CN105610921B - Correcting and eleting codes archiving method based on data buffer storage under a kind of cluster - Google Patents
Correcting and eleting codes archiving method based on data buffer storage under a kind of cluster Download PDFInfo
- Publication number
- CN105610921B CN105610921B CN201510979326.8A CN201510979326A CN105610921B CN 105610921 B CN105610921 B CN 105610921B CN 201510979326 A CN201510979326 A CN 201510979326A CN 105610921 B CN105610921 B CN 105610921B
- Authority
- CN
- China
- Prior art keywords
- deblocking
- archival
- node
- cluster
- data
- 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.)
- Expired - Fee Related
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/56—Provisioning of proxy services
- H04L67/568—Storing data temporarily at an intermediate stage, e.g. caching
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Memory System Of A Hierarchy Structure (AREA)
Abstract
The invention discloses the correcting and eleting codes archiving methods based on data buffer storage under a kind of cluster, include the following steps：(1) according to user access request, by required deblocking, production node where it reads node buffering area, and deblocking is sent to accessed node, while forwarding it to archival one according to filing data mode table；(2) archival one updates the deblocking number of each back end according to the deblocking received；Judge whether archival one receives the total data piecemeal of current band, if so, current band is filed；If it is not, then reading cold data block from production cluster, file after band is gathered together enough；This correcting and eleting codes archiving method provided by the invention, as a result of the deblocking of user's access cache, compared to common correcting and eleting codes archiving method, the number that archival one reads deblocking from production cluster can be greatly reduced, correcting and eleting codes archive operation can be carried out in the case where not influencing user and accessing, improve filing efficiency.
Description
Technical field
The invention belongs to computer memory technical fields, more particularly, to entangling based on data buffer storage under a kind of cluster
Delete a yard archiving method.
Background technology
Rapidly increase in information age data volume of today, large-scale storage cluster is more and more widely used.
According to the height of data access frequency, can the data in storage cluster be divided into dsc data (Hot Data), warm data successively
(Warm Data) and cold data (Cold Data).With the passage of system operation time, the visiting frequency of dsc data can decline,
Become warm data, finally become cold data, dsc data is usually stored in production cluster by three copies or more copy modes
(Production Cluster).It is a kind of improve production group system utilization rate method be by the low copy data of visiting frequency
Correcting and eleting codes filing cluster (Archival Cluster) is moved to, which stores number therein using RS codes come tissue
According to；Correcting and eleting codes filing (Erasure-coded Archival) refers to that data are moved to the behaviour that correcting and eleting codes store from copy storage
Make.
It is two mutually independent processes that existing correcting and eleting codes filing and user, which access copy cluster,；Existing correcting and eleting codes filing packet
It includes (1) and obtains and need to file data bitmap of the data in producing cluster；(2) copy data is read from production cluster；(3) will
It reads copy data and carries out correcting and eleting codes coding, generate verification piecemeal；(4) a copy data and corresponding verification piecemeal write-in are entangled
Delete code filing cluster；(5) remaining cold data on production cluster is recycled, deletes all participation correcting and eleting codes in production cluster and returns
The data copy of shelves.It includes metadata information of the deblocking in producing cluster needed for (1) obtains that user, which accesses copy cluster,；
(2) read request is sent out to node where deblocking；(3) node reads corresponding data and is chunked into node buffering area；(4) data
Node transmission data piecemeal, and deblocking is forwarded to client.
The correcting and eleting codes are filed, if the request of data filing is excessive, using offline filing, data access and filing two
Person is independent；But in the current big data epoch, special time of filing is there's almost no, online elite archiving need to be used；For returning online
Shelves produce and have remained dsc data and cold data on the node in cluster, are existed using above-mentioned correcting and eleting codes archiving method as follows
Problem：
(1) Archive request is directly handed down to production cluster, produces cluster at this time and is also providing the service of access to the user；Filing
Request bring additional external memory to access, the occupancy of memory cache, the occupancy of Internet resources, cause user request response time it is big
It is big to increase；
(2) for filing task, return with data due to producing the node needs in cluster while handling user access request
, there is resource contention between the two in shelves request；In the case of some storage cluster node users access request frequency height, the node
Archive request cannot will slowly respond, cause archival one that can not obtain deblocking in time, and then influence entire filing
Efficiency.
Invention content
For the disadvantages described above or Improvement requirement of the prior art, the present invention provides under a kind of cluster based on data buffer storage
Correcting and eleting codes archiving method, its object is to improve correcting and eleting codes to file efficiency.
To achieve the above object, according to one aspect of the present invention, entangling based on data buffer storage under a kind of cluster is provided
A yard archiving method is deleted, following steps are specifically included：
(1) according to user access request, by required deblocking, production node where it reads node buffering area, and
Deblocking is sent to accessed node, while archival one is forwarded it to according to filing data mode table；
(2) archival one updates the deblocking number of each back end according to the deblocking received；And judge
Whether archival one receives the total data piecemeal of current band, if so, current band is filed；If it is not, then from
It produces cluster and reads cold data block, file after band is gathered together enough.
Above-mentioned correcting and eleting codes archiving method, as a result of the deblocking of user's access cache, compared to common correcting and eleting codes
The number that archival one reads deblocking from production cluster can be greatly reduced, so as to have substantially no effect in filing
User carries out archive operation in the case of accessing, improve filing efficiency；
On the other hand, the deblocking of user's access cache is forwarded to archival one, avoids user and accesses and common
Correcting and eleting codes filing can efficiently reduce network in production cluster and pass to the repetition read requests of identical data piecemeal in production node
Defeated data volume mitigates the load of whole network.
Preferably, above-mentioned steps (1) specifically includes following sub-step：
(1.1) accessed node obtains address of the required deblocking in producing cluster from management server；
(1.2) read request is sent out to address corresponding node of the above-mentioned deblocking in generating cluster；
(1.3) deblocking is generated into the production node in cluster where it and reads node buffering area；
(1.4) according to filing data mode table, judge whether the deblocking has been forwarded to archival one, if so,
The deblocking is forwarded to accessed node；If it is not, the deblocking is then forwarded to accessed node and archival one simultaneously；It should
Step directly utilizes the data forwarding that the interchanger in cluster accesses user to archival one.
Preferably, above-mentioned steps (2) specifically include following sub-step：
(2.1) archival one receives deblocking (dsc data), updates each back end for including in archival one
Deblocking number；
(2.2) archival one initializes current band, analyzes the node composition of band, judges which the data block received belongs to
A back end, and initialize storage variable；In this step, node includes back end and check-node；
(2.3) the total data piecemeal for judging whether to receive current band by archival one, if so, entering step
(2.7)；If it is not, then entering step (2.4)；
(2.4) when a length of T time interval in, archival one receives deblocking from user cache；
(2.5) judge whether the k number of current band collects neat according to piecemeal, if so, entering step (2.7)；If it is not, then entering
Step (2.6)；Wherein, k values are equal to the number of back end；
(2.6) archival one directly reads cold data from production cluster, with the μ band that gather together enough, subsequently into step
(2.3)；
Wherein, the deblocking number read from back end j is Numj, Numj=max { (μ-aj- λ (j)), 0 }, μ is to return
The maximum band number that shelves node can be filed in time T, ajFor the deblocking comprising back end j in archival one
Number, λ (j) indicate the deblocking number that archival one will be received from back end j in next time interval T；
(2.7) file current band, and a value of reel number i is added 1, and update filing data mode table；
(2.8) filing for judging whether to complete whole bands according to filing data mode table, if it is not, then entering step
(2.3)；If so, terminating filing；
The above process, archival one, using data cached reception user's piecemeal, can rise while not influencing user's access
To the effect for improving filing efficiency.
In general, through the invention it is contemplated above technical scheme is compared with the prior art, can obtain down and show
Beneficial effect：
(1) the correcting and eleting codes archiving method based on data buffer storage under this cluster provided by the invention is visited not influencing user
In the case of asking, data cached filing efficiency is improved using what user accessed：Pass through the data cached piecemeal for accessing user
It is transmitted directly to archival one, compared to common correcting and eleting codes archiving method, method provided by the invention can reduce disk reading
Number reads to accelerate to produce node external memory deblocking, production node memory space hold and production node network interface card are sent out
It send filing deblocking these three links, improves filing efficiency；Only in the case where a band does not receive all, just meeting
Production company-data piecemeal is directly read, and reads the number of deblocking and data volume and frequency that block number is accessed compared to user
Spend it is much lower, have substantially no effect on user access；
(2) the correcting and eleting codes archiving method based on data buffer storage under this cluster provided by the invention, reduces production cluster
The data volume of middle network transmission；In existing common correcting and eleting codes filing, need deblocking being sent to from node buffering area
Interchanger in cluster, relays to archival one；And in the correcting and eleting codes archiving method of the offer in the present invention, it can be directly sharp
To archival one, avoiding will count in common correcting and eleting codes archiving method the data forwarding for being accessed user with the interchanger in cluster
It is sent to the interchanger in cluster network according to from node buffering area, so as to efficiently reduce network transmission in production cluster
Data volume, mitigate the load of whole network.
Description of the drawings
Fig. 1 is the flow chart of the correcting and eleting codes archiving method based on data buffer storage under cluster of the present invention；
Fig. 2 is the schematic diagram of the correcting and eleting codes archiving method based on data buffer storage under the cluster that embodiment provides；
Fig. 3 is the schematic diagram that user accesses and common correcting and eleting codes are filed；
Fig. 4 is the schematic diagram of user access and data filing based on data buffer storage.
Specific implementation mode
In order to make the purpose , technical scheme and advantage of the present invention be clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.As long as in addition, technical characteristic involved in the various embodiments of the present invention described below
It does not constitute a conflict with each other and can be combined with each other.
For ease of understanding, the term being related in the present invention is illustrated as follows：
Produce cluster：While the offer filing data in archiving process are provided, it is still desirable to upper layer application be given to provide data
Access the node general name of service.
Correcting and eleting codes file cluster：The cluster that partial data band is stored after the completion of filing, can be by production cluster itself
Free space substitutes.
Management server：The metadata information of the whole cluster of management, including the node of deblocking and its copy in cluster
Location information.
Coding nodes：During the data filing of one band, implement the node of coding calculating operation.
Node data piecemeal and band：In the cataloged procedure of filing, the unit for reading data is exactly deblocking.
In storage cluster, a band is exactly to be made of the deblocking unit of same offset address on different nodes, can be independent
Ground recovers the information aggregate of fail data.
Data bitmap：It is stored in the management server of storage cluster, indicates pair between deblocking and back end
It should be related to.
NUMj：In archiving process, when the current deblocking filed needed for band does not collect neat, archival one directly to
It produces back end j in cluster and reads NUMj=max { (μ-aj- λ (j)), 0 } a deblocking, μ is archival one in time T
The band number that can file, ajTo include the deblocking number of back end j in archival one, λ (j) is indicated at next section
Between archival one is received from node j in T deblocking number.
File data mode table：The global table that management server is safeguarded, record produce deblocking in cluster
Whether it has been archived, in embodiment, the deblocking ' not being archived ' and ' being archived ' has been indicated respectively with ' 0 ' and ' 1 '.
The present invention provides the correcting and eleting codes archiving methods based on data buffer storage under a kind of cluster, main that user is utilized to access
Data cached to accelerate correcting and eleting codes archiving process, detailed process is as shown in Figure 1, include the following steps：
(1) according to user access request, by required deblocking, production node where it reads node buffering area, then
Deblocking is sent to accessed node, while archival one is forwarded it to according to filing data mode table；This step is specific
Including following sub-step：
(1.1) accessed node is asked according to user, searching and managing server, and the deblocking for obtaining user's request is producing
Node location in cluster；
(1.2) the production clustered node position obtained according to step (1.1), the node reading deblocking for producing cluster arrive
Node buffering area；
(1.3) querying node management server is produced, filing data mode table is obtained；According to filing data mode table, sentence
The filing state of disconnected user's request data piecemeal；
(1.4) production node sends deblocking in buffering area；
(1.5) deblocking is sent to accessed node by interchanger, is asked with relative users, meanwhile, by what is be not archived
Deblocking is forwarded to archival one, with corresponding Archive request；
(1.6) judge that whether all user's request complete by processing, terminates if it has, then user accesses；If it has not, then going to
Step (1.1) continues to execute next user's request；
(2) archival one updates the deblocking number of each back end according to the user accesses data received,
If archival one receives the total data piecemeal of current band, directly filed；If can not by user accesses data
Current band total data piecemeal is received, then reads cold data from production cluster, is then filed；This step specifically includes
Following sub-step：
(2.1) archival one initialization filing band, current band number of filing is i, i=0；It initializes in archival one
Include the deblocking number a of back end jjSet a be [0,0,0,0]；
(2.2) archival one receives user accesses data (dsc data), and with newly set a；
(2.3) archival one judges whether to receive the total data piecemeal of filing band i, if so, entering step
(2.7)；If it is not, then entering step (2.4)；
(2.4) archival one latency period T；In the time interval of duration T, archival one is constantly from user cache
Receive deblocking；In the present embodiment, T is 1 second；
(2.5) archival one judges whether the total data piecemeal for receiving filing band i again with newly set a, if
It is to enter step (2.7)；If it is not, then entering step (2.6)；
(2.6) archival one directly reads cold data from production cluster, and the deblocking number read from back end j is
Numj, Numj=max { (μ-aj- λ (j)), 0 }, with the μ band that gather together enough, subsequently into step (2.3)；
(2.7) file current band, and a reel number i values are added 1；
(2.8) the filing data mode table in management server is updated；
(2.9) filing for judging whether whole bands in completion production cluster, if it is not, then entering step (2.3)；If so,
Then terminate to file.
The correcting and eleting codes archiving method provided by the invention based on data buffer storage is illustrated below in conjunction with specific embodiment；In example
Using (k+r, k) RS coding modes；Wherein, k refers to the number of initial data piecemeal, the verification piecemeal that r presentation codes generate
Number, k number can decode k initial data piecemeal according to arbitrary k in piecemeal and r verification piecemeal.
As shown in Fig. 2, being to use the correcting and eleting codes archiving method based on data buffer storage under RS (6,4) code clustering in embodiment
Exemplary plot；Wherein, there are six nodes in total in correcting and eleting codes filing cluster, wherein four are back end, two are to examine section
Data integrity is realized and ensured to point using the coding mode of (6,4) RS, and node serial number is { 1,2,3,4,5,6 }；
Wherein, the logical address for coming from any data piecemeal in production cluster both corresponds in correcting and eleting codes cluster four
One of in back end, archival one files band number μ=3 in time T.
The correcting and eleting codes archiving method based on data buffer storage, specific as follows under cluster provided in this embodiment：
(1) in embodiment, the deblocking number set that first time user accesses is [2,4,5,3], that is, from back end
1, back end 2, back end 3 and back end 4 obtain 2,4,5 and 3 deblockings respectively；
Accessed node searching and managing server first, obtains whole address informations of the deblocking set；Meanwhile filing
Node initializing band i,；At this point, the deblocking number of each back end is [0,0,0,0] in archival one；Wherein, i
Initial value is 0；
(2) production cluster reads deblocking set to buffering area；When accessing for the first time, interchanger is by deblocking set
It is forwarded to accessed node and archival one simultaneously；
(3) accessed node obtains deblocking set, response user's request；Meanwhile archival one obtains deblocking collection
It closes, the deblocking number for updating each back end is [2,4,5,3]；
1. flow indicated in above-mentioned steps (1) (2) (3) corresponding diagram 2；
(4) archival one carries out band i coding calculating, and 4 deblockings and 2 verification piecemeals are sent to correcting and eleting codes and are returned
Shelves cluster, i increase by 1, and i is 1 at this time；Archival one removes the deblocking for having filed band 0；
(5) if archival one receives all deblockings of band i, filed；And by 4 deblockings and 2
Verification piecemeal is sent to correcting and eleting codes filing cluster, and i increases by 1, and i is 2 at this time；
Delete the deblocking for the band 1 filed in archival one；Each node data piecemeal number in archival one
For [0,2,3,1]；
2. flow indicated in above-mentioned steps (4) (5) corresponding diagram 2；
(6) if not receiving the total data piecemeal of band i in archival one, enter wait state；Meanwhile user carries out
Back-call, deblocking collection are combined into [0,3,3,4], it is assumed that are not had in deblocking set and accessed repetition for the first time
Deblocking, then interchanger by deblocking set simultaneously be forwarded to accessed node and archival one；
(7) accessed node relative users back-call；Meanwhile archival one receives deblocking set [0,3,3,4],
It is [0,5,6,5] to update node data piecemeal number；
3. flow indicated in above-mentioned steps (6) (7) corresponding diagram 2；
(8) after archival one waiting period T, if the total data piecemeal of band i is not received yet, directly from production
Deblocking is read in cluster；
In next period T, the deblocking λ that user accesses is [1,1,1,1], according to NUMj=max { (μ-aj-λ
(j)) it is [2,0,0,0], 0 }, to obtain the deblocking number read from four back end；It needs to read two from back end 1
A corresponding deblocking；Wherein, band number μ=3 that archival one can be filed in time T；Include number in archival one
According to the deblocking number a of node jjSet a be [0,5,6,5]；
(9) third time user accesses data block collection is [1,1,2,1], and interchanger forwards deblocking set simultaneously
To accessed node and archival one；Meanwhile archival one is pressed deblocking set [2,0,0,0] and is directly read from production cluster
Deblocking；
(10) archival one receives the deblocking set [1,1,2,1] from user cache and the data directly read point
Set of blocks [2,0,0,0], update node data piecemeal number are [3,6,8,6]；
4. flow indicated in above-mentioned steps (8) (9) (10) corresponding diagram 2；
(11) archival one completes the filing calculating of 3 bands, and the deblocking of each band and verification piecemeal are sent out
It is sent to correcting and eleting codes filing cluster, deletes the deblocking filed in archival one；I is 5 at this time, each node in archival one
The number of deblocking is [0,3,5,3]；
5. flow indicated in above-mentioned steps (11) corresponding diagram 2；
(12) step (1)~(11) are repeated, the filing until producing whole bands in cluster is completed.
In the prior art, the process of common correcting and eleting codes filing is as shown in figure 3, be specially：Archival one directly collects from production
The k number needed for band is read in group according to piecemeal, r verification piecemeal of the band is then obtained, finally by k number according to piecemeal
Correcting and eleting codes filing cluster is sent to r verification piecemeal to preserve；Filing is all to read deblocking again from production cluster every time
Verification piecemeal is calculated, deblocking and verification piecemeal, which are then sent to correcting and eleting codes, files cluster, until filing is fully completed；
And the correcting and eleting codes archiving method based on data buffer storage that the invention of this hair provides, it is specific as follows：
(1) archival one initializes band, receives the deblocking of user's access cache；
(2) if currently the k number of filing band is all received according to piecemeal, it is put into step (4)；It otherwise waits for next
The data sent from the data buffer storage that user accesses in a period T；
(3) the total data piecemeal for judging whether to receive current band, if so, entering step (4)；If it has not, then
Corresponding deblocking is read with the μ band that gather together enough from production cluster；
(4) archival one carries out coding calculating to current band, and k number is sent to and is entangled according to piecemeal and r and verification piecemeal
It deletes and is preserved in code filing cluster；
(5) it repeats step (2) and arrives (4), the filing until producing whole bands in cluster is completed.
In embodiment, using the deblocking of reception user's access cache in step (1), return compared to common correcting and eleting codes
Shelves drastically reduce the number that archival one reads deblocking from production cluster, so as to have substantially no effect on use
Family carries out archive operation in the case of accessing, improve filing efficiency；On the other hand, by interchanger by the number of user's access cache
It is forwarded to archival one according to piecemeal, user is avoided and accesses with the filing of common correcting and eleting codes to identical data piecemeal in production node
Read requests are repeated, the data volume of network transmission in production cluster can be efficiently reduced, mitigate the load of whole network.
As it will be easily appreciated by one skilled in the art that the foregoing is merely illustrative of the preferred embodiments of the present invention, not to
The limitation present invention, all within the spirits and principles of the present invention made by all any modification, equivalent and improvement etc., should all include
Within protection scope of the present invention.
Claims (3)
1. the correcting and eleting codes archiving method based on data buffer storage under a kind of cluster, which is characterized in that the correcting and eleting codes archiving method tool
Body includes the following steps：
(1) according to user access request, required deblocking is read into node buffering area from production node, and by deblocking
It is sent to accessed node, while the deblocking is transmitted to by archival one according to filing data mode table；
It is described to be by the method that deblocking is transmitted to archival one according to filing data mode table：
Judge whether the deblocking has been forwarded to archival one, if so, the deblocking is forwarded to accessed node；
If it is not, the deblocking is then forwarded to accessed node and archival one simultaneously；
(2) archival one updates the deblocking number of each back end according to the deblocking received；And judge to file
Whether node receives the total data piecemeal of current band, if so, current band is filed；If it is not, then from production
Cluster reads cold data block, files after band is gathered together enough.
2. correcting and eleting codes archiving method as described in claim 1, which is characterized in that accessed and asked according to user in the step (1)
Ask the method that required deblocking is read into node buffering area from production node, including following sub-step：
(1.1) accessed node obtains address of the required deblocking in producing cluster from management server；
(1.2) read request is sent out to address corresponding node of the deblocking in generating cluster；
(1.3) deblocking is generated into the production node in cluster where it and reads node buffering area.
3. correcting and eleting codes archiving method as claimed in claim 1 or 2, which is characterized in that the step (2) specifically includes following son
Step：
(2.1) archival one receives deblocking, updates the deblocking number for each back end for including in archival one；
(2.2) archival one initializes current band, analyzes the node composition of band, judges which number is the data block received belong to
According to node, and initialize storage variable；The node composition of the band includes back end and check-node；
(2.3) the total data piecemeal for judging whether to receive current band by archival one, if so, entering step (2.7)；If
It is no, then it enters step (2.4)；
(2.4) when a length of T time interval in, archival one receives deblocking from user cache；
(2.5) judge whether the k number of current band collects neat according to piecemeal, if so, entering step (2.7)；If it is not, then entering step
(2.6)；Wherein, k values are equal to the number of back end；
(2.6) archival one directly reads cold data from production cluster, with the μ band that gather together enough, subsequently into step (2.3)；
Wherein, the maximum band number that μ can file for archival one in time interval T；
(2.7) file current band, and a value of reel number i is added 1, and update filing data mode table；
(2.8) filing for judging whether to complete whole bands according to filing data mode table, if it is not, then entering step (2.3)；If
It is then to terminate to file.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510979326.8A CN105610921B (en) | 2015-12-23 | 2015-12-23 | Correcting and eleting codes archiving method based on data buffer storage under a kind of cluster |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510979326.8A CN105610921B (en) | 2015-12-23 | 2015-12-23 | Correcting and eleting codes archiving method based on data buffer storage under a kind of cluster |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105610921A CN105610921A (en) | 2016-05-25 |
| CN105610921B true CN105610921B (en) | 2018-09-07 |
Family
ID=55990455
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510979326.8A Expired - Fee Related CN105610921B (en) | 2015-12-23 | 2015-12-23 | Correcting and eleting codes archiving method based on data buffer storage under a kind of cluster |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105610921B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108304264B (en) * | 2018-01-22 | 2022-02-15 | 华中科技大学 | Erasure code filing method based on SPARK streaming calculation |
| CN110190926B (en) * | 2019-04-26 | 2020-09-18 | 华中科技大学 | Erasure code updating method and system based on network computing |
| CN112764680B (en) * | 2021-01-20 | 2024-02-06 | 华云数据控股集团有限公司 | Disk stripe selection method and storage system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103177111A (en) * | 2013-03-29 | 2013-06-26 | 西安理工大学 | System and method for deleting repeating data |
| CN103605582A (en) * | 2013-11-27 | 2014-02-26 | 华中科技大学 | Erasure code storage and reconfiguration optimization method based on redirect-on-write |
-
2015
- 2015-12-23 CN CN201510979326.8A patent/CN105610921B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103177111A (en) * | 2013-03-29 | 2013-06-26 | 西安理工大学 | System and method for deleting repeating data |
| CN103605582A (en) * | 2013-11-27 | 2014-02-26 | 华中科技大学 | Erasure code storage and reconfiguration optimization method based on redirect-on-write |
Non-Patent Citations (2)
| Title |
|---|
| CoRec: A Cooperative Reconstruction Pattern for Multiple Failures in Erasure-Coded Storage Clusters;Jianzhong Huang等;《IEEE》;20150904;全文 * |
| 面向纠删码存储集群的弹性Ｉ/Ｏ调度机制研究;黄建忠等;《计算机研究与发展》;20141215;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105610921A (en) | 2016-05-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11023448B2 (en) | Data scrubbing method and apparatus, and computer readable storage medium | |
| CN110290217B (en) | Data request processing method and device, storage medium and electronic device | |
| CN104272274B (en) | Data processing method and equipment in a kind of distributed file storage system | |
| CN111291079A (en) | Data query method and device | |
| CN104346458B (en) | Date storage method and storage device | |
| CN105610921B (en) | Correcting and eleting codes archiving method based on data buffer storage under a kind of cluster | |
| CN103186537B (en) | A kind of data read-write method and device based on index order | |
| CN105868231A (en) | Cache data updating method and device | |
| CN106649349A (en) | Method, device and system for data caching, applicable to game application | |
| CN104125163A (en) | Data processing method, device and terminal | |
| US20220035786A1 (en) | Distributed database management system with dynamically split b-tree indexes | |
| CN107368608A (en) | The HDFS small documents buffer memory management methods of algorithm are replaced based on ARC | |
| CN109462631A (en) | Data processing method, device, storage medium and electronic device | |
| CN107872473A (en) | Message treatment method, device and system | |
| CN110399348A (en) | File deletes method, apparatus, system and computer readable storage medium again | |
| CN105868251A (en) | Cache data updating method and device | |
| CN110199512A (en) | The management method and device of equipment are stored in storage system | |
| CN109597903A (en) | Image file processing apparatus and method, document storage system and storage medium | |
| CN110502607A (en) | A kind of electronic medical record system, the method and server for inquiring electronic health record | |
| CN105450513B (en) | File the method and cloud storage service device of Email attachment | |
| CN113918857A (en) | Three-level cache acceleration method for improving performance of distributed WEB application system | |
| CN103825922B (en) | A kind of data-updating method and web server | |
| CN105915595A (en) | Cluster storage system data accessing method and cluster storage system | |
| CN106899558A (en) | The treating method and apparatus of access request | |
| CN106980618A (en) | File memory method and system based on MongoDB distributed type assemblies frameworks |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180907 Termination date: 20191223 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |