CN102904948A - Super-large-scale low-cost storage system - Google Patents
Super-large-scale low-cost storage system Download PDFInfo
- Publication number
- CN102904948A CN102904948A CN2012103766530A CN201210376653A CN102904948A CN 102904948 A CN102904948 A CN 102904948A CN 2012103766530 A CN2012103766530 A CN 2012103766530A CN 201210376653 A CN201210376653 A CN 201210376653A CN 102904948 A CN102904948 A CN 102904948A
- Authority
- CN
- China
- Prior art keywords
- node
- metadata
- management node
- volume
- storage system
- 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.)
- Pending
Links
Images
Abstract
The invention discloses a super-large-scale low-cost storage system. The system comprises a volume management node, a metadata management node, a block data storage node, a user mount client and a management monitoring center. According to the system, the problems, such as poor usability, low integration level, weak expansibility and relatively high cost, in the conventional storage scheme are solved, and the super-large-scale low-cost storage system which is high in reliability, excellent in performance, easy to maintain, strong in expansibility and high in density is realized. According to the system established by the invention, the volume management node and the metadata management node both adopt a dual-computer hot standby mode, and thus, the system reliability is improved greatly.
Description
Technical field
The present invention relates to computer and network information technology field, relate in particular to a kind of ultra-large low-cost storage system.
Background technology
Develop rapidly along with the information age, the explosive growth of amount of information has become a kind of characteristics of the times, thing followed problem is the storage problem of mass data, the storage of traditional hard-disc type obviously is difficult to satisfy the demands, the direct-connected storage of DAS(Direct-Attached Storage--that occurred afterwards) storage mode, solved the problem of storage data volume, but discrete DAS storage forms one by one isolated island, when a memory capacity saturated, even other DAS equipment has capacity more than needed also to need to buy new memory device, and newly add a server and also will newly add a DAS, carrying cost is higher, the storage of afterwards NAS(Network Attached Storage--network connection formula) and SAN(Storage Area Network--storage networking) solved the public problem of memory space, but the growth along with data volume, the performance of cluster has become again subject matter with extensibility, also just can't realize the structure of ultra-large low-cost storage system.
Summary of the invention
Purpose of the invention process is to solve that the ease for use that exists in the existing storage solution is poor, integrated level is not high, autgmentability is not strong, the problems such as cost is higher make up a kind of have simultaneously highly reliable, excellent performance, easy care, strong, the highdensity ultra-large cheaply storage system of autgmentability.
The present invention adopts following technical scheme:
A kind of ultra-large low-cost storage system comprises with lower module:
Volume management node (SuMaster)
The information of all meta data server sub-cluster of this node maintenance is for the carry client provides meta data server sign, IP address and port number information; Comprise master file management server (SuperMaster) and standby volume management server (Slave SuperMaster);
Metadata management node (Master)
Management of metadata is for the carry client provides the metadata access service; Comprise pivot data server (Master) and standby meta data server (Slave Master);
Blocks of data memory node (Data Node)
The storage data block, and according to the backup policy of configuration data block is sent to by another selected blocks of data memory node of metadata management node and to carry out the data block backup;
User's carry client (MountClient)
File system mounted client can be to using the transparent common single-point file system that is behind the carry.
Management and monitoring center (NMC)
Comprise configuration center and Surveillance center, be in charge of and monitor whole storage system, for the keeper provides powerful management function, the resource distribution of optimizing hardware.
Storage system all is meta data server, blocks of data server and the access client of separate unit in the prior art, metadata all is kept in the internal memory, because internal memory is limited can't accomplish extensive expansion, thereby the total number of files amount of storage is limited, each client can only be accessed the file in the fixing one by one little cluster, and the ultra-large low-cost storage system that the present invention makes up is take the ordinary magnetic disc of cheapness as medium, can realize that client accesses All Files automatically in a plurality of clusters.But 16 hard disks of storage server single node carry are that 3TB calculates according to current single-deck capacity, and the single node capacity can reach 48TB.Adopt the framework of volume management server+sub-cluster can satisfy the 100PB capacity requirement.The 42U frame of a standard can provide the capacity up to 1PB (1024TB).The storage system that makes up according to this method can provide mass data storage, the capacity no maximum, and the theory support capacity can reach the GP level, and namely 1024*1024*1024PB really accomplishes ultra-large low-cost storage.
The system that the present invention makes up, volume management node and metadata management node all adopt the two-node cluster hot backup mode, have greatly improved the reliability of system.
Description of drawings
Fig. 1 is system architecture diagram of the present invention;
Fig. 2 is the networking mode of system of the present invention and disposes Organization Chart;
Fig. 3 is the central server ideograph;
Fig. 4 is Hash ring schematic diagram.
Embodiment
Below in conjunction with accompanying drawing technical scheme of the present invention is elaborated:
As shown in Figure 1 and Figure 2, system of the present invention comprises volume management node (SuMaster), metadata management node (Master), blocks of data memory node (Data Node), user's carry client (MountClient) and management and monitoring center (NMC).
The information of all meta data server sub-cluster of volume management node maintenance, for the carry client provides meta data server sign, IP address and port number information, comprise master file management server (SuperMaster) and standby volume management server (Slave SuperMaster), active and standby volume management server is by same VIP(virtual IP address) service externally is provided, it is unified that active and standby volume management server adds both states of configure and maintenance by Surveillance center.
Metadata management node administration metadata, for the carry client provides the metadata access service, comprise pivot data server (Master) and standby meta data server (Slave Master), active and standby meta data server is by scoket network service, the each metadata synchronization that changes of real-time handle is to standby meta data server, if metadata is inconsistent, reach synchronously by full dose then that both are consistent, active and standby meta data server externally provides service by same VIP.
Blocks of data memory node storage data block, and according to the backup policy of configuration data block is sent to another blocks of data memory node and carry out the data block backup, backup node according to current loading condition, is selected the lightest blocks of data memory node of load by Master.
The management and monitoring center comprises configuration center and Surveillance center, is in charge of and monitors whole storage system, for the keeper provides powerful management function, and the resource distribution of optimizing hardware.Configuration center comprises volume configuration, node parameter configuration, stored parameter configuration, user's quotas administered and alarm configuration.Surveillance center comprises memory space monitoring, device status monitoring, program state monitoring, network status monitoring, Disk State monitoring, traffic monitoring and comprehensively alarm.Volume can be added and delete to the volume configuration; The node parameter configuration can arrange blocks of data memory node configuration parameter; The stored parameter configuration can arrange volume management node and metadata management node configuration parameter; User's quotas administered can arrange different user can use the size of memory space and the number of file; Alarm configuration is used for arranging alarm level and alarm type.Size and operating position that the memory space monitoring can be checked each volume space; Device status monitoring can check whether each node device is normal; Whether normally the program state monitoring can check the operation of each program; Network status monitoring is checked the network health situation; The Disk State monitoring checks whether disk has damage; The node flow situation is checked in traffic monitoring; Alarm comprises that disk failures, system space are full, the term of validity expires and the hardware damage alarm comprehensively.
The below further specifies framework and the organizational form of system of the present invention from the angle of data storage.
1. metadata store
● metadata store and is carried out persistence backup with the machine in the internal memory of metadata management node (Master) server;
● the metadata management node is active and standby two-shipper mode, and highly reliable uninterrupted metadata management service is provided, and can realize seamless quick switching during the unit fault, externally provides service by same VIP;
● quantity of documents is many when considering mass data storage, metadata problem capacious, and system carries out distributed storage with metadata, adopts a plurality of metadata management nodes to carry out metadata management;
● the storage of meta-data distribution formula, each metadata management node all is registered on the volume management server, and each metadata management node has unique identification; MasterID(meta data server sign), IP address, port number information configure and maintenance in sub-cluster information management node volume management node;
● high N position is MasterID in the metadata management node ID, has avoided different metadata management nodes to produce node ID and has repeated; The N here can be configured according to service application scene demand;
● the carry client determines according to the high N position of Archive sit sign which metadata management node this mails to when time request.
2. distributed storage
The present invention adopts the central server pattern to manage whole file system.Center management server ideograph such as accompanying drawing 3:
The metadata of a volume in the metadata management node administration distributed file system.Divide File is that piece is stored, and for the metadata management node, each blocks of data memory node all is registered to the metadata management node, and the blocks of data memory node is a memory space.The all operations that client is initiated all needs could carry out by the metadata management node first.Done so many benefits, increasing new blocks of data memory node is a very easy thing, and the blocks of data memory node only need to be registered on the metadata management node and get final product, between the blocks of data memory node without any relation.The metadata management node maintenance a unified NameSpace, grasp simultaneously the situation of blocks of data memory node in the whole system, can realize accordingly the load balancing of data storage in the whole system scope.Owing to only have a central server, the consistency problem of metadata solves naturally.
Client is carried out piecemeal to file, then can select different pieces to store on the different storage servers, and next piece stores on the smaller storage server of load, and relatively whole memory space has been realized distributed storage.
3. carry client-access
● client is obtained secondary metadata management nodal information from the volume management node first according to certain Hash strategy, then obtains metadata information from secondary metadata management node;
● client and metadata management node carry out the metadata associative operation;
● read and write data and frontly obtain blocks of data memory node information from the metadata management node first, then communicate by letter with the blocks of data memory node and carry out data read-write operation;
● for large file, data are divided into polylith take 64MB as unit, and Master is according to polling method allocation block memory node, and each blocks of data is distributed on the different blocks of data memory nodes simultaneously;
● the small documents (less than 64MB) for One's name is legion, then do not need piecemeal, Master is distributed to each small documents on the different blocks of data memory nodes simultaneously according to polling method allocation block memory node;
● in order to solve the reliability of Internet Transmission bottleneck and raising transfer of data, all data of writing on the blocks of data memory node, adopt formation (first-in first-out) caching data block, be cached to first client, all data are write client and are thought namely that later on data write end;
● in order to improve transmission rate, in data cached, start a plurality of threads and be sent to a plurality of blocks of data memory node the transmission of datas;
● above various strategies have fully guaranteed the saturated utilization to bandwidth when reading and writing data of single client.
4, client
The application of system is mainly reflected in client to the access of system:
1) client is obtained all metadata management node configuration informations of cluster to the volume management node, comprises MasterID, IP address, listening port;
2) client obtains the large integer (as sign) of a 128bit by MD5 algorithm (MasterID) Hash, and the information Hash of the sign of all metadata management nodes obtains a large integer circulus.Hash ring schematic diagram such as accompanying drawing 4
When 3) client is carried out file operation, if the operation under the mount point root is then carried out MD5 according to first order catalogue and calculated a value, go on foot from the 2nd by this value and find the Master sub-cluster of processing this document folder the circulus that obtains.If the operation of the sub-folder under the mount point root, known current file or folder metadata management node i node ID(node ID to be processed this moment), can obtain the MasterID numbering according to the former position that the position information that the volume management node returns is got inode ID, thereby determine to process the Master sub-cluster of this document folder.
4) to the All Files under root operation, all operate in first Master sub-cluster, and the file under the subdirectory or catalogue will be hashing onto different Master sub-cluster and operate.
Claims (9)
1. ultra-large low-cost storage system is characterized in that comprising:
The volume management node:
Safeguard the information of all meta data server sub-cluster, for the carry client provides meta data server sign, IP address and port number information; Comprise master file management server and standby volume management server;
The metadata management node:
Management of metadata is for the carry client provides the metadata access service; Comprise pivot data server and standby meta data server;
The blocks of data memory node:
The storage data block, and according to the backup policy of configuration data block is sent to by another selected blocks of data memory node of metadata management node and to carry out the data block backup;
User's carry client:
File system mounted client, transparent in common single-point file system to using behind the carry;
The management and monitoring center:
Comprise configuration center and Surveillance center, be in charge of and monitor whole storage system, for the keeper provides powerful management function, the resource distribution of optimizing hardware.
2. ultra-large low-cost storage system as claimed in claim 1, it is characterized in that described active and standby volume management server externally provides service by same VIP, it is unified that active and standby volume management server adds both states of configure and maintenance by the management and monitoring center.
3. ultra-large low-cost storage system as claimed in claim 1, it is characterized in that active and standby meta data server is by scoket network service, in real time the metadata synchronization of each variation is arrived standby meta data server, if metadata is inconsistent, reach synchronously by full dose then that both are consistent, active and standby meta data server externally provides service by same VIP.
4. ultra-large low-cost storage system as claimed in claim 1 is characterized in that configuration center comprises volume configuration, node parameter configuration, stored parameter configuration, user's quotas administered and alarm configuration; The volume configuration is used for adding and the deletion volume; The node parameter configuration is used for arranging blocks of data memory node configuration parameter; The stored parameter configuration is used for arranging volume management node and metadata management node configuration parameter; User's quotas administered is used for arranging different user and uses the size of memory space and the number of file; Alarm configuration is used for arranging alarm level and alarm type.
5. ultra-large low-cost storage system as claimed in claim 1 is characterized in that Surveillance center comprises memory space monitoring, device status monitoring, program state monitoring, network status monitoring, Disk State monitoring, traffic monitoring and comprehensively alarm; The memory space monitoring is for size and the operating position of checking each volume space; Device status monitoring is used for checking whether each node device is normal; Whether normally the program state monitoring is used for checking the operation of each program; Network status monitoring is used for checking the network health situation; The Disk State monitoring is used for checking whether disk has damage; Traffic monitoring is used for checking the node flow situation; Alarm comprises that disk failures, system space are full, the term of validity expires and the hardware damage alarm comprehensively.
6. ultra-large low-cost storage system as claimed in claim 1 is characterized in that: metadata adopts distributed storage, is managed by a plurality of metadata management nodes.
7. ultra-large low-cost storage system as claimed in claim 1, it is characterized in that: each metadata management node all is registered on the volume management server, each metadata management node has unique identification, meta data server sign, IP address, port number information configure and maintenance in the volume management node.
8. ultra-large low-cost storage system as claimed in claim 1, it is characterized in that: the metadata of a volume in each metadata management node administration distributed file system, Divide File is that piece is stored, each blocks of data memory node all is registered to the metadata management node, for the metadata management node, each blocks of data memory node is a memory space, and all operations that the carry client is initiated all needs could carry out by the metadata management node first.
9. ultra-large low-cost storage system as claimed in claim 8, it is characterized in that: the carry client is obtained secondary metadata management nodal information from the volume management node first according to certain Hash strategy, then obtains metadata information from secondary metadata management node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012103766530A CN102904948A (en) | 2012-09-29 | 2012-09-29 | Super-large-scale low-cost storage system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012103766530A CN102904948A (en) | 2012-09-29 | 2012-09-29 | Super-large-scale low-cost storage system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102904948A true CN102904948A (en) | 2013-01-30 |
Family
ID=47576983
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2012103766530A Pending CN102904948A (en) | 2012-09-29 | 2012-09-29 | Super-large-scale low-cost storage system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102904948A (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103647797A (en) * | 2013-11-15 | 2014-03-19 | 北京邮电大学 | Distributed file system and data access method thereof |
| CN104104739A (en) * | 2014-08-08 | 2014-10-15 | 成都致云科技有限公司 | Data synchronization method for metadata service cluster of cloud storage system |
| CN104539670A (en) * | 2014-12-16 | 2015-04-22 | 曙光信息产业(北京)有限公司 | Cluster management method and device |
| CN104967691A (en) * | 2015-07-08 | 2015-10-07 | 浪潮(北京)电子信息产业有限公司 | Distributed storage control method and system |
| CN105187517A (en) * | 2015-08-20 | 2015-12-23 | 浪潮(北京)电子信息产业有限公司 | Metadata clustering method and metadata cluster |
| CN105227643A (en) * | 2015-09-11 | 2016-01-06 | 武汉思捷云信息科技有限公司 | A kind of storage emerging system based on cloud storage platform facing video monitoring and method |
| CN105335095A (en) * | 2014-08-11 | 2016-02-17 | 北京兆易创新科技股份有限公司 | Flash file system processing method and apparatus |
| CN105939387A (en) * | 2016-06-23 | 2016-09-14 | 中国南方电网有限责任公司 | Multilevel load balancing and server role switching method based on network, province and district uniform dispatching |
| CN106815298A (en) * | 2016-12-09 | 2017-06-09 | 中电科华云信息技术有限公司 | Distributed sharing file system based on block storage |
| CN107111481A (en) * | 2014-10-03 | 2017-08-29 | 新加坡科技研究局 | Distribution actively mixes storage system |
| CN107343012A (en) * | 2016-04-28 | 2017-11-10 | 杭州海康威视数字技术股份有限公司 | Video Storage System and its video data transmitting method |
| CN107491360A (en) * | 2017-07-21 | 2017-12-19 | 山东九州信泰信息科技股份有限公司 | The method for being classified other redundant storage is carried out to record in tables of data |
| CN109522053A (en) * | 2017-09-20 | 2019-03-26 | 阿里巴巴集团控股有限公司 | A kind of massive parallel processing and data processing method |
| CN110119984A (en) * | 2018-02-07 | 2019-08-13 | 青岛农业大学 | A kind of processing system for international trade tick financing |
| CN110169040A (en) * | 2018-07-10 | 2019-08-23 | 深圳花儿数据技术有限公司 | Distributed data storage method and system based on multilayer consistency Hash |
| CN111813332A (en) * | 2020-05-21 | 2020-10-23 | 华北电力大学 | High-performance, high-expansion and high-safety intelligent distributed storage system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7181578B1 (en) * | 2002-09-12 | 2007-02-20 | Copan Systems, Inc. | Method and apparatus for efficient scalable storage management |
| CN101997884A (en) * | 2009-08-18 | 2011-03-30 | 升东网络科技发展(上海)有限公司 | Distributed storage system and method |
| CN101997918A (en) * | 2010-11-11 | 2011-03-30 | 清华大学 | Method for allocating mass storage resources according to needs in heterogeneous SAN (Storage Area Network) environment |
| CN102143215A (en) * | 2011-01-20 | 2011-08-03 | 中国人民解放军理工大学 | Network-based PB level cloud storage system and processing method thereof |
| CN102307221A (en) * | 2011-03-25 | 2012-01-04 | 国云科技股份有限公司 | Cloud storage system and implementation method thereof |
| CN102546823A (en) * | 2012-02-18 | 2012-07-04 | 南京云创存储科技有限公司 | File storage management system of cloud storage system |
-
2012
- 2012-09-29 CN CN2012103766530A patent/CN102904948A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7181578B1 (en) * | 2002-09-12 | 2007-02-20 | Copan Systems, Inc. | Method and apparatus for efficient scalable storage management |
| CN101997884A (en) * | 2009-08-18 | 2011-03-30 | 升东网络科技发展(上海)有限公司 | Distributed storage system and method |
| CN101997918A (en) * | 2010-11-11 | 2011-03-30 | 清华大学 | Method for allocating mass storage resources according to needs in heterogeneous SAN (Storage Area Network) environment |
| CN102143215A (en) * | 2011-01-20 | 2011-08-03 | 中国人民解放军理工大学 | Network-based PB level cloud storage system and processing method thereof |
| CN102307221A (en) * | 2011-03-25 | 2012-01-04 | 国云科技股份有限公司 | Cloud storage system and implementation method thereof |
| CN102546823A (en) * | 2012-02-18 | 2012-07-04 | 南京云创存储科技有限公司 | File storage management system of cloud storage system |
Cited By (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103647797A (en) * | 2013-11-15 | 2014-03-19 | 北京邮电大学 | Distributed file system and data access method thereof |
| CN104104739A (en) * | 2014-08-08 | 2014-10-15 | 成都致云科技有限公司 | Data synchronization method for metadata service cluster of cloud storage system |
| CN105335095B (en) * | 2014-08-11 | 2019-04-05 | 北京兆易创新科技股份有限公司 | The processing method and processing device of flash memory file system |
| CN105335095A (en) * | 2014-08-11 | 2016-02-17 | 北京兆易创新科技股份有限公司 | Flash file system processing method and apparatus |
| CN107111481A (en) * | 2014-10-03 | 2017-08-29 | 新加坡科技研究局 | Distribution actively mixes storage system |
| CN104539670A (en) * | 2014-12-16 | 2015-04-22 | 曙光信息产业(北京)有限公司 | Cluster management method and device |
| CN104539670B (en) * | 2014-12-16 | 2018-09-18 | 曙光信息产业(北京)有限公司 | cluster management method and device |
| CN104967691B (en) * | 2015-07-08 | 2018-06-15 | 浪潮(北京)电子信息产业有限公司 | A kind of distributed storage control method and system |
| CN104967691A (en) * | 2015-07-08 | 2015-10-07 | 浪潮(北京)电子信息产业有限公司 | Distributed storage control method and system |
| CN105187517A (en) * | 2015-08-20 | 2015-12-23 | 浪潮(北京)电子信息产业有限公司 | Metadata clustering method and metadata cluster |
| CN105227643A (en) * | 2015-09-11 | 2016-01-06 | 武汉思捷云信息科技有限公司 | A kind of storage emerging system based on cloud storage platform facing video monitoring and method |
| CN107343012A (en) * | 2016-04-28 | 2017-11-10 | 杭州海康威视数字技术股份有限公司 | Video Storage System and its video data transmitting method |
| CN107343012B (en) * | 2016-04-28 | 2019-12-03 | 杭州海康威视数字技术股份有限公司 | Video Storage System and its video data transmitting method |
| CN105939387A (en) * | 2016-06-23 | 2016-09-14 | 中国南方电网有限责任公司 | Multilevel load balancing and server role switching method based on network, province and district uniform dispatching |
| CN106815298A (en) * | 2016-12-09 | 2017-06-09 | 中电科华云信息技术有限公司 | Distributed sharing file system based on block storage |
| CN107491360B (en) * | 2017-07-21 | 2019-07-30 | 山东九州信泰信息科技股份有限公司 | The method for being classified other redundant storage is carried out to record in tables of data |
| CN107491360A (en) * | 2017-07-21 | 2017-12-19 | 山东九州信泰信息科技股份有限公司 | The method for being classified other redundant storage is carried out to record in tables of data |
| CN109522053A (en) * | 2017-09-20 | 2019-03-26 | 阿里巴巴集团控股有限公司 | A kind of massive parallel processing and data processing method |
| CN110119984A (en) * | 2018-02-07 | 2019-08-13 | 青岛农业大学 | A kind of processing system for international trade tick financing |
| CN110169040A (en) * | 2018-07-10 | 2019-08-23 | 深圳花儿数据技术有限公司 | Distributed data storage method and system based on multilayer consistency Hash |
| WO2020010503A1 (en) * | 2018-07-10 | 2020-01-16 | 深圳花儿数据技术有限公司 | Multi-layer consistent hashing-based distributed data storage method and system |
| CN110169040B (en) * | 2018-07-10 | 2021-09-28 | 深圳花儿数据技术有限公司 | Distributed data storage method and system based on multilayer consistent hash |
| US11461203B2 (en) | 2018-07-10 | 2022-10-04 | Here Data Technology | Systems and methods of handling node failure in a distributed data storage using multi-layer consistent hashing |
| CN111813332A (en) * | 2020-05-21 | 2020-10-23 | 华北电力大学 | High-performance, high-expansion and high-safety intelligent distributed storage system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102904948A (en) | Super-large-scale low-cost storage system | |
| CN107547653B (en) | Distributed file storage system | |
| CN103763383B (en) | Integrated cloud storage system and its storage method | |
| CN111124301B (en) | Data consistency storage method and system of object storage device | |
| TWI759789B (en) | Computer-implemented method, non-transitory computer readable media and storage system | |
| CN105335513B (en) | A kind of distributed file system and file memory method | |
| CN106066896B (en) | Application-aware big data deduplication storage system and method | |
| TWI766309B (en) | Log-structured storage systems | |
| CN107046563A (en) | A kind of implementation method, system and the cloud platform of distribution type high efficient cloud disk | |
| CN103067461B (en) | A kind of metadata management system of file and metadata management method | |
| CN103929500A (en) | Method for data fragmentation of distributed storage system | |
| CN102833580B (en) | Based on HD video application system and the method for infiniband | |
| CN108881942B (en) | Super-fusion normal state recorded broadcast system based on distributed object storage | |
| CN103455577A (en) | Multi-backup nearby storage and reading method and system of cloud host mirror image file | |
| TW202111564A (en) | Log-structured storage systems | |
| EP2062123A2 (en) | Automatic load spreading in a clustered network storage system | |
| US20150205819A1 (en) | Techniques for optimizing data flows in hybrid cloud storage systems | |
| CN102708158B (en) | PostgreSQL (postgres structured query language) cloud storage filing and scheduling system | |
| CN103888499A (en) | Distributed object processing method and system | |
| Zhang et al. | Survey of research on big data storage | |
| CN103067488A (en) | Implement method of unified storage | |
| CN102664914A (en) | IS/DFS-Image distributed file storage query system | |
| CN103501319A (en) | Low-delay distributed storage system for small files | |
| CN109522283A (en) | A kind of data de-duplication method and system | |
| CN108540510B (en) | Cloud host creation method and device and cloud service 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 | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20130130 |