CN101059807A - Method and system for promoting metadata service reliability - Google Patents
Method and system for promoting metadata service reliability Download PDFInfo
- Publication number
- CN101059807A CN101059807A CN 200710051407 CN200710051407A CN101059807A CN 101059807 A CN101059807 A CN 101059807A CN 200710051407 CN200710051407 CN 200710051407 CN 200710051407 A CN200710051407 A CN 200710051407A CN 101059807 A CN101059807 A CN 101059807A
- Authority
- CN
- China
- Prior art keywords
- data server
- metadata
- meta data
- module
- service
- 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
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Hardware Redundancy (AREA)
Abstract
A method for improving metadata service reliability and a relative system belongs to computer memory technical field, for fully utilizing node resource to improve the reliability of metadata service. The inventive method comprises normal operation, main metadata server accident, backup metadata server lift, sub metadata server accident and second backup metadata server lift. The inventive system comprises a client node, a memory node, a main metadata server and a sub metadata server, which respectively comprises an object metadata service module, a memory resource manage module, and a high available module, wherein the high available module comprises a metadata copy module, an accident shield module, a detect module and a select module, the select module can select the memory node with minimum load as the backup metadata server, and the accident shield server is switched to next metadata server, and selects the backup metadata server to form a master-salve backup chain service mode, to improve the reliability of metadata service.
Description
Technical field
The invention belongs to computer memory technical field, be specifically related to a kind of method and system thereof that improves metadata service reliability.
Background technology
Metadata often is referred to as " data of data ", network store system is for reducing the server expense, improve transfer rate, how with metadata and data separating, adopt the dedicated meta data server to be in charge of metadata, data transmission does not just need only need obtain enough metadata informations in advance through transit server like this.Meta data server becomes the hinge of contact customer node and memory node thus, also forms single failpoint easily, thereby need seek to improve the method for metadata service reliability.
More traditional network store system adopts the principal and subordinate to be equipped with mode mutually reliable metadata service is provided, as the pangli duckweed, and what leap, Yue Jianhui etc., " the centralized metadata management high-availability system design of parallel file system ", computer engineering and science, 2004 the 26th volume o. 11ths.Pivot data server and detect interconnection by " heartbeat " from meta data server.Hinder for some reason and after the break in service, bear pivot data server function when monitoring the pivot data server from meta data server.After introducing mutual mode fully, pivot data server and the probability that all breaks down from meta data server are little compared with single probability of malfunction, and the metadata reliability of service is also just corresponding to be improved.
Yet the principal and subordinate is equipped with mode mutually and limits because of own characteristic, fails to give full play to the effect of other node resources in the storage system, causes doing further breakthrough aspect reliability; After all breaking down when the pivot data server with from meta data server, although other parts of storage system still work well, because the central hub effect of server, system but can't provide stores service.For this reason, a kind of method and system thereof that improves metadata service reliability that the present invention proposes makes full use of other node resources in the storage system, improves the metadata reliability of service to a great extent.
Summary of the invention
The invention provides a kind of method and system thereof that improves metadata service reliability, purpose is the node resource that makes full use of system, improves the metadata reliability of service, reaches " zero failure " rate of system in metadata service angle.
A kind of method that improves metadata service reliability of the present invention comprises the steps:
(1) normal operating procedure, under the normal condition, pivot data server operation object metadata service processes and Storage Resource Management (SRM) process provide the metadata service of storage system; Connect from meta data server by sense channel, start monitoring process and monitor the other side's state, start simultaneously and choose process, from memory node, choose average load relatively the lightest as standby meta data server; Start the consistance that the metadata replication process guarantees all metadata on it and the pivot data server from meta data server;
(2) pivot data server fault step, when monitoring after the pivot data server breaks down, take over the metadata services request from meta data server, operation object metadata service processes and Storage Resource Management (SRM) process, enable the fault masking process, the influence that shielding pivot data server fault causes system starts and switches subprocess, the metadata request of client node is switched to from meta data server, and startup reparation subprocess is repaired the pivot data server; Start simultaneously and choose process, on memory node, select average load relatively the lightest as the second standby meta data server;
(3) standby meta data server raise-position step, standby meta data server is born the function from meta data server, start monitoring process, and monitor the other side's state mutually, start the metadata replication process and guarantee it and the consistance of all metadata from meta data server from meta data server;
(4) from meta data server fault step, when monitoring when meta data server breaks down, the standby meta data server of raise-position is taken over the metadata services request, operation object metadata service processes and Storage Resource Management (SRM) process, start the fault masking process, shielding is from the influence of meta data server fault to system, start and repair subprocess repairing from meta data server, and startup adds subprocess, judge whether the pivot data server is repaired, if the then adding system of repairing provides the metadata service of storage system, start the consistance that the metadata replication process guarantees all metadata on pivot data server and the standby meta data server, move object metadata service processes and Storage Resource Management (SRM) process then, start and switch subprocess, the request of client node switches to the pivot data server, and standby meta data server continues to bear from the meta data server function, stop object metadata service processes and Storage Resource Management (SRM) process, start the consistance that the metadata replication process guarantees all metadata on it and the pivot data server, the second standby meta data server continues to do standby meta data server; If the pivot data server is also at repairing phase, then standby meta data server is provided the metadata service of storage system, start and switch subprocess, client-requested switches to standby meta data server, start and to choose process, on memory node, select average load relatively the lightest as the 3rd standby meta data server.
(5) second standby meta data server raise-position steps, if the pivot data server is not finished reparation, standby meta data server is converted to the pivot data server, the second standby meta data server is born from the meta data server function, start monitoring process, monitor the other side's state mutually with standby meta data server, start the consistance that the metadata replication process guarantees all metadata on it and the standby meta data server.
The method of described raising metadata service reliability is characterized in that: (1) described object metadata service processes provides the inquiry location of metadata and upgrades operation; (2) all memory nodes in the described Storage Resource Management (SRM) management of process storage system; (3) described metadata replication process guarantees the consistance of all metadata on meta data server and the pivot data server; (4) whether normally described monitoring process periodically detects the operation of the other side's object metadata service processes and Storage Resource Management (SRM) process; (5) described fault masking process comprises that 3 subprocess: A switch subprocess, automatically the client metadata request is switched to another meta data server after breaking down, and continual metadata service is provided; B repairs subprocess, and the meta data server that breaks down is carried out necessary reparation; C adds subprocess, joins in the total system after the meta data server reparation is finished.
The method of described raising metadata service reliability, it is characterized in that: the described process of choosing, when choosing the lightest memory node of average load relatively, the load value of memory node is the weighted sum of system queue length, cpu busy percentage, memory usage, disk free space and network throughput; The memory node choosing is done after the standby meta data server, upgrades corresponding load value.
The system of raising metadata service reliability of the present invention comprises client node, memory node and meta data server, and the three is by network interconnection, and client node sends file or Object Operations request, comprises metadata request and request of data; Memory node provides stores service; Meta data server comprises the pivot data server and from meta data server, they comprise object metadata service module, Storage Resource Management (SRM) module and high available modules separately, three modules externally provide the metadata service by unified interface, and (1) object metadata service module provides the inquiry location of metadata and upgrades operation; (2) all memory nodes in the Storage Resource Management (SRM) module management storage system; (3) the high availability module guarantees the metadata reliability of service, comprises metadata replication module, fault masking module and monitoring module, and (3.1) metadata replication module guarantees the consistance of all metadata on meta data server and the pivot data server; (3.2) failure message of fault masking module shielding meta data server, transparent lasting metadata request service function is provided, comprise 3 submodules: the A. switching submodule, automatically the client metadata request is switched to another meta data server after breaking down, continual metadata service is provided; B. repair submodule, the meta data server that breaks down is carried out necessary reparation; C. add submodule, after the meta data server reparation is finished, join in the total system; (3.3) whether normally monitoring module periodically detects the operation of the other side's object metadata service module and Storage Resource Management (SRM) module; It is characterized in that:
(1) described high availability module comprises and chooses module, according to history access record and self load processing power, chooses the lighter memory node of load as standby meta data server from memory node;
When (2) described memory node is chosen for standby meta data server, load the pivot data server or from all functions module of meta data server.
According to method of the present invention, the server that shielding is broken down, the request of transparent switching client node is to next pivot data server, and choose new standby meta data server, constitute the principal and subordinate and be equipped with chain type metadata service mode, greatly improve metadata service reliability, system reaches " zero failure " rate in metadata service angle; System need not increase the continuity that extras guarantee the metadata service; Fully excavate the abundant resources of memory node, master and slave, be equipped with the three except reliable metadata service can be provided, under normal circumstances can also equally loaded by rational management.
Description of drawings
Fig. 1 is equipped with chain type metadata service mode synoptic diagram for the principal and subordinate that the present invention proposes;
Fig. 2 is the schematic flow sheet of the inventive method;
Fig. 3 is a system architecture synoptic diagram of the present invention;
Fig. 4 is equipped with the state transition diagram of chain type metadata service for the present invention adopts the principal and subordinate.
Embodiment
As shown in Figure 1, a kind of method that improves metadata service reliability proposed by the invention is that the principal and subordinate is equipped with chain type metadata service mode, pivot data server 11.1 and provide the private server of metadata service from meta data server 11.2, and follow-up that metadata service is provided is the lighter memory node (13.i of load selected in the storage system, 13.j, 13.k ...).By the chain type rule of this similar " translation ", reach the purpose that improves metadata service reliability.
The execution flow process of the inventive method as shown in Figure 2, under the normal condition, pivot data server MDS1 operation object metadata service processes and Storage Resource Management (SRM) process provide the metadata service of storage system; Connect from meta data server MDS2 by sense channel, start monitoring process and monitor the other side's state, start simultaneously and choose process, select standby meta data server MDS3; MDS2 starts the consistance that the metadata replication process guarantees all metadata on it and the pivot data server.
When monitoring after MDS1 breaks down, MDS2 takes over the metadata services request, operation object metadata service processes and Storage Resource Management (SRM) process, enable the fault masking process, shielding MDS1 fault is to the influence of system, start and switch subprocess, client requests is switched to MDS2, and startup reparation subprocess is repaired MDS1; Start simultaneously and choose process, select the second standby meta data server MDS4.MDS3 bears the function from meta data server, starts monitoring process, monitors the other side's state mutually with MDS2, starts the metadata unanimity of metadata replication process assurance with MDS2.
When monitoring MDS2 when breaking down, MDS3 takes over the metadata services request, operation object metadata service processes and Storage Resource Management (SRM) process, start the fault masking process, shielding MDS2 fault is to the influence of system, starting the reparation subprocess repairs MDS2, and startup adds subprocess, judge whether pivot data server MDS1 repairs, if the then adding system of repairing provides the metadata service of storage system, MDS1 starts the metadata replication process and guarantees to go up all metadata unanimities with MDS3, Yun Hang object metadata service processes and Storage Resource Management (SRM) process then, start and switch subprocess, the request of client node switches to MDS1, and MDS3 continues to bear from the meta data server function, stops object metadata service processes and Storage Resource Management (SRM) process, start the metadata replication process guarantee it and MDS1 metadata synchronously, MDS4 continues to do standby meta data server; If MDS1 also at repairing phase, then provides MDS3 the metadata service of storage system, start and switch subprocess, client-requested switches to MDS3, starts to choose process, selects the 3rd standby meta data server MDS5.MDS4 bears from the meta data server function, starts monitoring process, monitors the other side's state mutually with MDS3, starts the metadata replication process and guarantees consistent with the MDS3 metadata.
According to method of the present invention, the server that shielding is broken down is converted to the pivot data server from meta data server, the response element data service request, and constantly choose standby meta data server according to " underload " rule, constitute a kind of principal and subordinate and be equipped with chain type metadata service mode.
Figure 3 shows that system architecture synoptic diagram of the present invention, meta data server 11, client node 12 and memory node 13 are three important component parts of system, by network interconnection and mutual.Meta data server 11 mainly is responsible for providing the metadata service function, mainly is the metadata information of management object; Client node 12 is responsible for sending metadata or request of data; And storage node 13 provides stores service.
According to top algorithm implementation step, from the meta data server angle, described storage system state exchange of living in as shown in Figure 4, state 41 expression systems are in normal condition, two dedicated meta data servers provide the metadata service, select standby meta data server on the backstage, state 40 expression systems too much stop service because of breaking down.Wherein state 42 represents that enabling the principal and subordinate is equipped with the chain type pattern, the three finishes the reliability services of metadata jointly, state 43 expression faults are more, system only has only the pivot data server and from meta data server, do not have more equipment to do standby meta data server, state 44 expressions hinder for some reason has only unique meta data server that the metadata service is provided in the reason system.
According to algorithm recited above and state transitions analysis, after taking the principal and subordinate to be equipped with the chain type pattern, can constantly from memory node, choose the double meta data server of doing, like this, after and if only if two dedicated meta data servers and n memory node all broke down, the metadata service just can be ended (state 40), but like this, storage system is also ended because of the whole faults of memory node, can't provide stores service, thereby system reaches " zero failure " rate in the metadata angle.
Above-mentioned analysis explanation, the principal and subordinate is equipped with the abundant resources that the service of chain type metadata makes full use of memory node, a kind of implementation method that improves metadata service reliability is provided, though theory can not reach 100% reliability (state 40 exists in theory), but from needing memory node that the requirement of access service can be provided, the metadata service reaches very high reliability, has realized " zero " failure rate of storage system from the metadata angle.The method that the present invention proposes can be applicable in the various storage systems that comprise meta data server and memory node.
Claims (4)
1. a method that improves metadata service reliability comprises the steps:
(1) normal operating procedure, under the normal condition, pivot data server operation object metadata service processes and Storage Resource Management (SRM) process provide the metadata service of storage system; Connect from meta data server by sense channel, start monitoring process and monitor the other side's state, start simultaneously and choose process, from memory node, choose average load relatively the lightest as standby meta data server; Start the consistance that the metadata replication process guarantees all metadata on it and the pivot data server from meta data server;
(2) pivot data server fault step, when monitoring after the pivot data server breaks down, take over the metadata services request from meta data server, operation object metadata service processes and Storage Resource Management (SRM) process, enable the fault masking process, the influence that shielding pivot data server fault causes system starts and switches subprocess, the metadata request of client node is switched to from meta data server, and startup reparation subprocess is repaired the pivot data server; Start simultaneously and choose process, on memory node, select average load relatively the lightest as the second standby meta data server;
(3) standby meta data server raise-position step, standby meta data server is born the function from meta data server, start monitoring process, and monitor the other side's state mutually, start the metadata replication process and guarantee it and the consistance of all metadata from meta data server from meta data server;
(4) from meta data server fault step, when monitoring when meta data server breaks down, the standby meta data server of raise-position is taken over the metadata services request, operation object metadata service processes and Storage Resource Management (SRM) process, start the fault masking process, shielding is from the influence of meta data server fault to system, start and repair subprocess repairing from meta data server, and startup adds subprocess, judge whether the pivot data server is repaired, if the then adding system of repairing provides the metadata service of storage system, start the consistance that the metadata replication process guarantees all metadata on pivot data server and the standby meta data server, move object metadata service processes and Storage Resource Management (SRM) process then, start and switch subprocess, the request of client node switches to the pivot data server, and standby meta data server continues to bear from the meta data server function, stop object metadata service processes and Storage Resource Management (SRM) process, start the consistance that the metadata replication process guarantees all metadata on it and the pivot data server, the second standby meta data server continues to do standby meta data server; If the pivot data server is also at repairing phase, then standby meta data server is provided the metadata service of storage system, start and switch subprocess, client-requested switches to standby meta data server, start and to choose process, on memory node, select average load relatively the lightest as the 3rd standby meta data server.
(5) second standby meta data server raise-position steps, if the pivot data server is not finished reparation, standby meta data server is converted to the pivot data server, the second standby meta data server is born from the meta data server function, start monitoring process, monitor the other side's state mutually with standby meta data server, start the consistance that the metadata replication process guarantees all metadata on it and the standby meta data server.
2. the method for raising metadata service reliability as claimed in claim 1 is characterized in that: (1) described object metadata service processes provides the inquiry location of metadata and upgrades operation; (2) all memory nodes in the described Storage Resource Management (SRM) management of process storage system; (3) described metadata replication process guarantees the consistance of all metadata on meta data server and the pivot data server; (4) whether normally described monitoring process periodically detects the operation of the other side's object metadata service processes and Storage Resource Management (SRM) process; (5) described fault masking process comprises that 3 subprocess: A. switch subprocess, automatically the client metadata request is switched to another meta data server after breaking down, and continual metadata service is provided; B. repair subprocess, the meta data server that breaks down is carried out necessary reparation; C. add subprocess, after the meta data server reparation is finished, join in the total system.
3. the method for raising metadata service reliability as claimed in claim 1 or 2, it is characterized in that: the described process of choosing, when choosing the lightest memory node of average load relatively, the load value of memory node is the weighted sum of system queue length, cpu busy percentage, memory usage, disk free space and network throughput; The memory node choosing is done after the standby meta data server, upgrades corresponding load value.
4. a system that improves metadata service reliability comprises client node, memory node and meta data server, and the three is by network interconnection, and client node sends file or Object Operations request, comprises metadata request and request of data; Memory node provides stores service; Meta data server comprises the pivot data server and from meta data server, they comprise object metadata service module, Storage Resource Management (SRM) module and high available modules separately, three modules externally provide the metadata service by unified interface, and (1) object metadata service module provides the inquiry location of metadata and upgrades operation; (2) all memory nodes in the Storage Resource Management (SRM) module management storage system; (3) the high availability module guarantees the metadata reliability of service, comprises metadata replication module, fault masking module and monitoring module, and (3.1) metadata replication module guarantees the consistance of all metadata on meta data server and the pivot data server; (3.2) failure message of fault masking module shielding meta data server, transparent lasting metadata request service function is provided, comprise 3 submodules: the A. switching submodule, automatically the client metadata request is switched to another meta data server after breaking down, continual metadata service is provided; B. repair submodule, the meta data server that breaks down is carried out necessary reparation; C. add submodule, after the meta data server reparation is finished, join in the total system; (3.3) whether normally monitoring module periodically detects the operation of the other side's object metadata service module and Storage Resource Management (SRM) module; It is characterized in that:
(1) described high availability module comprises and chooses module, according to history access record and self load processing power, from memory node, choose load the lightest as standby meta data server;
When (2) described memory node is chosen for standby meta data server, load the pivot data server or from all functions module of meta data server.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200710051407A CN100587692C (en) | 2007-01-26 | 2007-01-26 | Method and system for promoting metadata service reliability |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200710051407A CN100587692C (en) | 2007-01-26 | 2007-01-26 | Method and system for promoting metadata service reliability |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101059807A true CN101059807A (en) | 2007-10-24 |
| CN100587692C CN100587692C (en) | 2010-02-03 |
Family
ID=38865913
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200710051407A Expired - Fee Related CN100587692C (en) | 2007-01-26 | 2007-01-26 | Method and system for promoting metadata service reliability |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100587692C (en) |
Cited By (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010148988A1 (en) * | 2009-06-24 | 2010-12-29 | 成都市华为赛门铁克科技有限公司 | Method, device and system for taking over fault metadata server |
| CN102024022A (en) * | 2010-11-04 | 2011-04-20 | 曙光信息产业(北京)有限公司 | Method for copying metadata in distributed file system |
| CN102171995A (en) * | 2011-04-19 | 2011-08-31 | 华为技术有限公司 | Method for processing packet when server fails and router thereof |
| CN102368711A (en) * | 2011-10-25 | 2012-03-07 | 曙光信息产业(北京)有限公司 | Communication system facing parallel file system |
| CN101751415B (en) * | 2008-12-09 | 2012-03-28 | 中国移动通信集团公司 | Metadata service system, metadata synchronized method and writing server updating method |
| CN102523105A (en) * | 2011-11-30 | 2012-06-27 | 广东电子工业研究院有限公司 | Failure recovery method of data storage and applied data distribution framework thereof |
| CN102522121A (en) * | 2011-12-13 | 2012-06-27 | 记忆科技(深圳)有限公司 | Solid state disk and automatic restoration method thereof |
| CN102694825A (en) * | 2011-03-22 | 2012-09-26 | 腾讯科技（深圳）有限公司 | Data processing method and data processing system |
| CN103235748A (en) * | 2013-04-24 | 2013-08-07 | 曙光信息产业(北京)有限公司 | Method and system for managing metadata |
| WO2014087287A1 (en) * | 2012-12-04 | 2014-06-12 | International Business Machines Corporation | Replication control using eventually consistent meta-data |
| CN104202387A (en) * | 2014-08-27 | 2014-12-10 | 华为技术有限公司 | Metadata recovery method and related device |
| CN104239425A (en) * | 2014-08-25 | 2014-12-24 | 华为技术有限公司 | File access method, file access device and file access system |
| CN104506625A (en) * | 2014-12-22 | 2015-04-08 | 国云科技股份有限公司 | Method for improving reliability of metadata nodes of cloud databases |
| CN105205156A (en) * | 2015-09-25 | 2015-12-30 | 浪潮(北京)电子信息产业有限公司 | Data management method, device and system |
| CN105373431A (en) * | 2015-10-29 | 2016-03-02 | 武汉联影医疗科技有限公司 | Computer system resource management method and computer resource management system |
| CN106469087A (en) * | 2015-08-19 | 2017-03-01 | 阿里巴巴集团控股有限公司 | Metadata output intent, client and meta data server |
| CN107633070A (en) * | 2017-09-22 | 2018-01-26 | 郑州云海信息技术有限公司 | Balance Control Scheme method, apparatus and storage medium without the MDS of configuration |
| CN107665253A (en) * | 2017-09-22 | 2018-02-06 | 郑州云海信息技术有限公司 | Configurable MDS balance control method, device and storage medium |
| CN107710164A (en) * | 2015-03-31 | 2018-02-16 | 伊姆西知识产权控股有限责任公司 | As a kind of disaster recovery of service |
| CN107729178A (en) * | 2017-09-28 | 2018-02-23 | 郑州云海信息技术有限公司 | A kind of Metadata Service process takes over method and device |
| CN108647248A (en) * | 2018-04-16 | 2018-10-12 | 新华三技术有限公司成都分公司 | WORM state monitors transfer method and device |
| CN109492013A (en) * | 2018-11-02 | 2019-03-19 | 北京京东金融科技控股有限公司 | Data processing method, device and system applied to data-base cluster |
| CN109582213A (en) * | 2017-09-29 | 2019-04-05 | 杭州海康威视系统技术有限公司 | Data reconstruction method and device, data-storage system |
| CN109936613A (en) * | 2017-12-19 | 2019-06-25 | 北京京东尚科信息技术有限公司 | Disaster recovery method and device applied to server |
| CN111381965A (en) * | 2020-03-03 | 2020-07-07 | 百度在线网络技术（北京）有限公司 | Method and apparatus for processing requests |
| CN112346940A (en) * | 2019-08-08 | 2021-02-09 | 北京神州泰岳软件股份有限公司 | Performance acquisition method, device and equipment |
| CN115757330A (en) * | 2022-12-08 | 2023-03-07 | 丝路信息港云计算科技有限公司 | Highly reliable metadata service system of distributed file system |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102891868B (en) * | 2011-07-19 | 2016-09-28 | 上海可鲁系统软件有限公司 | The load-balancing method of a kind of distributed system and device |
-
2007
- 2007-01-26 CN CN200710051407A patent/CN100587692C/en not_active Expired - Fee Related
Cited By (46)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101751415B (en) * | 2008-12-09 | 2012-03-28 | 中国移动通信集团公司 | Metadata service system, metadata synchronized method and writing server updating method |
| CN101577735B (en) * | 2009-06-24 | 2012-04-25 | 成都市华为赛门铁克科技有限公司 | Method, device and system for taking over fault metadata server |
| WO2010148988A1 (en) * | 2009-06-24 | 2010-12-29 | 成都市华为赛门铁克科技有限公司 | Method, device and system for taking over fault metadata server |
| CN102024022A (en) * | 2010-11-04 | 2011-04-20 | 曙光信息产业(北京)有限公司 | Method for copying metadata in distributed file system |
| CN102024022B (en) * | 2010-11-04 | 2013-08-28 | 曙光信息产业(北京)有限公司 | Method for copying metadata in distributed file system |
| CN102694825A (en) * | 2011-03-22 | 2012-09-26 | 腾讯科技（深圳）有限公司 | Data processing method and data processing system |
| CN102171995B (en) * | 2011-04-19 | 2013-09-11 | 华为技术有限公司 | Method for processing packet when server fails and router thereof |
| CN102171995A (en) * | 2011-04-19 | 2011-08-31 | 华为技术有限公司 | Method for processing packet when server fails and router thereof |
| US9391854B2 (en) | 2011-04-19 | 2016-07-12 | Huawei Technologies Co., Ltd. | Method and router for packet processing during server failure |
| CN102368711A (en) * | 2011-10-25 | 2012-03-07 | 曙光信息产业(北京)有限公司 | Communication system facing parallel file system |
| CN102368711B (en) * | 2011-10-25 | 2014-05-21 | 曙光信息产业(北京)有限公司 | Communication system facing parallel file system |
| CN102523105B (en) * | 2011-11-30 | 2014-03-26 | 广东电子工业研究院有限公司 | Failure recovery method of data storage and applied data distribution framework thereof |
| CN102523105A (en) * | 2011-11-30 | 2012-06-27 | 广东电子工业研究院有限公司 | Failure recovery method of data storage and applied data distribution framework thereof |
| CN102522121A (en) * | 2011-12-13 | 2012-06-27 | 记忆科技(深圳)有限公司 | Solid state disk and automatic restoration method thereof |
| CN102522121B (en) * | 2011-12-13 | 2015-09-02 | 记忆科技(深圳)有限公司 | The method that solid state hard disc is repaired automatically and solid state hard disc thereof |
| WO2014087287A1 (en) * | 2012-12-04 | 2014-06-12 | International Business Machines Corporation | Replication control using eventually consistent meta-data |
| US10747776B2 (en) | 2012-12-04 | 2020-08-18 | International Business Machines Corporation | Replication control using eventually consistent meta-data |
| US10726042B2 (en) | 2012-12-04 | 2020-07-28 | International Business Machines Corporation | Replication control using eventually consistent meta-data |
| CN103235748A (en) * | 2013-04-24 | 2013-08-07 | 曙光信息产业(北京)有限公司 | Method and system for managing metadata |
| CN104239425A (en) * | 2014-08-25 | 2014-12-24 | 华为技术有限公司 | File access method, file access device and file access system |
| CN104239425B (en) * | 2014-08-25 | 2017-11-17 | 华为技术有限公司 | The method, apparatus and system of file access |
| CN104202387A (en) * | 2014-08-27 | 2014-12-10 | 华为技术有限公司 | Metadata recovery method and related device |
| CN104202387B (en) * | 2014-08-27 | 2017-11-24 | 华为技术有限公司 | A kind of metadata restoration methods and relevant apparatus |
| CN104506625A (en) * | 2014-12-22 | 2015-04-08 | 国云科技股份有限公司 | Method for improving reliability of metadata nodes of cloud databases |
| CN104506625B (en) * | 2014-12-22 | 2018-04-17 | 国云科技股份有限公司 | A kind of method for lifting cloud database metadata node reliability |
| CN107710164A (en) * | 2015-03-31 | 2018-02-16 | 伊姆西知识产权控股有限责任公司 | As a kind of disaster recovery of service |
| CN106469087A (en) * | 2015-08-19 | 2017-03-01 | 阿里巴巴集团控股有限公司 | Metadata output intent, client and meta data server |
| CN105205156A (en) * | 2015-09-25 | 2015-12-30 | 浪潮(北京)电子信息产业有限公司 | Data management method, device and system |
| CN105373431B (en) * | 2015-10-29 | 2022-09-27 | 武汉联影医疗科技有限公司 | Computer system resource management method and computer resource management system |
| CN105373431A (en) * | 2015-10-29 | 2016-03-02 | 武汉联影医疗科技有限公司 | Computer system resource management method and computer resource management system |
| CN107633070A (en) * | 2017-09-22 | 2018-01-26 | 郑州云海信息技术有限公司 | Balance Control Scheme method, apparatus and storage medium without the MDS of configuration |
| CN107665253A (en) * | 2017-09-22 | 2018-02-06 | 郑州云海信息技术有限公司 | Configurable MDS balance control method, device and storage medium |
| CN107665253B (en) * | 2017-09-22 | 2022-02-18 | 郑州云海信息技术有限公司 | Configurable MDS balance control method and device and storage medium |
| CN107729178A (en) * | 2017-09-28 | 2018-02-23 | 郑州云海信息技术有限公司 | A kind of Metadata Service process takes over method and device |
| CN109582213A (en) * | 2017-09-29 | 2019-04-05 | 杭州海康威视系统技术有限公司 | Data reconstruction method and device, data-storage system |
| CN109582213B (en) * | 2017-09-29 | 2020-10-30 | 杭州海康威视系统技术有限公司 | Data reconstruction method and device and data storage system |
| CN109936613A (en) * | 2017-12-19 | 2019-06-25 | 北京京东尚科信息技术有限公司 | Disaster recovery method and device applied to server |
| CN108647248B (en) * | 2018-04-16 | 2021-03-09 | 新华三技术有限公司成都分公司 | WORM state monitoring transfer method and device |
| CN108647248A (en) * | 2018-04-16 | 2018-10-12 | 新华三技术有限公司成都分公司 | WORM state monitors transfer method and device |
| CN109492013A (en) * | 2018-11-02 | 2019-03-19 | 北京京东金融科技控股有限公司 | Data processing method, device and system applied to data-base cluster |
| CN109492013B (en) * | 2018-11-02 | 2021-01-26 | 京东数字科技控股有限公司 | Data processing method, device and system applied to database cluster |
| CN112346940A (en) * | 2019-08-08 | 2021-02-09 | 北京神州泰岳软件股份有限公司 | Performance acquisition method, device and equipment |
| CN112346940B (en) * | 2019-08-08 | 2024-06-07 | 北京神州泰岳软件股份有限公司 | Performance acquisition method, device and equipment |
| CN111381965A (en) * | 2020-03-03 | 2020-07-07 | 百度在线网络技术（北京）有限公司 | Method and apparatus for processing requests |
| CN111381965B (en) * | 2020-03-03 | 2023-01-31 | 百度在线网络技术（北京）有限公司 | Method and apparatus for processing requests |
| CN115757330A (en) * | 2022-12-08 | 2023-03-07 | 丝路信息港云计算科技有限公司 | Highly reliable metadata service system of distributed file system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100587692C (en) | 2010-02-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100587692C (en) | Method and system for promoting metadata service reliability | |
| CN102394774B (en) | Service state monitoring and failure recovery method for controllers of cloud computing operating system | |
| US9632887B2 (en) | Automatic client side seamless failover | |
| CN111917846A (en) | Kafka cluster switching method, device and system, electronic equipment and readable storage medium | |
| CN101179432A (en) | Method of implementing high availability of system in multi-machine surroundings | |
| CN102467508A (en) | Method for providing database service and database system | |
| CN102394914A (en) | Cluster brain-split processing method and device | |
| CN1493031A (en) | Twin-tailed fail-over for file servers maintaniing full performance in presence of failure | |
| CN102739453A (en) | Master-standby switching method, device and system | |
| CN113867226B (en) | Redundant data acquisition system for industrial control system | |
| CN109254876A (en) | The management method and device of database in cloud computing system | |
| CN101056254A (en) | An expansion method, system and device of network storage device | |
| WO2006005251A1 (en) | Method and system for carrying out the function of switching in a communication system | |
| US20040153704A1 (en) | Automatic startup of a cluster system after occurrence of a recoverable error | |
| CN102111281A (en) | Hot standby method for implementing dual-machine hot standby system | |
| JP2008535294A (en) | Fail-safe stream processing | |
| CN1598778A (en) | Data disaster tocerance backup control system | |
| JP5154843B2 (en) | Cluster system, computer, and failure recovery method | |
| WO2011083778A1 (en) | Computer, computer management method, and computer management program | |
| CN1275476C (en) | Clustering system for utilizing sharing internal memory in mobile communiation system and realizing method thereof | |
| CN110675614A (en) | Transmission method of power monitoring data | |
| CN111314129B (en) | High-availability architecture based on file type storage service | |
| CN1710864A (en) | Upgradable communication system and method for upgrading communication system | |
| CN106534262A (en) | Network information system fault switching method | |
| CN116400860A (en) | Fusion type storage method, device, system and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20100203 Termination date: 20220126 |