CN103731505A - Data distributed storage method and system - Google Patents
Data distributed storage method and system Download PDFInfo
- Publication number
- CN103731505A CN103731505A CN201410022973.5A CN201410022973A CN103731505A CN 103731505 A CN103731505 A CN 103731505A CN 201410022973 A CN201410022973 A CN 201410022973A CN 103731505 A CN103731505 A CN 103731505A
- Authority
- CN
- China
- Prior art keywords
- memory node
- performance parameter
- data
- memory
- priority
- 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
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a data distributed storage method and system which includes acquiring performance parameters of all storage nodes, determining the priority of the storage nodes according to the performance parameters of the storage nodes and the corresponding preset weights, and storing data blocks finishing data block division and coding into the storage nodes with high priority. The method and system is used for collecting various performance parameters of the storage nodes to serve as the basis of follow-up data block storage, completely considers the performance of the storage nodes and well guarantees data reliability.
Description
Technical field
The present invention relates to and environmental data memory technology, espespecially a kind of based on error correcting code (Erasure Code) coded data distributed storage method and system.
Background technology
Along with the development of cloud environment data center technology, increasing service application all depends on the flexible resource enabling capabilities that cloud environment provides, and realizes the deployment of its applied logic and the storage of business datum.In order to improve the reliability of data, data backup becomes one of matter of utmost importance that cloud data center must solve.Traditional data backup is that the complete copy of data is copied on different memory nodes, thereby improves the reliability of data, and still, this method need to take a large amount of memory spaces, is unfavorable for the reasonable distribution of storage resources.In addition, complete data Replica process also needs to consume the long period.
Based on Erasure Code coded data distributed storage mode, effectively alleviated the space waste that partial data copies generation, yet, data based on Erasure Code coded data distributed storage mode distribute at present, only depend on utilization rate of equipment and installations as the reference frame of data storage.
At present, based on Erasure Code coded data distributed storage mode, roughly comprise: first, data file is divided into n data block; Then, by the mode of encoding, n original data block is encoded to (n+m) individual data block (wherein, m is the quantity of checking data piece), like this, only needs just can recover complete initial data by the data block after any n coding; Finally, by (n+m) the individual data block after coding is distributed and is stored on different memory nodes, thereby avoided the insecure problem of data that causes due to tables of equipment point failure.
Conventionally Erasure Code data distributed storage mode adopts the mode of random selection or relies on the resource utilization of memory node as the decision-making foundation of data distribution, although this mode can effectively realize the mean allocation of resource utilization.But, performance from memory device, the multidimensional attributes such as resource utilization, input and output (IO) load and network delay all can affect the performance of memory device, and different service application types may be embodied in different aspect (as accessed mutual delay, high concurrent reading and writing etc.) to the performance requirement focus of data storage.That is to say, the existing properties that does not take into full account memory node based on Erasure Code coded data distributed storage mode, thus there is the risk that can not guarantee data reliability.
Summary of the invention
In order to solve the problems of the technologies described above, the invention provides a kind of data distributed storage method and system, take into full account the performance of memory node, can guarantee well the reliability of data.
In order to reach the object of the invention, the invention provides a kind of data distributed storage method, comprising: the performance parameter of obtaining all memory nodes;
According to the performance parameter of memory node and the corresponding weight that sets in advance, determine the priority of memory node;
To complete deblocking and coded data piece is stored in the memory node that priority is high.
Described performance parameter comprises infrastructure resources service condition, input and output IO load and the network delay of described memory node equipment.
The infrastructure resources service condition of described memory node equipment comprises cpu busy percentage, memory space utilance.
The priority of described definite memory node comprises: by described performance parameter normalization;
According to the weight coefficient of normalization result and each performance parameter of setting in advance, calculate respectively the weighted priority of each memory node.
Describedly complete deblocking and coded data piece is:
(n+m) blocks of data piece based on obtaining after error correcting code Erasure Code mechanism coding;
Wherein, n represents the piece number of service application data block; M represents the piece number of checking data piece.
The invention also discloses a kind of data distributed memory system, at least comprise a plurality of memory nodes and control node; Wherein,
Memory node, for storing data block;
Control node, for gathering the performance parameter of each memory node; According to the performance parameter of each memory node and the weight that sets in advance, determine the priority of each memory node; To complete deblocking and coded data piece is stored in the memory node that priority is high.
Described control node comprises acquisition module, processing module, wherein,
Acquisition module, for gathering the performance parameter of each memory node;
Processing module, sets in advance a weight corresponding to performance parameter, for according to the performance parameter of each memory node and corresponding weight, determines the priority of each memory node; To complete deblocking and coded data piece is stored in the memory node that priority is high.
Compared with prior art, the present invention includes the performance parameter of obtaining all memory nodes; According to the performance parameter of memory node and the corresponding weight that sets in advance, determine the priority of memory node; To complete deblocking and coded data piece is stored in the memory node that priority is high.The present invention gathers by the multinomial performance parameter to memory node, as the foundation of subsequent data blocks storage, has taken into full account the performance of memory node, has guaranteed well the reliability of data.
Other features and advantages of the present invention will be set forth in the following description, and, partly from specification, become apparent, or understand by implementing the present invention.Object of the present invention and other advantages can be realized and be obtained by specifically noted structure in specification, claims and accompanying drawing.
Accompanying drawing explanation
Accompanying drawing is used to provide the further understanding to technical solution of the present invention, and forms a part for specification, is used from explanation technical scheme of the present invention with the application's embodiment mono-, does not form the restriction to technical solution of the present invention.
Fig. 1 is the flow chart of data distributed storage method of the present invention;
Fig. 2 is the composition structural representation of data distributed memory system of the present invention.
Embodiment
For making the object, technical solutions and advantages of the present invention clearer, hereinafter in connection with accompanying drawing, embodiments of the invention are elaborated.It should be noted that, in the situation that not conflicting, the embodiment in the application and the feature in embodiment be combination in any mutually.
In the step shown in the flow chart of accompanying drawing, can in the computer system such as one group of computer executable instructions, carry out.And, although there is shown logical order in flow process, in some cases, can carry out shown or described step with the order being different from herein.
Fig. 1 is the flow chart of data distributed storage method of the present invention, as shown in Figure 2, comprising:
Step 100: the performance parameter of obtaining all memory nodes.
In this step, the performance parameter of memory node is Real-time Collection, the performance parameter collecting includes but not limited to the infrastructure resources service condition (as cpu busy percentage, memory space utilance etc.) of memory node equipment, and I/O (I/O) load and network delay etc.
The conventional techniques means that the collection of the performance parameter of memory node belonged to those skilled in the art, repeat no more here.What this step was emphasized is that the multinomial performance parameter of memory node is gathered, and as the foundation of subsequent data blocks storage, rather than picture resource utilization of only usining memory node of the prior art is as the foundation of distributed data storage.
Step 101: according to the performance parameter of memory node and the corresponding weight that sets in advance, determine the priority of memory node.
In this step, weight should reflect the demand of sector application to data access.Such as: for the application of deflection analysis type, its data characteristics normally write-once repeatedly reads, and has stronger concurrency, therefore simultaneously, to having relatively high expectations of I/O load, network delay and the bandwidth etc. of memory node, the weight of these performance parameters is just large; For another example: for the application of the deflection course of processing, often need on resource node, carry out simultaneously and calculate and read-write operation, preserve the intermediate data result of the course of processing simultaneously, therefore, require node to possess more CPU computing capability and memory space, that is to say, for the weight of the performance parameters such as cpu busy percentage, memory space utilance, want large.
Suppose to be provided with i memory node in cloud environment, first, the i that cloud environment an is provided memory node is converted into the candidate collection { A consisting of i kind discrete solution
i, meanwhile, the j dimension performance parameter that affects storage device performance is mapped as to the j kind attribute of discrete solution; Correspondingly, each memory node respectively tie up the assignment that resource using status (be performance parameter show state) just becomes the various attributes of each candidate scheme, input matrix D as shown in formula (1),
D=(x
i,j)
n*m， (1)
In formula (1), x
i,jrepresent candidate collection { A
iin candidate scheme A
iat dimension X
jon assignment, i.e. the use state q of corresponding stored node i in j dimension
i,j.N represents that service application data are divided into the quantity of data block, and m represents the quantity of checking data piece.
In this step, determine that the priority of memory node comprises:
First, by each performance parameter normalization.
Because each dimension resource has different unit of measurement, such as: the unit of space utilisation is that the unit of %, I/O load is Mbps etc., can adopt range transformation as shown in formula (2), matrix is carried out to standardization, and re-using regularization trans formation, as formula (3), to eliminate metering particle size differences be normalization on the impact of result.
Wherein,
represent matrix r
ijj row in maximum numerical value,
represent matrix r
ijj row in minimum numerical value, x
i,jrepresent candidate collection { A
iin candidate scheme A
iat dimension X
jon assignment, i.e. the use state q of corresponding stored node i in j dimension
i,j.
Then, set in advance the weight coefficient w of each performance parameter
j, according to above-mentioned weight, should reflect the demand of sector application to data access, can use existing expert's method, by system or user, according to actual service condition or protracted experience, set in advance each weight coefficient.The normal standard matrix r generating according to formula (3)
ijand each weight coefficient w
j, calculate respectively the weighted priority of each memory node, i.e. priority
i∑
jr
ij* w
j.
Step 102: will complete deblocking and coded data piece is stored in the memory node that priority is high.
In this step, complete deblocking and coded data piece refers to, service application data are divided into n piece, and based on Erasure Code mechanism, this n blocks of data are encoded into (n+m) blocks of data piece, newly-generated m blocks of data piece is checking data piece.Wherein, n and m are the value setting in advance, and can arrange according to existing experimental result and construction experiences, such as value n=6, m=2.It should be noted that, in real system, can suitably increase according to the degree of scatter of data the value of n, according to data reliability demand and honor demand, suitably adjust the value of m, how many concrete adjustment belongs to those skilled in the art's conventional techniques means, repeats no more here.The implementation of conventional Erasure Code has the encoding mechanisms such as Reed-Solomon Code and Tonardo Code; specific implementation belongs to those skilled in the art's conventional techniques means; here repeat no more, its specific implementation is also not intended to limit the scope of the invention.
This step, stores (n+m) individual deblocking of each service application in (n+m) platform memory node that priority is high into.
Fig. 2 is the composition structural representation of data distributed memory system of the present invention, as shown in Figure 1, at least comprises a plurality of memory nodes and controls node; Wherein,
Memory node, for storing data block.
Control node, for gathering the performance parameter of each memory node; According to the performance parameter of each memory node and the weight that sets in advance, determine the priority of each memory node; To complete deblocking and coded data piece is stored in the memory node that priority is high.
Control node in the present invention can comprise acquisition module, processing module, wherein,
Acquisition module, for gathering the performance parameter of each memory node;
Processing module, sets in advance a weight corresponding to performance parameter, for according to the performance parameter of each memory node and corresponding weight, determines the priority of each memory node; To complete deblocking and coded data piece is stored in the memory node that priority is high.
Although the disclosed execution mode of the present invention as above, the execution mode that described content only adopts for ease of understanding the present invention, not in order to limit the present invention.Those of skill in the art under any the present invention; do not departing under the prerequisite of the disclosed spirit and scope of the present invention; can in the form of implementing and details, carry out any modification and variation; but scope of patent protection of the present invention, still must be as the criterion with the scope that appending claims was defined.
Claims (7)
1. a data distributed storage method, is characterized in that, comprising: the performance parameter of obtaining all memory nodes;
According to the performance parameter of memory node and the corresponding weight that sets in advance, determine the priority of memory node;
To complete deblocking and coded data piece is stored in the memory node that priority is high.
2. data distributed storage method according to claim 1, is characterized in that, described performance parameter comprises infrastructure resources service condition, input and output IO load and the network delay of described memory node equipment.
3. data distributed storage method according to claim 2, is characterized in that, the infrastructure resources service condition of described memory node equipment comprises cpu busy percentage, memory space utilance.
4. according to the data distributed storage method described in claim 2 or 3, it is characterized in that, the priority of described definite memory node comprises: by described performance parameter normalization;
According to the weight coefficient of normalization result and each performance parameter of setting in advance, calculate respectively the weighted priority of each memory node.
5. specified number according to claim 1, according to distributed storage method, is characterized in that, described in complete deblocking and coded data piece is:
(n+m) blocks of data piece based on obtaining after error correcting code Erasure Code mechanism coding;
Wherein, n represents the piece number of service application data block; M represents the piece number of checking data piece.
6. a data distributed memory system, is characterized in that, at least comprises a plurality of memory nodes and controls node; Wherein,
Memory node, for storing data block;
Control node, for gathering the performance parameter of each memory node; According to the performance parameter of each memory node and the weight that sets in advance, determine the priority of each memory node; To complete deblocking and coded data piece is stored in the memory node that priority is high.
7. data distributed memory system according to claim 6, is characterized in that, described control node comprises acquisition module, processing module, wherein,
Acquisition module, for gathering the performance parameter of each memory node;
Processing module, sets in advance a weight corresponding to performance parameter, for according to the performance parameter of each memory node and corresponding weight, determines the priority of each memory node; To complete deblocking and coded data piece is stored in the memory node that priority is high.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410022973.5A CN103731505A (en) | 2014-01-17 | 2014-01-17 | Data distributed storage method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410022973.5A CN103731505A (en) | 2014-01-17 | 2014-01-17 | Data distributed storage method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103731505A true CN103731505A (en) | 2014-04-16 |
Family
ID=50455427
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410022973.5A Pending CN103731505A (en) | 2014-01-17 | 2014-01-17 | Data distributed storage method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103731505A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105554149A (en) * | 2015-12-31 | 2016-05-04 | 康佳集团股份有限公司 | Video cloud storage load equalizing method and system based on video ranks |
| CN105975345A (en) * | 2016-05-20 | 2016-09-28 | 江苏得得空间信息科技有限公司 | Video frame data dynamic equilibrium memory management method based on distributed memory |
| CN106156317A (en) * | 2016-06-30 | 2016-11-23 | 电子科技大学 | A kind of secure storage method of data based on Attribute transposition |
| CN106230982A (en) * | 2016-09-08 | 2016-12-14 | 哈尔滨工程大学 | A kind of dynamic self-adapting secure cloud storage method considering node reliability |
| CN106933492A (en) * | 2015-12-30 | 2017-07-07 | 伊姆西公司 | The method and apparatus for contributing to the abrasion equilibration of solid state hard disc |
| CN108241552A (en) * | 2016-12-23 | 2018-07-03 | 航天星图科技（北京）有限公司 | A kind of client file restoration methods |
| CN108875035A (en) * | 2018-06-25 | 2018-11-23 | 郑州云海信息技术有限公司 | The date storage method and relevant device of distributed file system |
| CN109460426A (en) * | 2018-11-05 | 2019-03-12 | 郑州云海信息技术有限公司 | A kind of system and method, the routing node of the classification storage based on MongoDB |
| WO2020001287A1 (en) * | 2018-06-28 | 2020-01-02 | 阿里巴巴集团控股有限公司 | Data verification method and apparatus, and storage medium |
| CN111447044A (en) * | 2020-03-10 | 2020-07-24 | 深圳市大数据研究院 | Distributed storage method and transmission decoding method |
| CN112202910A (en) * | 2020-10-10 | 2021-01-08 | 上海威固信息技术股份有限公司 | Computer distributed storage system |
| CN112947843A (en) * | 2019-12-10 | 2021-06-11 | 北京金山云网络技术有限公司 | Configuration and scheduling method and device of storage system and electronic equipment |
| CN115865989A (en) * | 2023-02-21 | 2023-03-28 | 中国市政工程西南设计研究总院有限公司 | Wide area network configuration method for efficient and safe interconnection of information of enterprise headquarters and branch offices |
| CN117591039A (en) * | 2024-01-18 | 2024-02-23 | 济南浪潮数据技术有限公司 | Distributed storage method, system, equipment and medium |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101135994A (en) * | 2007-09-07 | 2008-03-05 | 杭州华三通信技术有限公司 | Method and apparatus for dividing cache space and cache controller thereof |
| CN101808095A (en) * | 2010-03-22 | 2010-08-18 | 华中科技大学 | Encryption copy organization method under distributed storage environment |
| CN101827121A (en) * | 2010-03-12 | 2010-09-08 | 成都市华为赛门铁克科技有限公司 | Method, service end and system for creating files in RAID (Redundant Array of Independent Disk) |
| CN101872320A (en) * | 2010-04-16 | 2010-10-27 | 浪潮电子信息产业股份有限公司 | Method for reliability, performance test and statistic of SSD(Solid State Disk) |
| US20100274762A1 (en) * | 2009-04-24 | 2010-10-28 | Microsoft Corporation | Dynamic placement of replica data |
| CN102148871A (en) * | 2011-03-18 | 2011-08-10 | 浪潮(北京)电子信息产业有限公司 | Storage resource scheduling method and device |
| CN102300240A (en) * | 2011-08-26 | 2011-12-28 | 北京邮电大学 | Output performance parameter-based method for evaluating similarity of two systems |
| CN102301367A (en) * | 2008-10-24 | 2011-12-28 | Ilt创新公司 | Distributed data storage |
| CN102726031A (en) * | 2011-07-22 | 2012-10-10 | 华为技术有限公司 | Content processing method, device and system |
| CN103064914A (en) * | 2012-12-20 | 2013-04-24 | 曙光信息产业(北京)有限公司 | Data processing system and method |
-
2014
- 2014-01-17 CN CN201410022973.5A patent/CN103731505A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101135994A (en) * | 2007-09-07 | 2008-03-05 | 杭州华三通信技术有限公司 | Method and apparatus for dividing cache space and cache controller thereof |
| CN102301367A (en) * | 2008-10-24 | 2011-12-28 | Ilt创新公司 | Distributed data storage |
| EP2908257A1 (en) * | 2008-10-24 | 2015-08-19 | Compuverde AB | Distributed data storage |
| US20100274762A1 (en) * | 2009-04-24 | 2010-10-28 | Microsoft Corporation | Dynamic placement of replica data |
| CN101827121A (en) * | 2010-03-12 | 2010-09-08 | 成都市华为赛门铁克科技有限公司 | Method, service end and system for creating files in RAID (Redundant Array of Independent Disk) |
| CN101808095A (en) * | 2010-03-22 | 2010-08-18 | 华中科技大学 | Encryption copy organization method under distributed storage environment |
| CN101872320A (en) * | 2010-04-16 | 2010-10-27 | 浪潮电子信息产业股份有限公司 | Method for reliability, performance test and statistic of SSD(Solid State Disk) |
| CN102148871A (en) * | 2011-03-18 | 2011-08-10 | 浪潮(北京)电子信息产业有限公司 | Storage resource scheduling method and device |
| CN102726031A (en) * | 2011-07-22 | 2012-10-10 | 华为技术有限公司 | Content processing method, device and system |
| CN102300240A (en) * | 2011-08-26 | 2011-12-28 | 北京邮电大学 | Output performance parameter-based method for evaluating similarity of two systems |
| CN103064914A (en) * | 2012-12-20 | 2013-04-24 | 曙光信息产业(北京)有限公司 | Data processing system and method |
Non-Patent Citations (2)
| Title |
|---|
| 任飞 等: "大规模分布式存储系统中数据修复策略的研究", 《互联网天地》 * |
| 郑彬: "数字版权管理系统的应用与研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106933492B (en) * | 2015-12-30 | 2020-05-22 | 伊姆西Ip控股有限责任公司 | Method and apparatus for facilitating wear leveling of solid state disk |
| CN106933492A (en) * | 2015-12-30 | 2017-07-07 | 伊姆西公司 | The method and apparatus for contributing to the abrasion equilibration of solid state hard disc |
| US10372349B2 (en) | 2015-12-30 | 2019-08-06 | EMC IP Holding Company LLC | Method and apparatus for facilitating wear leveling of solid state disk |
| CN105554149A (en) * | 2015-12-31 | 2016-05-04 | 康佳集团股份有限公司 | Video cloud storage load equalizing method and system based on video ranks |
| CN105975345B (en) * | 2016-05-20 | 2019-03-15 | 江苏得得空间信息科技有限公司 | A kind of video requency frame data dynamic equalization memory management method based on distributed memory |
| CN105975345A (en) * | 2016-05-20 | 2016-09-28 | 江苏得得空间信息科技有限公司 | Video frame data dynamic equilibrium memory management method based on distributed memory |
| CN106156317A (en) * | 2016-06-30 | 2016-11-23 | 电子科技大学 | A kind of secure storage method of data based on Attribute transposition |
| CN106156317B (en) * | 2016-06-30 | 2019-05-10 | 电子科技大学 | A kind of secure storage method of data based on Attribute transposition |
| CN106230982A (en) * | 2016-09-08 | 2016-12-14 | 哈尔滨工程大学 | A kind of dynamic self-adapting secure cloud storage method considering node reliability |
| CN106230982B (en) * | 2016-09-08 | 2019-07-16 | 哈尔滨工程大学 | A kind of dynamic self-adapting secure cloud storage method considering node reliability |
| CN108241552B (en) * | 2016-12-23 | 2022-04-12 | 中科星图股份有限公司 | Client file recovery method |
| CN108241552A (en) * | 2016-12-23 | 2018-07-03 | 航天星图科技（北京）有限公司 | A kind of client file restoration methods |
| CN108875035A (en) * | 2018-06-25 | 2018-11-23 | 郑州云海信息技术有限公司 | The date storage method and relevant device of distributed file system |
| CN108875035B (en) * | 2018-06-25 | 2022-02-18 | 郑州云海信息技术有限公司 | Data storage method of distributed file system and related equipment |
| WO2020001287A1 (en) * | 2018-06-28 | 2020-01-02 | 阿里巴巴集团控股有限公司 | Data verification method and apparatus, and storage medium |
| CN110659151A (en) * | 2018-06-28 | 2020-01-07 | 阿里巴巴集团控股有限公司 | Data verification method and device and storage medium |
| US11537304B2 (en) | 2018-06-28 | 2022-12-27 | Alibaba Group Holding Limited | Data verification method and apparatus, and storage medium |
| CN110659151B (en) * | 2018-06-28 | 2023-05-02 | 阿里巴巴集团控股有限公司 | Data verification method and device and storage medium |
| CN109460426A (en) * | 2018-11-05 | 2019-03-12 | 郑州云海信息技术有限公司 | A kind of system and method, the routing node of the classification storage based on MongoDB |
| CN112947843A (en) * | 2019-12-10 | 2021-06-11 | 北京金山云网络技术有限公司 | Configuration and scheduling method and device of storage system and electronic equipment |
| CN111447044A (en) * | 2020-03-10 | 2020-07-24 | 深圳市大数据研究院 | Distributed storage method and transmission decoding method |
| CN111447044B (en) * | 2020-03-10 | 2022-12-09 | 深圳市大数据研究院 | Distributed storage method and transmission decoding method |
| CN112202910A (en) * | 2020-10-10 | 2021-01-08 | 上海威固信息技术股份有限公司 | Computer distributed storage system |
| CN112202910B (en) * | 2020-10-10 | 2021-10-08 | 上海威固信息技术股份有限公司 | Computer distributed storage system |
| CN115865989A (en) * | 2023-02-21 | 2023-03-28 | 中国市政工程西南设计研究总院有限公司 | Wide area network configuration method for efficient and safe interconnection of information of enterprise headquarters and branch offices |
| CN115865989B (en) * | 2023-02-21 | 2023-05-12 | 中国市政工程西南设计研究总院有限公司 | Wide area network configuration method for high-efficiency and safe interconnection of enterprise headquarter and branch office information |
| CN117591039A (en) * | 2024-01-18 | 2024-02-23 | 济南浪潮数据技术有限公司 | Distributed storage method, system, equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103731505A (en) | Data distributed storage method and system | |
| CN102411616B (en) | Method and system for storing data and data management method | |
| US20120221373A1 (en) | Estimating Business Service Responsiveness | |
| CN113821332B (en) | Method, device, equipment and medium for optimizing efficiency of automatic machine learning system | |
| CN104077280A (en) | Community discovery parallelization method, community discovery parallelization system, host node equipment and computing node equipment | |
| WO2023082629A1 (en) | Data storage method and apparatus, electronic device, and storage medium | |
| CN115080248B (en) | Scheduling optimization method for scheduling device, and storage medium | |
| CN110891087B (en) | Log transmission method and device, electronic equipment and storage medium | |
| CN111858146A (en) | Method, apparatus and computer program product for recovering data | |
| CN113255263B (en) | Particle band dividing method, device, computer equipment and storage medium | |
| CN107220271A (en) | A kind of method and system of distributed digital resource storage processing and management | |
| Cheng et al. | Convex contractive interval linear programming for resources and environmental systems management | |
| CN114866563A (en) | Capacity expansion method, device, system and storage medium | |
| CN113325998A (en) | Read-write speed control method and device | |
| CN103248622B (en) | A kind of Online Video QoS guarantee method of automatic telescopic and system | |
| US9424945B2 (en) | Linear programming based decoding for memory devices | |
| RU2615072C2 (en) | Information processing method and device and recording medium | |
| Yang et al. | Reliability assurance of big data in the cloud: Cost-effective replication-based storage | |
| CN111723907B (en) | Model training device, method, system and computer readable storage medium | |
| CN116991334B (en) | Data storage method, system, device, electronic equipment and readable storage medium | |
| CN117032954B (en) | Memory optimization method, system, equipment and medium for terminal training model | |
| Jewson | Application of uncertain hurricane climate change projections to catastrophe risk models | |
| CN111598390B (en) | Method, device, equipment and readable storage medium for evaluating high availability of server | |
| CN117349075A (en) | Data processing method and related equipment | |
| CN115034351A (en) | Data processing method, convolutional neural network training method and device and FPGA |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20140416 |