同
Original assignee: 华为技术有限公司
Priority date: 2013-12-13
Filing date: 2014-10-20
Publication date: 2015-06-18
Also published as: CN103716383B; CN103716383A

Abstract

The present invention relates to the technical field of data communications, and disclosed are a method and an apparatus for accessing a shared resource. In a scheme described in embodiments of the present invention, after receiving a lock permission request sent by a client of a first node, the first node forwards the lock permission request only when a lock identifier comprising an identifier of a shared resource is not found in the first node, that is, the first node is searched first to determine whether the lock identifier comprising the identifier of the shared resource exists in the first node; the lock permission request is forwarded only when the lock identifier comprising the identifier of the shared resource does not exist in the first node, and when the lock identifier comprising the identifier of the shared resource exists in the first node, the lock permission request is directly allocated. It takes less time for the client to acquire the lock identifier from the first node than from other nodes. Therefore, the problem that an existing method for accessing a shared resource is time-consuming is solved.

Description

Method and device for accessing shared resources

The present application claims priority to Chinese Patent Application No. 201310689369.3, entitled "A Method and Apparatus for Accessing Shared Resources" on December 13, 2013, the entire contents of which are incorporated herein by reference. in.

Technical field

The present invention relates to the field of data communication technologies, and in particular, to a method and an apparatus for accessing shared resources.

Background technique

With the development of storage technology, various data information is exploding, and people are looking for a powerful data center architecture. The active-active data center is one of the most important solutions. Through the dual-active technology, service disaster recovery and load sharing can be implemented between multiple data centers (storage devices) across multiple regions, thereby consolidating the storage resources of storage devices. Figure 1A shows.

The storage array is the storage carrier of the storage device. In the dual-active data center solution, the storage array provides virtualized shared resources for the host application. When the system is running, data is simultaneously written to the storage arrays of the two storage devices. When one of the storage devices fails, the other storage device can still provide external services to implement data disaster recovery. In order to achieve virtualization integration and data disaster recovery of storage resources, a mutual exclusion mechanism needs to be provided between two storage devices to prevent concurrent access of shared resources (each storage device includes multiple nodes, and each node includes multiple clients) As shown in FIG. 1B, where the client refers to a process in which a node accesses a shared resource, because if multiple clients access and manipulate data at the same time, it may happen that each client overlaps the shared resource, resulting in The shared resources being accessed are in an inconsistent state.

In the cluster system, the lock manager is a commonly used mutual exclusion protection mechanism between nodes. It allows a shared resource to be accessed by only one client at a time. The process of accessing shared resources through the lock manager is as follows (as shown in the figure). 1C):

Assume that there are three nodes in storage device 1: node 1, node 2, node 3

Step a: assign a lock manager to the shared resource A;

Step b: When the node 2 needs to access the shared resource A, apply to the lock manager for the lock authority corresponding to the shared resource A;

Step c: After obtaining the lock permission corresponding to the shared resource A, the node 2 accesses the shared resource A;

Step d: Node 3 applies to the lock manager for the lock permission corresponding to the shared resource A. Since the lock has been authorized, the lock manager suspends the authorization of the lock request of other rights conflicts, and only the lock holder of the conflict releases the lock permission. After that, the new lock request will be authorized.

Through the above process, it can be seen that when multiple clients access a certain shared resource at the same time, the lock manager is used to achieve the orderly access of the same shared resource by multiple clients, and the consistency of the accessed data is ensured. However, the current lock manager has the following drawbacks: when some nodes are not the same node as the lock manager node, when acquiring the lock identifier corresponding to the lock authority of any shared resource, the other node is required to acquire the lock permission. A problem that takes a long time.

In summary, the current method of accessing shared resources has a long time-consuming problem.

Summary of the invention

The embodiment of the invention provides a method and a device for accessing a shared resource, which are used to solve the problem that the current method for accessing the shared resource has a long time.

In a first aspect, a method of accessing a shared resource is provided, including:

Receiving a lock permission request for carrying the identifier of the shared resource forwarded by the first node;

The lock permission request is forwarded when the first node sends the lock identifier that is sent by the client of the first node and does not find the identifier of the shared resource in the first node;

Determining, from the stored lock identifier, a lock identifier including an identifier of the shared resource;

Assigning the determined lock identifier to the client by using the first node, and giving the client a lock right corresponding to the lock identifier of the shared resource, where the client uses the share for the share Access to resources.

With reference to the first aspect, in a first possible implementation, the first node is configured to cache a lock identifier corresponding to a lock authority occupied by each client in the first node, or an occupied idle time The lock identifier corresponding to the lock authority that is smaller than the delete threshold.

With reference to the first aspect, or the first possible implementation manner of the first aspect, in a second possible implementation manner, the receiving, by the first node, the lock permission request that carries the identifier of the shared resource, specifically includes:

Receiving, by the first node, a lock permission request that carries the identifier of the shared resource that is forwarded by the second node;

The lock permission request is forwarded by the second node when receiving the lock identifier sent by the first node and not finding the identifier of the shared resource in the first node.

In conjunction with the second possible implementation of the first aspect, in a third possible implementation, the second node is configured to cache a lock identifier corresponding to the occupied lock authority in the specified lock authority, or The lock identifier corresponding to the lock authority that is occupied by the lock authority and whose idle time is less than the deletion threshold.

With reference to the first aspect, or the second to third possible implementation manners of the first aspect, in a fourth possible implementation, the determining, by the first node, the lock identifier is Client, specifically:

And transmitting the determined lock identifier to the first node by the second node, so that the first node assigns the lock identifier to the client.

With reference to the first aspect, or the first to fourth possible implementation manners of the first aspect, in a fifth possible implementation, the determining, by the first node, the lock identifier is Client, specifically:

If the lock permission corresponding to the lock identifier is occupied by another client, the lock permission recall instruction is sent to the client occupying the lock permission, so that the client occupying the lock authority completes accessing the shared resource and then releases The lock authority;

In a second aspect, an apparatus for accessing a shared resource is provided, including:

a receiving unit, configured to receive a lock permission request that is forwarded by the first node and that carries the identifier of the shared resource, where the lock permission request is sent by the first node at the client that receives the first node, and is in the first Forwarded when the lock identifier of the identifier of the shared resource is not found in the node;

a determining unit, configured to determine, from the stored lock identifier, a lock label including an identifier of the shared resource knowledge;

An allocating unit, configured to allocate the determined lock identifier to the client by using the first node, and give the client a lock right corresponding to the lock identifier of the shared resource, for the client Access to the shared resource.

With reference to the second aspect, in a first possible implementation, the lock permission request received by the receiving unit is sent by the first node, and the first node is used to cache the first node. The lock identifier corresponding to the lock authority occupied by each client in the account, or the lock identifier corresponding to the lock authority whose idle time is less than the deletion threshold.

With reference to the second aspect, or the first possible implementation manner of the second aspect, in a second possible implementation manner, the receiving unit is specifically configured to:

With reference to the second possible implementation of the second aspect, in a third possible implementation, the lock permission request received by the receiving unit is sent by the second node, and the second node is used by the second node. The lock identifier corresponding to the lock permission that is occupied in the specified lock permission, or the lock identifier corresponding to the lock authority that is occupied by the specified lock authority and whose idle time is less than the delete threshold.

With reference to the second aspect, or the second to third possible implementation manners of the second aspect, in a fourth possible implementation, the allocation unit is specifically configured to:

With reference to the second aspect, or the first to fourth possible implementation manners of the second aspect, in a fifth possible implementation manner, the allocation unit is specifically configured to:

If the lock permission corresponding to the lock identifier is occupied by another client, the lock permission recall instruction is sent to the client occupying the lock permission, so that the client occupying the lock authority completes accessing the total Release the lock permission after enjoying the resource;

And assigning, by the first node, the lock identifier corresponding to the lock authority released by the client occupying the lock authority to the client.

The beneficial effects of the present invention are as follows:

In the prior art, when some nodes are not the same node as the lock manager node, when acquiring the lock identifier corresponding to the lock authority of any shared resource, it is necessary to go to other nodes to acquire the lock permission, which takes a long time. The problem is that, in the embodiment of the present invention, the lock permission request that carries the identifier of the shared resource forwarded by the first node is received, where the lock permission request is sent by the first node to the client that receives the first node. And forwarding, when the lock identifier including the identifier of the shared resource is not found in the first node, and then determining a lock identifier including the identifier of the shared resource from the stored lock identifier, and finally, determining The lock identifier is allocated to the client by the first node, and the client is given a lock right corresponding to the lock identifier of the shared resource, and is used by the client to access the shared resource. In the solution, after receiving the lock permission request sent by the client of the first node, the first node does not find the target including the shared resource in the first node. The lock is only forwarded when the lock is identified. That is, the lock identifier of the identifier of the shared resource is first searched in the first node, and only when it does not exist, the forwarding is performed, and when it exists, the direct assignment is performed. However, it takes less time for the client to obtain the lock identifier from the first node than to obtain the lock identifier from other nodes. Therefore, the problem of the current method of accessing the shared resource is long.

DRAWINGS

The drawings are intended to provide a further understanding of the invention, and are intended to be a In the drawing:

1A is a schematic structural diagram of a dual active data center in the prior art;

1B is a schematic structural diagram of a node and a client in the prior art;

1C is a schematic diagram of accessing shared resources in a dual-active data center in the prior art;

2 is a schematic diagram of accessing a shared resource according to an embodiment of the present invention;

FIG. 3 is a first embodiment of accessing a shared resource according to an embodiment of the present invention;

4 is a second embodiment of accessing a shared resource according to an embodiment of the present invention;

FIG. 5 is a schematic structural diagram of a device for accessing a shared resource according to an embodiment of the present invention.

detailed description

The technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. It is a partial embodiment of the invention, and not all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

Additionally, the terms "system" and "network" are used interchangeably herein. The term "and/or" in this context is merely an association describing the associated object, indicating that there may be three relationships, for example, A and / or B, which may indicate that A exists separately, and both A and B exist, respectively. B these three situations. In addition, the letter "/" in this article generally indicates that the contextual object is an "or" relationship.

In order to solve the problem that the time-consuming problem exists in the prior art, in the embodiment of the present invention, the lock permission request that carries the identifier of the shared resource forwarded by the first node is received, where the lock permission request is received by the first node. And forwarding, when the lock identifier sent by the client of the first node and not identifying the identifier of the shared resource is found in the first node, and then determining the lock identifier including the identifier of the shared resource from the stored lock identifier, Finally, the determined lock identifier is assigned to the client through the first node, and the lock permission corresponding to the lock identifier of the shared resource is given to the client for accessing the shared resource by the client. In the solution, the first node is receiving After the lock permission request sent by the client of the first node is forwarded, the lock identifier including the identifier of the shared resource is not found in the first node, that is, the first node is first searched for whether to include the share. The lock identifier of the identifier of the resource, when it exists, directly allocates it, and only forwards when it does not exist, and the client obtains the lock identifier from the first node. To identify the other nodes to acquire the lock takes less time, therefore, to solve the long time-consuming problems of current methods to access shared resources.

The preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings, and it should be understood that The preferred embodiments described herein are only illustrative of the invention and are not intended to limit the invention, and the features of the embodiments and embodiments of the present application can be combined with each other without conflict.

Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

Referring to FIG. 2, in the embodiment of the present invention, the detailed process of accessing the shared resource is as follows:

Step 200: Receive a lock permission request that carries the identifier of the shared resource forwarded by the first node, where the lock permission request is sent by the first node after receiving the client of the first node, and is not found in the first node. Forwarded when the lock identifier of the shared resource identifier is used;

Step 210: Determine, from the stored lock identifier, a lock identifier that includes an identifier of the shared resource.

Step 220: Assign the determined lock identifier to the client through the first node, and give the client the lock permission corresponding to the lock identifier of the shared resource, and use the client to access the shared resource.

In the embodiment of the present invention, the lock identifier includes the following information, the identifier of the protected shared resource, and the manner of accessing the shared resource. In actual applications, other information than the above two types of information may also be included, and details are not further described herein. Said.

In the embodiment of the present invention, there are multiple ways to access a shared resource, for example, a method of reading a shared resource. At this time, the client acquires the permission to read the shared resource, and, for example, writes the shared resource. The permission obtained by the client to write the shared resource, wherein if the acquired permission to read the shared resource is obtained, the client can only perform the read operation on the shared resource, and cannot perform the write operation; if the obtained write share is obtained The permission of the resource, the client can not only read the shared resource, but also perform the write operation. Of course, in the actual application, there are other ways to access the shared resource, and the details are not detailed here. .

In the embodiment of the present invention, the first node has multiple forms, for example, may be used to cache a lock identifier corresponding to the lock authority occupied by each client in the first node, or the occupied idle time is less than the deletion threshold. The lock identifier corresponding to the lock authority. Of course, in practical applications, the first node may also be used to cache other information, which is not detailed here.

There are a plurality of execution entities of the steps 200-220. The following describes the lock manager as an example. It should be noted that the lock manager is only an embodiment of the execution subject. In practical applications, the execution entity may also be used. Therefore, other details are not repeated here.

In an actual application, if the client sends a lock permission request to the other node to access the shared resource, the interaction between the different nodes is a problem that takes a long time. In the embodiment of the present invention, in order to shorten the client sending access The lock permission request of the shared resource, or the time taken to receive the lock identifier, when the client sends a lock permission request for accessing the shared resource, the client may send a lock permission request for accessing the shared resource to the first node where the client is located, If the lock identifier of the identifier of the shared resource to be accessed is currently stored in the first node, the first node allocates the corresponding lock identifier to the client, if the first node does not currently store the shared resource to be accessed. The lock identifier of the identifier, the first node sends a lock permission request for accessing the shared resource to the lock manager that manages the lock authority of the shared resource of the different node, and then assigns the lock identifier returned by the received lock manager to The client, and stores the lock identifier, then the first node receives the lock permission for accessing the shared resource sent by the client again. While, if the lock identifier is not allocated, then the lock identification assigned to clients, thereby avoiding a longer acquire the lock identified problems exist in the process time-consuming.

For example, the node where the client 1 is located is in the first storage device, and the lock manager is located in the second storage device, wherein the lock manager manages the lock permission of the first storage device and the second storage device to access the shared resource, and the client 1 When the lock identifier of the first shared resource is obtained, that is, the lock permission, the lock permission request for accessing the first shared resource may be directly sent to the lock manager located at the second storage device, and at this time, because it is between the two storage devices Communication interaction, therefore, there is a problem that takes a long time. In order to shorten the time when the client 1 sends a lock permission request to access the first shared resource, or obtains the lock identifier of the first shared resource, the client 1 can send the client to the client. The first node that is located at 1 sends a lock permission request for accessing the first shared resource. If the lock identifier of the first shared resource is cached in the first node, and the lock identifier is not allocated, the first node sends the lock identifier. To the client 1, that is, to assign the lock permission to the first shared resource to the client 1; if the lock of the first shared resource is not cached in the first node The first node sends a lock permission request for accessing the first shared resource to the lock manager, sends the lock identifier returned by the received lock manager to the client, and caches the lock identifier, so as to receive the client again. 1 or when the client sends a lock permission request to access the first shared resource, if the lock identifier is not If it is assigned, the lock ID is sent to the client.

In the embodiment of the present invention, the request for the lock permission to carry the identifier of the shared resource forwarded by the first node may be received by the first node, for example, the lock permission request of the identifier of the shared resource that is forwarded by the first node by the second node may be received, where The lock permission request is forwarded by the second node when receiving the lock identifier sent by the first node and not identifying the identifier of the shared resource in the first node, where the foregoing is only receiving the shared resource carried by the first node. An implementation manner of the identified lock permission request, in actual application, there are other implementation manners, which are not described in detail herein.

The implementation of the lock permission request for receiving the identifier of the shared resource forwarded by the first node by the second node is as follows:

For example, the first storage device and the second storage device share 10 shared resources: shared resource 1, shared resource 2, ..., each shared resource corresponds to a lock identifier, and the second node located at the first storage device can store 10 The lock identifier corresponding to the shared resource, that is, the lock permission for accessing the 10 shared resources. If the client 1 located in the second storage device wants to access the shared resource 1, the client 1 can directly access the first storage device. The second node sends a lock permission request for accessing the shared resource 1, and acquires a lock identifier corresponding to the shared resource 1 from the second node; in order to improve efficiency, the client 1 may also first send the access shared resource 1 to the first node where it is located. The lock permission request, if the first node caches a lock identifier corresponding to the shared resource 1, the first node sends the lock identifier to the client 1, and assigns the lock permission to the client 1, if the first node is not cached The lock identifier corresponding to the shared resource 1 is forwarded by the first node to the second node to access the lock permission request of the shared resource 1, and the received second node sends the shared resource 1 corresponding to the client ID to the lock 1, the lock authority to assign a client.

In the embodiment of the present invention, the second node has multiple forms, optionally, used to cache the lock identifier corresponding to the occupied lock authority in the specified lock authority, or the idle time occupied by the specified lock authority is less than Delete the lock identifier corresponding to the lock permission of the threshold.

In the embodiment of the present invention, a storage device may further include a plurality of second nodes, where the second node is configured to cache a lock identifier corresponding to the occupied lock authority in the specified lock authority, or is occupied by the specified lock authority. The lock identifier corresponding to the lock authority whose idle time is less than the deletion threshold. At this time, the second node is different. You can manage lock identifiers for different shared resources.

For example, the first storage device and the second storage device share 10 shared resources: shared resource 1, shared resource 2, ..., each shared resource corresponds to a lock identifier, and the second node 1 located at the first storage device manages sharing. The lock identifiers corresponding to the resources 1-5 and the lock identifiers corresponding to the shared resources 6-10 of the second node 2 of the first storage device respectively, and the lock manager of the second storage device manages the locks corresponding to the 10 shared resources respectively. The identifier, that is, the lock permission for accessing 10 shared resources. If the client 1 located at the first storage device wants to access the shared resource 1, the client 1 can directly send the access shared resource to the lock manager located at the second storage device. The lock permission request acquires the lock identifier corresponding to the shared resource 1 from the lock manager; in order to improve the efficiency, the client 1 may first send a lock permission request for accessing the shared resource 1 to the first node where the first node is located, if the first node The cache has a lock identifier corresponding to the shared resource 1, and the first node sends the lock identifier to the client 1, and assigns the lock permission to the client 1, if the first node is not cached The lock identifier corresponding to the shared resource 1 is forwarded to the second node 1 by the first node to access the lock permission request of the shared resource 1. If the lock identifier of the identifier including the shared resource is not found in the second node 1, the switch forwards to the lock. The manager sends the lock identifier corresponding to the shared resource 1 sent by the received lock manager to the client 1 through the first node, and assigns the lock authority to the client 1.

Similarly, if the client 2 located at the first storage device wants to access the shared resource 10, the client 2 can directly send a lock permission request for accessing the shared resource 10 to the lock manager located at the second storage device, from the lock manager. Obtaining a lock identifier corresponding to the shared resource 10; in order to improve the access efficiency, the client 2 may first send a lock permission request for accessing the shared resource 10 to the third node where the third node is located, and if the third node caches the cached resource corresponding to the shared resource 10 The lock identifier is sent by the third node to the client 2, and the lock authority is assigned to the client 2. If the lock identifier corresponding to the shared resource 10 is not cached in the third node, the third node is The second node 2 forwards the lock permission request for accessing the shared resource 10. If the lock identifier corresponding to the shared resource 10 is not cached in the second node 2, the second node 2 forwards the lock permission request for accessing the shared resource 10 to the lock manager. And the lock identifier corresponding to the shared resource 10 sent by the received lock manager is sent to the client 2 through the third node, and the lock authority is allocated to the client 2.

The above embodiment only enumerates the case where only one second node exists in one storage device, in practice. In an application, there may also be a case where a storage device has multiple second nodes. If a storage device has multiple second nodes, when the client sends a lock permission request for accessing the shared resource, the client first sends a lock permission request for accessing the shared resource to the first node where the client is located, if the client is located at the first node. A lock identifier corresponding to the shared resource to be accessed is cached in a node, and the first node where the client is located sends a lock identifier corresponding to the shared resource to be accessed to the client, if the first node where the client is located is not If the lock identifier corresponding to the shared resource to be accessed is cached, the lock permission request for accessing the shared resource is forwarded to the corresponding second node, and if the lock identifier corresponding to the shared resource to be accessed is cached in the second node, The two nodes send the lock identifier corresponding to the shared resource to be accessed to the client. If the lock identifier corresponding to the shared resource to be accessed is not cached in the second node, the lock permission request for accessing the shared resource is forwarded to the lock management. Device.

In the embodiment of the present invention, there are multiple ways to receive the lock permission request of the identifier of the shared resource that is forwarded by the first node. For example, the first node may directly receive the lock permission request of the identifier of the shared resource that is forwarded by the first node. Alternatively, the second node may receive a lock permission request that receives the identifier of the shared resource forwarded by the first node.

Then, in the same way, the determined lock identifier is allocated to the client through the first node. For example, the determined lock identifier may be directly allocated to the client through the first node; or, The determined lock identifier is sent to the first node by the second node such that the first node assigns the lock identifier to the client.

In the embodiment of the present invention, the client sends a lock permission request for accessing the shared resource to the first node or the second node, whether the lock permission request for accessing the shared resource is directly sent to the lock manager, and the lock for receiving the request is received. The lock identifier corresponding to the accessed shared resource may be stored in the manager or the node, but the lock identifier is sent to the client, that is, the lock authority corresponding to the lock identifier is occupied. At this time, the lock corresponding to the determined lock identifier is used. When the privilege is occupied by other clients, the lock permission recall instruction is sent to the client occupying the lock permission, so that the client occupying the lock privilege releases the lock privilege after accessing the shared resource; and the lock privilege released by the client occupying the lock privilege corresponds to The lock identifier is assigned to the client through the first node.

That is to say, there are multiple ways to assign the determined lock identifier to the client through the first node. For example, if the lock permission corresponding to the determined lock identifier is occupied by other clients, the lock permission is sent to the client occupying the lock permission. The recall indication is such that the client occupying the lock authority releases the lock permission after accessing the shared resource, and the lock identifier corresponding to the lock authority released by the client occupying the lock authority is allocated to the client through the first node.

For example, the client 1 sends a lock permission request for accessing the shared resource 1 to the second node, and if the lock identifier corresponding to the shared resource 1 cached in the second node is assigned to another node, that is, the lock corresponding to the lock identifier When the privilege is assigned to the client 2, the lock privilege recall instruction is sent to the client 2 occupying the lock privilege, so that the client 2 occupying the lock privilege releases the lock privilege when accessing the shared resource, and then the second node occupies the lock privilege again. The lock identifier corresponding to the lock permission released by the client 2 is assigned to the client 1.

In the above embodiment, the lock identifier corresponding to the shared resource to be accessed in the second node is allocated to the other client as an example. In the actual application, the first node where the client is located may be used, or The lock identifier corresponding to the shared resource to be accessed cached in the lock manager is allocated to other clients. At this time, the first node or the lock manager sends a lock permission recall instruction to the client occupying the lock authority, and the process is as follows. The embodiments are the same and will not be described in detail herein.

In order to better understand the embodiments of the present invention, a specific application scenario is given below, and a process for accessing shared resources is further described in detail:

An embodiment of obtaining a lock identifier by a primary lock authority agent node.

Embodiment 1 (shown in Figure 3):

Step 300: The client 1 sends a lock permission request for accessing the shared resource 1 to the first node where the client 1 is located;

Step 310: The first node determines whether the lock identifier corresponding to the shared resource 1 is stored in the first node, and the lock identifier is not allocated, and if so, step 350 is performed; otherwise, step 320 is performed;

Step 320: The first node sends a lock permission request for accessing the shared resource 1 to the lock manager.

The lock manager is located at the second storage device.

Step 330: The lock manager determines whether a lock label corresponding to the shared resource 1 is stored in the first node. Knowledge, and the lock identifier is not assigned; if yes, step 340 is performed, otherwise, step 360 is performed;

Step 340: The lock manager sends the lock identifier corresponding to the shared resource 1 to the first node;

Step 350: The first node sends a lock identifier corresponding to the shared resource 1 to the client 1;

Step 360: The lock manager sends a lock permission recall indication to the client 2 occupying the lock authority, and sends the lock identifier corresponding to the lock authority released by the client 2 occupying the lock authority to the client 1 through the first node.

Embodiment 2 (as shown in Figure 4):

Step 400: The client 1 sends a lock permission request for accessing the shared resource 1 to the first node where the client 1 is located;

Step 410: The first node determines whether the lock identifier corresponding to the shared resource 1 is stored in the first node, and the lock identifier is not allocated, and if so, step 480 is performed; otherwise, step 420 is performed;

Step 420: The first node sends a lock permission request for accessing the shared resource 1 to the second node.

In this step, the client 1 and the second node are located at the first storage device.

Step 430: The second node determines whether the lock identifier corresponding to the shared resource 1 is stored in the first node, and the lock identifier is not allocated, and if so, step 470 is performed; otherwise, step 440 is performed;

Step 440: The second node sends a lock permission request for accessing the shared resource 1 to the lock manager.

The lock manager is located at the second storage device.

Step 450: The lock manager determines whether the lock identifier corresponding to the shared resource 1 is stored in the first node, and the lock identifier is not allocated; if yes, step 460 is performed; otherwise, step 490 is performed;

Step 460: The lock manager sends the lock identifier corresponding to the shared resource 1 to the second node;

Step 470: The second node sends the lock identifier corresponding to the shared resource 1 to the first node.

Step 480: The first node sends a lock identifier corresponding to the shared resource 1 to the client 1;

Step 490: The lock manager sends a lock permission recall indication to the client 2 that occupies the lock authority, and forwards the lock identifier corresponding to the lock authority released by the client 2 that occupies the lock authority to the first node through the second node, so that the first The node sends to client 1.

Referring to FIG. 5, an embodiment of the present invention further provides an apparatus for accessing a shared resource, where the apparatus includes a receiving unit 500, a determining unit 510, and an allocating unit 520, where:

The receiving unit 500 is configured to receive, by the first node, a lock permission request that carries the identifier of the shared resource, where the lock permission request is sent by the first node on the client that receives the first node, and is not in the first node. Forwarded when a lock identifier including the identity of the shared resource is found;

a determining unit 510, configured to determine, from the stored lock identifier, a lock identifier that includes an identifier of the shared resource;

The allocating unit 520 is configured to allocate the determined lock identifier to the client by using the first node, and give the client a lock right corresponding to the lock identifier of the shared resource, for the client to access the shared resource.

In the embodiment of the present invention, optionally, the lock permission request received by the receiving unit 500 is sent by the first node, and the first node is configured to cache a lock identifier corresponding to the lock authority occupied by each client in the first node. Or the lock identifier corresponding to the lock authority whose idle time is less than the deletion threshold.

In the embodiment of the present invention, optionally, the receiving unit 500 is specifically configured to:

Receiving a lock permission request that carries the identifier of the shared resource forwarded by the first node by using the second node;

The lock permission request is forwarded by the second node when receiving the lock identifier sent by the first node and not identifying the identifier including the shared resource in the first node.

In the embodiment of the present invention, optionally, the lock permission request received by the receiving unit 500 is sent by the second node, and the second node is configured to cache the lock identifier corresponding to the occupied lock authority in the specified lock authority, or specify The lock identifier corresponding to the lock authority that is occupied by the lock authority and whose idle time is less than the delete threshold.

In the embodiment of the present invention, optionally, the allocating unit 520 is specifically configured to:

The determined lock identifier is sent to the first node through the second node such that the first node assigns the lock identifier to the client.

If the lock permission corresponding to the determined lock identifier is occupied by another client, the lock permission recall instruction is sent to the client occupying the lock permission, so that the client occupying the lock permission releases the lock right after accessing the shared resource;

The lock identifier corresponding to the lock authority released by the client occupying the lock authority is allocated to the client through the first node.

In summary, in the embodiment of the present invention, the lock permission request that carries the identifier of the shared resource forwarded by the first node is received, where the lock permission request is sent by the first node on the client that receives the first node, and And forwarding, when the lock identifier of the identifier of the shared resource is not found in the first node, and then determining a lock identifier including the identifier of the shared resource from the stored lock identifier, and finally, determining the determined lock identifier by using the first node Assigned to the client, giving the client the lock permission corresponding to the lock identifier of the shared resource, for the client to access the shared resource, in the scheme, the first node receives the lock permission sent by the client of the first node After the request, the lock identifier of the identifier including the shared resource is not found in the first node, that is, the lock identifier of the identifier of the shared resource is first searched in the first node, and only exists in the absence of the lock identifier of the identifier of the shared resource. When the forwarding is performed, when it exists, the direct allocation is performed, and the client obtains the lock identifier from the first node than when the other node acquires the lock identifier. Less, therefore, to solve the problem of a longer time-consuming methods currently exist to access shared resources.

The present invention has been described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (system), and computer program products according to embodiments of the invention. 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. Means for implementing the functions in one or more of the flow or in a block or blocks of the flowchart.

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 functions in one or more blocks of a flow or a flow diagram and/or block diagram of a flowchart.

These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operational steps are performed on a computer or other programmable device to produce computer-implemented processing for execution on a computer or other programmable device. The instructions provide steps for implementing the functions in one or more blocks of the flowchart or in a flow or block of the flowchart.

While the preferred embodiment of the invention has been described, it will be understood that Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and the modifications and

It is apparent that those skilled in the art can make various modifications and variations to the embodiments of the invention without departing from the spirit and scope of the embodiments of the invention. Thus, it is intended that the present invention cover the modifications and modifications of the embodiments of the invention.

Claims

A method for accessing a shared resource, comprising:

Receiving a lock permission request for carrying the identifier of the shared resource forwarded by the first node;

The lock permission request is forwarded when the first node sends the lock identifier that is sent by the client of the first node and does not find the identifier of the shared resource in the first node;

Determining, from the stored lock identifier, a lock identifier including an identifier of the shared resource;

Assigning the determined lock identifier to the client by using the first node, and giving the client a lock right corresponding to the lock identifier of the shared resource, where the client uses the share for the share Access to resources.
The method according to claim 1, wherein the first node is configured to cache a lock identifier corresponding to a lock authority occupied by each client in the first node, or the occupied idle time is less than the deletion. The lock identifier corresponding to the lock permission of the threshold.
The method according to claim 1 or 2, wherein the receiving the lock permission request that carries the identifier of the shared resource forwarded by the first node includes:

Receiving, by the first node, a lock permission request that carries the identifier of the shared resource that is forwarded by the second node;

The lock permission request is forwarded by the second node when receiving the lock identifier sent by the first node and not finding the identifier of the shared resource in the first node.
The method according to claim 3, wherein the second node is configured to cache a lock identifier corresponding to the occupied lock authority in the specified lock authority, or the idle time occupied by the specified lock authority is less than Delete the lock identifier corresponding to the lock permission of the threshold.
The method of claim 3 or 4, wherein the assigning the determined lock identifier to the client by using the first node comprises:

And transmitting the determined lock identifier to the first node by the second node, so that the first node assigns the lock identifier to the client.
The method according to any one of claims 1-5, wherein the assigning the determined lock identifier to the client by using the first node comprises:

If the lock permission corresponding to the lock identifier is occupied by another client, the lock permission recall instruction is sent to the client occupying the lock permission, so that the client occupying the lock authority completes accessing the shared resource and then releases The lock authority;

And assigning, by the first node, the lock identifier corresponding to the lock authority released by the client occupying the lock authority to the client.
An apparatus for accessing a shared resource, comprising:

a receiving unit, configured to receive a lock permission request that is forwarded by the first node and that carries the identifier of the shared resource, where the lock permission request is sent by the first node at the client that receives the first node, and is in the first Forwarded when the lock identifier of the identifier of the shared resource is not found in the node;

a determining unit, configured to determine, from the stored lock identifier, a lock identifier including an identifier of the shared resource;

An allocating unit, configured to allocate the determined lock identifier to the client by using the first node, and give the client a lock right corresponding to the lock identifier of the shared resource, for the client Access to the shared resource.
The device according to claim 7, wherein the lock permission request received by the receiving unit is sent by the first node, and the first node is configured to cache each of the first nodes. The lock identifier corresponding to the lock authority occupied by the client, or the lock identifier corresponding to the lock authority whose idle time is less than the deletion threshold.
The device according to claim 7 or 8, wherein the receiving unit is specifically configured to:

Receiving, by the first node, a lock permission request that carries the identifier of the shared resource that is forwarded by the second node;

The lock permission request is sent by the second node after receiving the first node, And forwarding when the lock identifier including the identifier of the shared resource is not found in the first node.
The apparatus according to claim 9, wherein the lock authority request received by the receiving unit is sent by the second node, and the second node is configured to cache an occupied one of the specified lock rights. The lock identifier corresponding to the lock authority, or the lock identifier corresponding to the lock authority that is occupied by the specified lock authority and whose idle time is less than the deletion threshold.
The device according to claim 9 or 10, wherein the allocating unit is specifically configured to:

And transmitting the determined lock identifier to the first node by the second node, so that the first node assigns the lock identifier to the client.
The device according to any one of claims 7 to 11, wherein the allocating unit is specifically configured to:

If the lock permission corresponding to the lock identifier is occupied by another client, the lock permission recall instruction is sent to the client occupying the lock permission, so that the client occupying the lock authority completes accessing the shared resource and then releases The lock authority;

And assigning, by the first node, the lock identifier corresponding to the lock authority released by the client occupying the lock authority to the client.