CN104750562A - Lock processing system and method in distributed system - Google Patents
Lock processing system and method in distributed system Download PDFInfo
- Publication number
- CN104750562A CN104750562A CN201310742719.8A CN201310742719A CN104750562A CN 104750562 A CN104750562 A CN 104750562A CN 201310742719 A CN201310742719 A CN 201310742719A CN 104750562 A CN104750562 A CN 104750562A
- Authority
- CN
- China
- Prior art keywords
- lock
- node
- status
- locking
- inquiry request
- Prior art date
- Legal status (The legal status 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 status listed.)
- Granted
Links
Abstract
The invention provides a lock processing system in a distributed system. The distributed system comprises a plurality of nodes. The lock processing system comprises lock storage modules arranged on each node, lock processing modules arranged on each node and lock communication modules arranged on each node, wherein the lock storage modules are used for storing lock information related to the node where the lock storage modules are located, the lock information comprises a lock status of the node where the lock storage module is located, the lock status comprises an idle status, a pre-locking status and a locking status, the lock information also comprises priority data when the lock status is in the pre-locking status, the lock processing modules are used for processing data related to the lock of the node where the lock processing modules are located and processing exchange data related to the lock status between the node where the lock processing modules are located and other nodes, and the lock communication modules are used for communication between the node where the lock communication modules are located and other nodes. A corresponding lock processing method is further provided. According to the lock processing system and method in the distributed system, the abnormal problem of the lock processing caused by single node faults in a general distributed system is solved.
Description
Technical field
The present invention relates to the technology relevant with distributed system, more specifically, relate to the technology relevant with the distributed lock in distributed system.
Background technology
The lock implementation method of existing distributed system, mostly adopts Centroid system, namely adopts a main frame as Centroid, is responsible for the lock request of all nodes, lock is set up, lock is cancelled, locked overtime control etc. by this Centroid.The benefit of this pattern is that structure is simple and is easy to realize and control; but Centroid can only be a main frame under this structure can only be a node in other words; when this Centroid generation Single Point of Faliure, easily cause the lock process of whole system that abnormal or paralysis occurs.In addition, when lock-based transaction is occurred frequently, each node all goes to access this individual node, causes this individual node pressure increase, probably becomes the bottleneck of whole system.
In addition, also has the scheme adopting distributed lock system, it adopts the pattern of non-stop layer node, but in existing this distributed schemes, lock information needs between each node synchronous, this just likely produces the inconsistent problem of the lock information that causes because of reasons such as the delay of lock synchronizing information, node failures, also limit the scale of cluster simultaneously.
Summary of the invention
The invention provides the lock disposal system in a kind of distributed system, can solve the problem.Lock disposal system in this distributed system is arranged in the distributed system comprising multiple node, this lock disposal system comprises: the lock memory module being arranged on each node, for storing the lock information relevant with its place node, described lock information comprises the lock status of its place node, described lock status comprises free time, pre-locked and locking, and described lock information also comprises priority data when described lock status is pre-locked; Be arranged on the lock processing module of each node, for the treatment of its place node with lock relevant data and process interaction data relevant with lock status between its place node and other node; Be arranged on the lock communication module of each node, for the communication between its place node and other node.
According to the lock disposal system in distributed system of the present invention, preferably, described lock processing module is arranged to further: when the lock status of its place node is pre-locked, the lock inquiry request comprising priority data and lock mark is sent to other node, when the response of other node allows locking, the lock status of place node is set to locking, when any node response does not allow to lock in the other node, the lock status of place node is set to the free time, and after the lock inquiry request receiving from other node, inquire about the lock status of its place node, when described lock status is idle, generate and allow the signal of locking and this signal sent to other node described, when described lock status is pre-locked, the priority data of the priority data in more described lock inquiry request and this lock processing module place node, and the priority data in described lock inquiry request represent priority height time, generate and allow the signal of locking and this signal is sent to send other node described in lock inquiry request, otherwise, generate and do not allow the signal of locking and this signal is sent to send other node described in lock inquiry request.
According to the lock disposal system in distributed system of the present invention, preferably, described lock processing module is when the lock status of its place node is pre-locked, priority data is generated according to R × 1000+N mod 10000, wherein R is random number, N is node serial number, and the node serial number of each node is different.
According to a further aspect in the invention, also provide a kind of application lock disposal route in a distributed system, this distributed system comprises multiple node.This application lock disposal route in a distributed system comprises: the lock status arranging each node, and described lock status shows the held state of node to lock, when any node will lock lock, perform following steps: the lock status of inquiring about this node, when described lock status is idle, this lock is set to pre-locked, generates the lock inquiry request comprising priority data and this lock lock mark, send described lock inquiry request to other node, receive the response of other node to lock inquiry request, when described response is permission locking, put this lock for locking, and arbitrary response is when not allowing locking in described response, put this lock for idle, and after the lock inquiry request to receive from other node in any node, perform following steps: its lock status of querying node receiving this lock inquiry request, when its lock status state is for locking, generate and send and do not allow the data of locking to other node described, when the state of its lock status is idle, generate and send and allow the data of locking to other node described, when its lock status is pre-locked, priority data in more described lock inquiry request and the priority data of other node described, and the priority data in described lock inquiry request represent priority height time, generate and allow the signal of locking and this signal is sent to send other node described in lock inquiry request, otherwise, generate and do not allow the signal of locking and this signal is sent to send other node described in lock inquiry request.
According to the lock disposal route in distributed system of the present invention, preferably, when the lock status of node is pre-locked, priority data is generated according to R × 1000+N mod 10000, wherein R is random number, and N is node serial number, and the node serial number of each node is different.
According to lock processing scheme provided by the present invention, the information relevant with lock status is stored in each node of distributed system by it, and this information is without the need to synchronous between each node, but by lock initiator to the querying node lock status likely holding this lock, and solve the problem of lock conflict by lock priority data, thus avoid in prior art and cause because the single node of centralized management lock breaks down the situation that system lock process is paralysed.
Accompanying drawing explanation
Fig. 1 is the schematic configuration diagram according to the lock disposal system in the distributed system of example of the present invention.
The process flow diagram of Fig. 2 application according to the present invention lock disposal route relevant with sending the node of locking inquiry request in a distributed system.
Fig. 3 is the process flow diagram of the application according to the present invention lock disposal route relevant with receiving the node of locking inquiry request in a distributed system.
Embodiment
Describe schematic example of the present invention referring now to accompanying drawing, identical drawing reference numeral represents identical element.Each embodiment described below contributes to those skilled in the art and thoroughly understands the present invention, and is intended to example and unrestricted.Unless otherwise defined, the term (comprising science, technology and industry slang) used in literary composition has the implication identical with the implication that those skilled in the art in the invention generally understand.
Fig. 1 is the schematic configuration diagram according to the lock disposal system in the distributed system of example of the present invention.In this example, show four nodes of this distributed system, in embody rule, the data of node more, also can be two.In shown distributed system
lock disposal systemcomprise the lock memory module being arranged on each node, for storing the lock information relevant with its place node, this lock information comprises the lock status of its place node, and described lock status comprises free time, pre-locked and locking, and described lock information also comprises priority data when described lock status is pre-locked; Be arranged on the lock processing module of each node, for the treatment of its place node with lock relevant data and process interaction data relevant with lock status between its place node and other node; Be arranged on the communication module of each node, for the communication between communication module place node and other node.
As shown in Figure 1, node 1 comprises lock memory module 10 disposed thereon, lock processing module 12 and communication module 14; Node 2 comprises lock memory module 20 disposed thereon, lock processing module 22 and communication module 24; Node 3 comprises lock memory module 30 disposed thereon, lock processing module 32 and communication module 34; Node 4 comprises lock memory module 40 disposed thereon, lock processing module 42 and communication module 44.Lock memory module in each node, lock processing module and lock communication module are similar, hereafter mainly for node 1, lock memory module, lock processing module and lock communication module will be described.
The lock memory module 10 of node 1 stores the lock information relevant to node 1, and lock information comprises the current lock status of node 1 and also comprises priority data when lock status is pre-locked.The data relevant with lock of lock processing module 12 processing node 1, and processing node 1 and other node 2, interaction data relevant with lock status between 3 and 4; Exemplarily, lock processing module 12 receives from the inquiry request for a certain lock of other node from communication module 14 and then responds to this request, and when node 1 needs to lock a certain lock, inquire about this state being locked in node 1 and be further processed according to Query Result, being sent inquiring about the request of a certain lock status and the response according to other node 2,3 and 4 processes the state etc. that this is locked in node 1 to other node 2,3 and 4 by communication module 14, hereafter will set forth lock processing module 12 in more detail.In addition, lock processing module 12 to go back in processing node lock and cancel, lock the affairs relevant with lock such as overtime control.Lock communication module 1 is for the communication of the data relevant with lock any between node 1 with other node, concrete example is as comprised: receive the request signal relevant with lock status from other node and this request signal be transferred to lock processing module 12, receive the lock status request signal that other node sent by the node 1 that transmits of lock processing module 12 and this signal is sent to other node, send the response of other node being locked to request signal generated by the lock processing module 12 of node 1, and receive other node and be sent to the response signal of the lock request of node 1 and this signal sent to lock processing module 12 etc.
For convenience of description, a certain lock in this distributed system is referred to hereinafter with " A lock ".
The current lock status of each node comprises " free time ", " pre-locked " and " locking " three kinds, and wherein, " free time " represents that A lock is not held by present node, is in idle condition; " pre-locked " represents the locked A lock of this node, is attempting the lock-out state of inquiring that other node is locked A, but is not yet receiving inquiry result; " locking " represents that A is locked in this node and is in the lock state and other node also agrees unanimously that A lock is locked.Specifically, for " free time ", the lock status of node 1 represents that A lock is not locked by node 1; For " pre-locked ", the lock status of node 1 represents that A lock has been locked by node 1, and node 1 is inquiring the lock-out state that other node is locked A, but not yet receives the response of other node to this inquiry; When the lock status of node 1 is " locking ", represent that A lock has been locked by node 1 and other node all agrees to that A lock is locked by node 1.
When node 1 needs locking A to lock, first lock processing module 12 inquires about to lock memory module 10 current state that A is locked in node 1, if A lock is in locking, then no longer process, if A lock is current be in idle condition, the status modifier then locked by A is " pre-locked ", generate the lock priority data for A lock and A lock status inquiry request simultaneously, and this inquiry request is sent to lock communication module 14, such as node 2 is sent to the form of broadcast by lock communication module 14, 3 and 4, wherein, this inquiry request comprises the lock priority data and A lock mark locked for A, this lock mark be intended to other node specification inquire about for A lock, this lock priority data shows the priority of node 1 pair of A lock.The communication module 24,34 and 44 of node 2,3 and 4 receives this inquiry request respectively, this request is sent to the lock processing module 22 of node 2 by communication module 24, this request is sent to the lock processing module 32 of node 3 by communication module 34, and this request is sent to the lock processing module 42 of node 4 by communication module 44.After the lock processing module 22 of node 2 receives this request, namely the lock memory module 20 of query node 2 is to know that A is locked in the state of node 2; After the lock processing module 32 of node 3 receives this request, namely the lock memory module 30 of query node 3 is to know that A is locked in the state of node 3; After lock processing module 42 receives this request, namely the lock memory module 40 of query node 4 is to know that A is locked in the state of node 4.Lock processing module 22 obtains A and locks current state for " free time ", therefore generates the signal allowing locking, and by communication module 24, this signal is sent to node 1.Lock processing module 32 obtains A and is locked in the current state of node 3 for " pre-locked ", then query node 3 locks the priority data of A lock in memory module 30 further, thus the priority data of node 3 pairs of A locks is compared with the priority data that received inquiry request interior joint 1 couple of A locks, if the priority of node 3 pairs of A locks is high, then lock processing module 32 generates and does not allow the signal of locking and by communication module 34, this signal sent to node 1, otherwise then generate allow locking signal and by A, the state be locked in node 3 makes into " free time ", and send to node 1 by communication module 24 by allowing the signal of locking.Lock processing module 42 obtains A and is locked in the current state of node 4 for " locking ", then generate and do not allow locking signal, and this signal is sent to the communication module 14 of node 1 via communication module 44.If in this example, node 1 receive from node 2,3,4 to lock inquiry response be respectively allow locking, allow locking, do not allow locking, then node 1 finally can not lock A lock, and can by A lock state be arranged to the free time.
Priority data can be each node being manually arranged at distributed system, also can stochastic generation.In example discussed herein, by (1) generation according to the following formula when lock status is pre-locked of lock processing module:
Priority data=R × 1000+N mod 10000 (1)
Wherein, R is random number, and N is node serial number, and the numbering of each node is not identical.Be noted that the algorithm of the calculating priority level data that the present invention provides is example and unrestricted, also can generate this priority data according to other algorithm.
In addition, in lock disposal system as above, each lock memory module can adopt the existing memory module of each node, also can arrange in addition again, each lock processing module can realize separately also realizing in the existing processing element of each node (as processor), and each communication module then can adopt the existing communication module of each node.
Lock disposal system in distributed system as above, the information relevant with lock status is stored in each node of distributed system by it, and this information is without the need to synchronous between each node, but by lock initiator (node 1 in such as example) to the querying node lock status likely holding this lock, and solve the problem of lock conflict by lock priority data, thus avoid in prior art and cause because the single node of centralized management lock breaks down the situation that system lock process is paralysed.
Fig. 2 is the process flow diagram of the application according to the present invention lock disposal route relevant with sending the node of locking inquiry request in a distributed system, and wherein, this distributed system comprises multiple node, such as node 1,2,3 and 4.
In step 200, arrange the lock status of each node, this lock status shows the held state of node to lock.In example of the present invention, lock status comprises free time, locking and pre-locked.Any node in this distributed system, such as, when node 1 needs locking A to lock (as described above, for simplicity, referring to arbitrary lock with A lock), performs step 210.In step 210, inquire about this node, such as the lock status of node 1.In step 212, according to Query Result, state A being locked in node 1 is set to pre-locked or no longer carries out subsequent treatment; Specifically, when the lock status of node 1 is idle, lock status A being locked in node 1 is set to pre-locked, when lock status is for locking, then no longer carries out subsequent treatment.A is locked in node 1 when being set to pre-locked, namely performs step 214, generate the lock inquiry request comprising priority data and A lock lock mark; Priority data can be manual configuration for each node in this distributed system, also when being set to pre-locked by this lock status, can generate according to formula (1).In step 216, the lock inquiry request of generation is sent to other node 2,3 and 4.In step 218, the node 1 sending lock inquiry request receives the response of the lock inquiry request sent for node 1 from node 2,3 and 4.In step 220, the node 1 sending lock inquiry request, according to received response, puts A lock for locking or idle, specifically, when described response is permission locking, put this lock for locking, and arbitrary response is when not allowing locking in described response, put this lock for idle.
Fig. 3 is the process flow diagram of the application according to the present invention lock disposal route relevant with receiving the node of locking inquiry request in a distributed system.In conjunction with the example above for Fig. 2, the method shown in key diagram 3, that is, in conjunction with receiving the node 2,3 and 4 of lock inquiry request so that the method shown in Fig. 3 to be described.
Any one receiving in the node 2,3 and 4 of lock inquiry request performs the method.In step 300, the node 2,3 and 4 receiving lock inquiry request is inquired about A respectively and is locked in its state located, and specifically, node 2 inquires about that A is locked in the state of node 2, node 3 is inquired about A and is locked in the state of node 3 and node 4 inquires about the state that A is locked in node 4.
In step 302, according to Query Result, generate corresponding signal; Particularly, when lock status is for locking, generate the signal not allowing to lock, when lock status is idle, generate the signal allowing locking, when lock status is pre-locked, relatively receive from the priority data in the lock inquiry request of node 1 and the priority data in this node, when the priority height that the priority data of the request of locking represents, generate the signal allowing to lock, otherwise, generate the signal not allowing to lock.Exemplarily, the Query Result of node 2 is that A lock is idle, and the Query Result of node 3 is that A lock is pre-locked, and the Query Result of node 4 is the locking of A lock.Node 2, when A lock is idle, generates the signal allowing locking; The priority data that node 3 is locked for the A priority data locked and the lock inquiry request interior joint 1 couple of A received compares by node 3 further, if the priority of node 3 pairs of A locks is high, then generate do not allow lock signal, otherwise then generate allow locking signal and by A, the state be locked in node 3 makes into " free time ".When the current state of node 4 is " locking ", generate the signal not allowing to lock.
In step 304, the signal generated in step 302 is sent to the node 1 sending lock inquiry request, namely, the signal that the permission of generation locks is sent to node 1 by node 2, node 3 is by the permission of generation locking or do not allow locking signal to send to node 1, and node 4 sends to node 1 by not allowing the signal locked.
Method as above can be embodied as software, hardware or this both combination.In practical application, method as above can perform in the lock disposal system described in composition graphs 1.
Method as above, it makes each node in distributed system all participate in lock management, when it is intended to lock a certain lock, the node can may holding this lock to other sends inquiry request, to understand the held state of other node to this lock, and solve lock conflict problem by priority data.Compared with prior art, perform the distributed system of lock disposal route of the present invention, avoid and cause Single Point of Faliure to cause the problem of whole system lock process paralysis because lock relies on Centroid, also solve the problem of lock process bottleneck in distributed system simultaneously, improve the high availability of distributed system.
Claims (5)
1. the lock disposal system in distributed system, this distributed system comprises multiple node, it is characterized in that, the lock disposal system in described distributed system comprises:
Be arranged on the lock memory module of each node, for storing the lock information relevant with its place node, described lock information comprises the lock status of its place node, and described lock status comprises free time, pre-locked and locking, and described lock information also comprises priority data when described lock status is pre-locked;
Be arranged on the lock processing module of each node, for the treatment of its place node with lock relevant data and process interaction data relevant with lock status between its place node and other node;
Be arranged on the lock communication module of each node, for the communication between its place node and other node.
2. the lock disposal system in distributed system as claimed in claim 1, it is characterized in that, described lock processing module is arranged to further:
When the lock status of its place node is pre-locked, send the lock inquiry request comprising priority data and lock mark to other node,
When the response of other node allows locking, the lock status of place node is set to locking,
When any node response does not allow to lock in the other node, the lock status of place node is set to the free time; And
After receiving the lock inquiry request from other node, inquire about the lock status of its place node,
When described lock status is idle, generates and allow the signal of locking and this signal sent to other node described,
When described lock status is pre-locked, the priority data of the priority data in more described lock inquiry request and this lock processing module place node, and the priority data in described lock inquiry request represent priority height time, generate and allow the signal of locking and this signal is sent to send other node described in lock inquiry request, otherwise, generate and do not allow the signal of locking and this signal is sent to send other node described in lock inquiry request.
3. the lock disposal system in distributed system as claimed in claim 2, it is characterized in that, described lock processing module is when the lock status of its place node is pre-locked, priority data is generated according to R × 1000+N mod10000, wherein R is random number, N is node serial number, and the node serial number of each node is different.
4. an application lock disposal route in a distributed system, this distributed system comprises multiple node, it is characterized in that, described application lock disposal route in a distributed system comprises:
Arrange the lock status of each node, described lock status shows the held state of node to lock;
When any node will lock lock, perform following steps:
Inquire about the lock status of this node, when described lock status is idle, this lock be set to pre-locked,
Generate the lock inquiry request comprising priority data and this lock lock mark,
Send described lock inquiry request to other node,
Receive the response of other node to lock inquiry request,
When described response is permission locking, put this lock for locking, in described response, arbitrary response is when not allowing locking, and it is idle for putting this lock; And
After any node receives the lock inquiry request from other node, perform following steps:
Receive its lock status of querying node of lock inquiry request,
When its lock status is for locking, generates and sends and do not allow the data of locking to other node described,
When its lock status state is idle, generates and sends and allow the data of locking to other node described,
When its lock status is pre-locked, priority data in more described lock inquiry request and the priority data of other node described, and the priority data in described lock inquiry request represent priority height time, generate and allow the signal of locking and this signal is sent to send other node described in lock inquiry request, otherwise, generate and do not allow the signal of locking and this signal is sent to send other node described in lock inquiry request.
5. application as claimed in claim 4 lock disposal route in a distributed system, it is characterized in that, described priority is generated by R × 1000+N mod10000 when described lock status is pre-locked, and wherein R is random number, N is node serial number, and the node serial number of each node is different.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310742719.8A CN104750562B (en) | 2013-12-30 | 2013-12-30 | Lock processing system in distributed system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310742719.8A CN104750562B (en) | 2013-12-30 | 2013-12-30 | Lock processing system in distributed system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104750562A true CN104750562A (en) | 2015-07-01 |
| CN104750562B CN104750562B (en) | 2018-09-21 |
Family
ID=53590303
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310742719.8A Active CN104750562B (en) | 2013-12-30 | 2013-12-30 | Lock processing system in distributed system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104750562B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105069008A (en) * | 2015-07-03 | 2015-11-18 | 曙光信息产业股份有限公司 | Distributed system data processing method and apparatus |
| CN106293954A (en) * | 2016-08-08 | 2017-01-04 | 浪潮(北京)电子信息产业有限公司 | A kind of High Availabitity service management based on distributed lock |
| CN106484542A (en) * | 2016-09-06 | 2017-03-08 | 华为技术有限公司 | The method and apparatus of overlapping nodes event in a kind of process distributed system |
| CN106775974A (en) * | 2016-12-07 | 2017-05-31 | 国云科技股份有限公司 | A kind of implementation method of distributed priority queuing lock |
| CN107517277A (en) * | 2017-10-09 | 2017-12-26 | 郑州云海信息技术有限公司 | A kind of sanlock lock implementation methods and device |
| CN109697127A (en) * | 2017-10-24 | 2019-04-30 | 华为技术有限公司 | The method and apparatus that the access operation of a kind of pair of shared resource locks |
| CN112035269A (en) * | 2020-09-10 | 2020-12-04 | 蔡国凤 | Multi-node locking method for distributed system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101256510A (en) * | 2008-04-11 | 2008-09-03 | 中兴通讯股份有限公司 | Cluster system and method for implementing centralized management thereof |
| CN102412988A (en) * | 2011-11-14 | 2012-04-11 | 浪潮(北京)电子信息产业有限公司 | Service information system and method for realizing continuous operation by using same |
-
2013
- 2013-12-30 CN CN201310742719.8A patent/CN104750562B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101256510A (en) * | 2008-04-11 | 2008-09-03 | 中兴通讯股份有限公司 | Cluster system and method for implementing centralized management thereof |
| CN102412988A (en) * | 2011-11-14 | 2012-04-11 | 浪潮(北京)电子信息产业有限公司 | Service information system and method for realizing continuous operation by using same |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105069008A (en) * | 2015-07-03 | 2015-11-18 | 曙光信息产业股份有限公司 | Distributed system data processing method and apparatus |
| CN106293954A (en) * | 2016-08-08 | 2017-01-04 | 浪潮(北京)电子信息产业有限公司 | A kind of High Availabitity service management based on distributed lock |
| CN106484542A (en) * | 2016-09-06 | 2017-03-08 | 华为技术有限公司 | The method and apparatus of overlapping nodes event in a kind of process distributed system |
| CN106484542B (en) * | 2016-09-06 | 2020-05-19 | 华为技术有限公司 | Method and device for processing overlapping node event in distributed system |
| CN106775974A (en) * | 2016-12-07 | 2017-05-31 | 国云科技股份有限公司 | A kind of implementation method of distributed priority queuing lock |
| CN106775974B (en) * | 2016-12-07 | 2019-12-10 | 国云科技股份有限公司 | Method for realizing distributed priority queuing lock |
| CN107517277A (en) * | 2017-10-09 | 2017-12-26 | 郑州云海信息技术有限公司 | A kind of sanlock lock implementation methods and device |
| CN107517277B (en) * | 2017-10-09 | 2020-05-19 | 苏州浪潮智能科技有限公司 | Method and device for realizing sanlock |
| CN109697127A (en) * | 2017-10-24 | 2019-04-30 | 华为技术有限公司 | The method and apparatus that the access operation of a kind of pair of shared resource locks |
| CN109697127B (en) * | 2017-10-24 | 2020-12-01 | 华为技术有限公司 | Method and device for locking access operation of shared resource |
| CN112035269A (en) * | 2020-09-10 | 2020-12-04 | 蔡国凤 | Multi-node locking method for distributed system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104750562B (en) | 2018-09-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104750562A (en) | Lock processing system and method in distributed system | |
| US10764369B2 (en) | Data storage method and server applicable to distributed server cluster | |
| WO2015096606A1 (en) | Network device, cluster storage system and distributed lock management method | |
| CN106982136B (en) | Multi-domain layered multi-domain Internet of things platform and multi-domain management method | |
| US20080209007A1 (en) | Methods, systems, and computer program products for accessing data associated with a plurality of similarly structured distributed databases | |
| CN107153660B (en) | Fault detection processing method and system for distributed database system | |
| CN106294713A (en) | The method of data synchronization resolved based on Incremental Log and data synchronization unit | |
| CN103488526A (en) | System and method for locking business resource in distributed system | |
| CN106612339A (en) | Domain name updating method, system and main DNS (Domain Name System) server | |
| CN106814975A (en) | A kind of memory management method and its system for polymorphic type storage device | |
| CN104135378A (en) | Method of management control of Internet of Things gateways and management control entity for Internet of Things gateways | |
| CN106648858A (en) | Interactive management system for C# interface data | |
| WO2015194937A1 (en) | System and method for distributed secure data storage in torus network topology | |
| EP3128715B1 (en) | Resource creation method and apparatus | |
| CN111104250B (en) | Method, apparatus and computer readable medium for data processing | |
| US10680974B2 (en) | Method and device for monitoring data processing status | |
| CN107239235A (en) | A kind of many RAID synchronous method living of many controls and system | |
| US9385884B2 (en) | Node joining protocol for cluster federation | |
| US20100303055A1 (en) | Method for Assigning Data Messages to Time Slots in a Wireless Data Bus System Having Hidden Subscribers | |
| CN107423454B (en) | Method, device and equipment for processing file lock in distributed file system | |
| EP3346671B1 (en) | Service processing method and equipment | |
| CN107231394B (en) | Construction method of data source address distribution tree and method for copying data | |
| CN103679615A (en) | High-availability logistics information management system based on distributed system architecture | |
| CN113518126A (en) | Cross fault-tolerant method for alliance chain | |
| CN107153594B (en) | HA component main selection method and system of distributed database system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |