CN103605483A - Feature processing method for block-level data in hierarchical storage system - Google Patents
Feature processing method for block-level data in hierarchical storage system Download PDFInfo
- Publication number
- CN103605483A CN103605483A CN201310587772.5A CN201310587772A CN103605483A CN 103605483 A CN103605483 A CN 103605483A CN 201310587772 A CN201310587772 A CN 201310587772A CN 103605483 A CN103605483 A CN 103605483A
- Authority
- CN
- China
- Prior art keywords
- data
- piece
- block
- reference frequency
- storage
- 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 provides a feature processing method for block-level data in a hierarchical storage system. The method includes: setting a hierarchical storage system architecture comprising a data feature collector, a storage resource manager, a strategy configuration manger and a migration manager; allowing the storage resource manager to integrate storage resources to form a storage resource pool capable of hierarchical storage, and mapping the storage resources into a logical volume to use in an upper-level file system; allowing the data feature collector to be responsible for monitoring upper-level transmitted I/O requests, counting reference frequency features for a certain data block, analyzing by a data feature analysis method, and informing the migration manager of analysis results to allow for subsequent hierarchical management. Compared to the prior art, the feature processing method for block-level data in the hierarchical storage system allows hot spot data to be discovered and judged during the process of hierarchical data storage and management and allows the hot spot data to be judged more accurately.
Description
Technical field
The present invention relates to intelligent data management domain, specifically a kind of piece DBMS characteristic processing method in hierarchical stor.
Background technology
Data staging memory technology is mainly at storing virtual layer, storage resources and hot spot data to be carried out to Proper Match according to the data characteristics of monitored object, realizes the efficient utilization of storage resources.In existing classification storage data characteristics analytical approach, a class is the several data characteristic attribute that utilizes file object to comprise, as file size, type etc. are carried out the analysis of obtaining of data characteristics, and the data with different characteristic are carried out to Classification Management.But, the base unit of file as data characteristics statistics usingd in classification based on file-level storage, if file is larger, and when only having local message frequent when accessed, with regard to existing, for hot spot data, locate inaccurate problem so, thereby cause by the data placement that really need to be graded on efficient and expensive storage resources, being unfavorable for improving storage administration efficiency.Classification based on piece level storage is a kind of fine-grained data staging Managed Solution, can be by the granularity refinement of data management to according to the defined extension blocks level of application demand.Yet, for obtaining of data characteristics and analysis, be difficult point, particularly the judgement for piece level hot spot data has problems.If the comparative result based on piece citation times and threshold values is as the criterion of the cold and hot degree of data separately, thereby make the decision-making of Data Migration, so probably cause data dithering problem, be that data will repeat the operation of lifting/lowering level at short notice, this mode will cause the consumption of system resource, is unfavorable for improving system performance.
Summary of the invention
Technical assignment of the present invention is to solve the deficiencies in the prior art, and a kind of piece DBMS characteristic processing method in hierarchical stor is provided.
Technical scheme of the present invention realizes in the following manner, piece DBMS characteristic processing method in this kind of hierarchical stor, first hierarchical stor framework is set, this framework comprises data characteristics gatherer, Storage Resource Management (SRM) device, policy configuration management device and migration manager, and wherein various piece to the processing procedure of piece DBMS feature is:
One, Storage Resource Management (SRM) device is integrated storage resources, formation possesses the memory resource pool of classification storage capacity, utilize Storage Resource Management (SRM) device according to performance sequential organization disk from high to low, and build multilevel device chained list, equipment is carried out to unified management, then storage resources is mapped as to logical volume for upper strata file system;
Two, data characteristics gatherer is responsible for monitoring the I/O request of upper strata transmission, and statistics is directed to the reference frequency feature of certain data block, then according to data characteristics analytical approach, analyze, analysis result is notified to migration manager and carried out follow-up differentiated control behaviour;
Three, migration manager utilizes policy configuration management device to arrange parameters such as the scan period of blocks of data, maximum unused times simultaneously, to can provide decision information for migration operation.
Data characteristics gatherer in described step 2 carries out monitoring management to the complete I/O event of blocks of data, wherein I/O event comprises transmission size, response time, LBA (Logical Block Addressing) LBA and the magnetic disc ID of I/O event, the queue of execution I/O occurs, data characteristics gatherer shines upon each I/O bis-tuple to unique piece, and add up the I/O number of each piece, then, calculate periodically the reference frequency of each piece, statistics is read number of references and is write number of references respectively.
The detailed process of described step 2 is: the request queue of data characteristics gatherer receives the I/O request from generic block equipment, when I/O enters after queue, the running of notice hierarchy system worker thread, worker thread is write data according to the different device block of the dissimilar division of I/O to I/O request storage, data write the block device of distribution after I/O finish; When processing write requests, the write information of recording data blocks, comprises the write request quantity of equipment, the write request quantity of piece; For access reference count and the access time information of reading I/O request statistics equipment and piece, and then I/O is forwarded; Finally, the reference frequency information of whole queue is added up, according to data characteristics computing method, carry out the calculating of eigenwert, thus the cold and hot degree of decision data.
Described data feature values comprises reference frequency deviation E
i(t) with quote deviation variation rate DE
i(t): piece reference frequency deviation E
i(t) represent piece S
iin data management period T, data block S
ithe difference of quoting threshold values threshold of accessed actual number of references M and this piece level equipment of living in, reflects the difference level of this blocks of data and average temperature; Quote deviation variation rate DE
i(t) represent piece S
ithe change frequency of deviation within certain period, this value reflection data block S
iactive degree.
Described reference frequency deviation E
i(t) account form is as follows: data block S
iin data management period T, S
ithe difference of quoting threshold values threshold of accessed actual number of references M and this piece level equipment of living in;
Quote deviation variation rate DE
i(t) account form is as follows: data block S
icurrent sampling instant t reference frequency deviation E
i(t) with a upper sampling instant t-1 reference frequency deviation E
i(t-1) difference between, then divided by sampling interval time △ t.
The beneficial effect that the present invention compared with prior art produced is:
In a kind of hierarchical stor of the present invention, piece DBMS characteristic processing method solves in the hierarchical stor based on piece level the decision problem for hot spot data, in hierarchical stor, piece DBMS characteristic analysis method can be realized for the blocks of data of a part of fixed size in volume and monitoring and the statistical study of data temperature, then completes the differentiated control operation for blocks of data.In volume, need the object granularity of management thinner, the location of hot spot data is also more accurate, therefore can obtain higher data allocations efficiency, when the operation such as moving, loss is less, be conducive to improve utilization factor and the efficiency of management of storage resources, this invention can be used for hierarchical storage management in intelligent data management domain, by the access characteristic information to bottom physical block, obtain and analyze, realization is for discovery and the judgement of hot spot data in data staging storage administration process, improve the accuracy that hot spot data is judged, practical, be easy to promote.
Accompanying drawing explanation
Accompanying drawing 1 is hierarchical stor framework map of the present invention.
Accompanying drawing 2 is implementation schematic diagram of data characteristics gatherer of the present invention.
Accompanying drawing 3 is computing formula of number of references of the present invention.
Accompanying drawing 4 is computing formula of reference frequency deviation of the present invention.
Accompanying drawing 5 is computing formula of quoting deviation variation rate of the present invention.
Embodiment
Below in conjunction with accompanying drawing, piece DBMS characteristic processing method in a kind of hierarchical stor of the present invention is elaborated.
The present invention carries out analytic statistics to the data characteristics based on piece level, due to the blocks of data object logical OR physics bottom in memory hierarchy substantially, relatively simple than file object data structure, and comprise and read and write for the main operation of bottom data piece, therefore, for the read/write frequency of data block, become and judge that whether these data are the principal character of focus.In data characteristics analytical approach based on piece level, be mainly that the number of references of data block is added up, when data block records respectively its read/write number of times during by read/write, to calculate reference frequency, this blocks of data within a period of time by the number of times sum of read/write.On this basis, the reference frequency of data block is carried out to mathematical computations, defined respectively reference frequency deviation E
i(t) with quote deviation variation rate DE
i(t), to reflect the use temperature of data block and the activity of blocks of data.
As shown in Figure 1, the invention provides a kind of piece DBMS characteristic processing method in hierarchical stor, first hierarchical stor framework is set, this framework comprises data characteristics gatherer, Storage Resource Management (SRM) device, policy configuration management device and migration manager, in hierarchical stor, between different assemblies, association process is: Storage Resource Management (SRM) device is integrated storage resources, formation possesses the memory resource pool of classification storage capacity, utilize Storage Resource Management (SRM) device according to performance sequential organization disk from high to low, and build multilevel device chained list, equipment is carried out to unified management.Then storage resources is mapped as to logical volume for upper strata file system etc.; Data characteristics gatherer is responsible for monitoring the I/O request of upper strata transmission, and statistics is directed to the features such as reference frequency of certain data block, then according to data characteristics analytical approach, analyze, analysis result is notified to migration manager and carried out follow-up differentiated control operation.Migration manager utilizes policy configuration management device to arrange parameters such as the scan period of blocks of data, maximum unused times simultaneously, to can provide decision information for migration operation.
In the present invention, related data characteristics gatherer is responsible for the characteristic information of monitor data piece, and the characteristic information of blocks of data is carried out to statistical study according to temperature eigenwert, and wherein, temperature eigenwert comprises reference frequency deviation E
i(t) with quote deviation variation rate DE
i(t).Piece reference frequency deviation E
i(t) represent piece S
iin data management period T, data block S
ithe difference of quoting threshold values threshold of accessed actual number of references M and this piece level equipment of living in, has reflected the difference level of this blocks of data with average temperature; Quote deviation variation rate DE
i(t) represent piece S
ithe change frequency of deviation within certain period, this value has reflected data block S
iactive degree.The request queue of data characteristics gatherer receives the I/O request from generic block equipment, when I/O enters after queue, the running of notice hierarchy system worker thread, worker thread is write data according to the different device block of the dissimilar division of I/O to I/O request storage, data write the block device of distribution after I/O finish.When processing write requests, the write information of recording data blocks, comprises the write request quantity of equipment, the write request quantity of piece; For access reference count and the access time information of reading I/O request statistics equipment and piece, and then I/O is forwarded.Finally, the reference frequency information of whole queue is added up, according to data characteristics computing method, carry out the calculating of eigenwert, thus the cold and hot degree of decision data.
Described reference frequency deviation E
i(t) account form is as follows: data block S
iin data management period T, S
ithe difference of quoting threshold values threshold of accessed actual number of references M and this piece level equipment of living in;
Quote deviation variation rate DE
i(t) account form is as follows: data block S
icurrent sampling instant t reference frequency deviation E
i(t) with a upper sampling instant t-1 reference frequency deviation E
i(t-1) difference between, then divided by sampling interval time △ t.
Classification prototype system is the classification foundation using hot spot data access characteristics such as data time axle, deviation ratio and deviation variation rates as data resource in the process of implementation, and according to the readwrite performance difference of memory device, carry out the differentiated control of storage resources and hot spot data.Therefore, for the collection of data characteristics, be the basis of carrying out signature analysis, whole data characteristics collection process is as shown in Figure 2.Data characteristics gatherer carries out monitoring management to the complete I/O event of blocks of data, and wherein I/O event comprises transmission size, response time, LBA (Logical Block Addressing) LBA and the magnetic disc ID that I/O event occurs, and also comprises the queue of carrying out I/O.Data characteristics gatherer shines upon each I/O bis-tuple (LBA, magnetic disc ID) to unique piece, and adds up the I/O number of each piece.Then, calculate periodically the reference frequency of each piece, statistics is read number of references and is write number of references respectively.Feature collection and analytic process are as follows:
First, data characteristics gatherer utilizes hsm_do_bio function to monitor the data block access I/O of general layer transmission, and the I/O request of each piece is joined in the middle of the I/O queue, IOQ that hierarchical stor safeguards.Utilize bio_for_each_segment function traversal I/O queue, IOQ, judge respectively I/O action type, if write operation, so just by the write operation number of references W of this data block
iadd up, process equally the read operation number of references R of this data block
i.
Then, data characteristics gatherer adds up the number of references of all data blocks, obtains total number of references of distinct device, wherein total_read_hitcount be level total read number of references, total_write_hitcount be level total write number of references.Then utilize this layer of capacity tiersize and cell block size blocksize to calculate the average number of references of this layer, average_read_hitcount is that level on average reads number of references, and average_write_hitcount is that level on average writes number of references.Computation process formula 1 as shown in Figure 3.
Finally, implement the analysis phase to data characteristics.Characteristic variable E for reflection data temperature
iand DE (t)
i(t) carry out statistical study.To piece S
ireference frequency deviation E
i(t) formula 2 as shown in Figure 4, quotes deviation variation rate DE
i(t) shown in the formula 3 shown in calculating accompanying drawing 5, in accompanying drawing 5, S
ireference frequency deviation variation rate refer to current sampling instant reference frequency deviation E
i(t) with a upper sampling instant reference frequency deviation E
i(t-1) difference between, wherein, △ t is the sampling time.Calculating E
i(t), time, need to the average reference frequency average_hitcount in different levels be calculated.
By data feature values is analyzed, according to DE
iand E (t)
i(t) determine the temperature of data, then carry out relevant Data Migration operation.In carrying out data staging management process, for the bookkeeping of data, be no longer simply according to threshold method, to carry out the operation of lifting/lowering level, but need to be according to characteristic variable DE
iand E (t)
i(t) to data block S
itemperature judge rear enforcement.If data block S
ireference frequency may surpass threshold values, still, its access frequency is the statistics completing within several cycles, so DE
i(t) value will reduce, thereby reflects S
iactive degree not high, probably within following one period of cycle, there will not be intensive access situation, there is certain degradation expection, will continue to carry out the statistic analysis of reference frequency so, and updating operation can not occur.Having avoided like this upgrading or degraded operation blindly, retained an operation buffer and make whole transition process more level and smooth, is not a kind of either-or process.
Except the technical characterictic described in instructions, be the known technology of those skilled in the art.
Claims (5)
1. piece DBMS characteristic processing method in a hierarchical stor, it is characterized in that, first hierarchical stor framework is set, this framework comprises data characteristics gatherer, Storage Resource Management (SRM) device, policy configuration management device and migration manager, and wherein various piece to the processing procedure of piece DBMS feature is:
One, Storage Resource Management (SRM) device is integrated storage resources, formation possesses the memory resource pool of classification storage capacity, utilize Storage Resource Management (SRM) device according to performance sequential organization disk from high to low, and build multilevel device chained list, equipment is carried out to unified management, then storage resources is mapped as to logical volume for upper strata file system;
Two, data characteristics gatherer is responsible for monitoring the I/O request of upper strata transmission, and statistics is directed to the reference frequency feature of certain data block, then according to data characteristics analytical approach, analyze, analysis result is notified to migration manager and carried out follow-up differentiated control behaviour;
Three, migration manager utilizes policy configuration management device to arrange the scan period of blocks of data, maximum unused time parameter simultaneously, to can provide decision information for migration operation.
2. piece DBMS characteristic processing method in a kind of hierarchical stor according to claim 1, it is characterized in that, data characteristics gatherer in described step 2 carries out monitoring management to the complete I/O event of blocks of data, wherein I/O event comprises transmission size, response time, LBA (Logical Block Addressing) LBA and the magnetic disc ID that I/O event occurs, carry out the queue of I/O, data characteristics gatherer shines upon each I/O bis-tuple to unique piece, and add up the I/O number of each piece, then, calculate periodically the reference frequency of each piece, statistics is read number of references and is write number of references respectively.
3. piece DBMS characteristic processing method in a kind of hierarchical stor according to claim 2, it is characterized in that, the detailed process of described step 2 is: the request queue of data characteristics gatherer receives the I/O request from generic block equipment, when I/O enters after queue, the running of notice hierarchy system worker thread, worker thread is write data according to the different device block of the dissimilar division of I/O to I/O request storage, data write the block device of distribution after I/O finish; When processing write requests, the write information of recording data blocks, comprises the write request quantity of equipment, the write request quantity of piece; For access reference count and the access time information of reading I/O request statistics equipment and piece, and then I/O is forwarded; Finally, the reference frequency information of whole queue is added up, according to data characteristics computing method, carry out the calculating of eigenwert, thus the cold and hot degree of decision data.
4. piece DBMS characteristic processing method in a kind of hierarchical stor according to claim 3, is characterized in that, described data feature values comprises reference frequency deviation E
i(t) with quote deviation variation rate DE
i(t): piece reference frequency deviation E
i(t) represent piece S
iin data management period T, data block S
ithe difference of quoting threshold values threshold of accessed actual number of references M and this piece level equipment of living in, reflects the difference level of this blocks of data and average temperature; Quote deviation variation rate DE
i(t) represent piece S
ithe change frequency of deviation within certain period, this value reflection data block S
iactive degree.
5. piece DBMS characteristic processing method in a kind of hierarchical stor according to claim 4, is characterized in that, described reference frequency deviation E
i(t) account form is as follows: data block S
iin data management period T, S
ithe difference of quoting threshold values threshold of accessed actual number of references M and this piece level equipment of living in;
In a kind of hierarchical stor according to claim 4, piece DBMS characteristic processing method, is characterized in that, quotes deviation variation rate DE
i(t) account form is as follows: data block S
icurrent sampling instant t reference frequency deviation E
i(t) with a upper sampling instant t-1 reference frequency deviation E
i(t-1) difference between, then divided by sampling interval time △ t.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310587772.5A CN103605483A (en) | 2013-11-21 | 2013-11-21 | Feature processing method for block-level data in hierarchical storage system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310587772.5A CN103605483A (en) | 2013-11-21 | 2013-11-21 | Feature processing method for block-level data in hierarchical storage system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103605483A true CN103605483A (en) | 2014-02-26 |
Family
ID=50123715
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310587772.5A Pending CN103605483A (en) | 2013-11-21 | 2013-11-21 | Feature processing method for block-level data in hierarchical storage system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103605483A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106598501A (en) * | 2016-12-16 | 2017-04-26 | 郑州云海信息技术有限公司 | Data migration device and method for storage automatic hierarchy |
| CN109144417A (en) * | 2018-08-16 | 2019-01-04 | 广州杰赛科技股份有限公司 | A kind of cloud storage method, system and equipment |
| CN109634520A (en) * | 2018-11-29 | 2019-04-16 | 南京航空航天大学 | A kind of storage system based on HDFS CD server |
| WO2019085769A1 (en) * | 2017-10-30 | 2019-05-09 | 阿里巴巴集团控股有限公司 | Tiered data storage and tiered query method and apparatus |
| CN110515947A (en) * | 2019-08-23 | 2019-11-29 | 苏州浪潮智能科技有限公司 | A kind of storage system |
| CN111587423A (en) * | 2017-11-13 | 2020-08-25 | 维卡艾欧有限公司 | Hierarchical data policy for distributed storage systems |
| CN112035498A (en) * | 2020-08-31 | 2020-12-04 | 北京奇艺世纪科技有限公司 | Data block scheduling method and device, scheduling layer node and storage layer node |
| CN112612417A (en) * | 2020-12-24 | 2021-04-06 | 深圳市科力锐科技有限公司 | Data migration method, device, equipment and storage medium |
| CN112685337A (en) * | 2021-01-15 | 2021-04-20 | 浪潮云信息技术股份公司 | Method for hierarchically caching read and write data in storage cluster |
| WO2022116778A1 (en) * | 2020-12-02 | 2022-06-09 | International Business Machines Corporation | Enhanced application performance using storage system optimization |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0805396A1 (en) * | 1996-05-01 | 1997-11-05 | Sun Microsystems, Inc. | Multi-tier cache system for mass storage device and method for implementing such a system |
| CN101101563A (en) * | 2007-07-23 | 2008-01-09 | 清华大学 | Migration management based on massive data classified memory system |
| CN101989999A (en) * | 2010-11-12 | 2011-03-23 | 华中科技大学 | Hierarchical storage system in distributed environment |
-
2013
- 2013-11-21 CN CN201310587772.5A patent/CN103605483A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0805396A1 (en) * | 1996-05-01 | 1997-11-05 | Sun Microsystems, Inc. | Multi-tier cache system for mass storage device and method for implementing such a system |
| CN101101563A (en) * | 2007-07-23 | 2008-01-09 | 清华大学 | Migration management based on massive data classified memory system |
| CN101989999A (en) * | 2010-11-12 | 2011-03-23 | 华中科技大学 | Hierarchical storage system in distributed environment |
Non-Patent Citations (2)
| Title |
|---|
| 施光源等: "基于块级的分级存储数据特征模型及其应用研究", 《计算机研究与发展》 * |
| 施光源等: "基于模糊逻辑的数据分级存储模型研究", 《计算机科学》 * |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106598501A (en) * | 2016-12-16 | 2017-04-26 | 郑州云海信息技术有限公司 | Data migration device and method for storage automatic hierarchy |
| CN106598501B (en) * | 2016-12-16 | 2019-06-28 | 郑州云海信息技术有限公司 | For storing the Data Migration device and method of AUTOMATIC ZONING |
| WO2019085769A1 (en) * | 2017-10-30 | 2019-05-09 | 阿里巴巴集团控股有限公司 | Tiered data storage and tiered query method and apparatus |
| CN111587423A (en) * | 2017-11-13 | 2020-08-25 | 维卡艾欧有限公司 | Hierarchical data policy for distributed storage systems |
| CN111587423B (en) * | 2017-11-13 | 2023-09-19 | 维卡艾欧有限公司 | Hierarchical data policies for distributed storage systems |
| CN109144417A (en) * | 2018-08-16 | 2019-01-04 | 广州杰赛科技股份有限公司 | A kind of cloud storage method, system and equipment |
| CN109634520A (en) * | 2018-11-29 | 2019-04-16 | 南京航空航天大学 | A kind of storage system based on HDFS CD server |
| CN109634520B (en) * | 2018-11-29 | 2021-12-07 | 南京航空航天大学 | Storage system based on HDFS optical disc library |
| CN110515947A (en) * | 2019-08-23 | 2019-11-29 | 苏州浪潮智能科技有限公司 | A kind of storage system |
| CN112035498A (en) * | 2020-08-31 | 2020-12-04 | 北京奇艺世纪科技有限公司 | Data block scheduling method and device, scheduling layer node and storage layer node |
| CN112035498B (en) * | 2020-08-31 | 2023-09-05 | 北京奇艺世纪科技有限公司 | Data block scheduling method and device, scheduling layer node and storage layer node |
| WO2022116778A1 (en) * | 2020-12-02 | 2022-06-09 | International Business Machines Corporation | Enhanced application performance using storage system optimization |
| US11726692B2 (en) | 2020-12-02 | 2023-08-15 | International Business Machines Corporation | Enhanced application performance using storage system optimization |
| GB2616789A (en) * | 2020-12-02 | 2023-09-20 | Ibm | Enhanced application performance using storage system optimization |
| CN112612417B (en) * | 2020-12-24 | 2023-08-08 | 深圳市科力锐科技有限公司 | Data migration method, device, equipment and storage medium |
| CN112612417A (en) * | 2020-12-24 | 2021-04-06 | 深圳市科力锐科技有限公司 | Data migration method, device, equipment and storage medium |
| CN112685337B (en) * | 2021-01-15 | 2022-05-31 | 浪潮云信息技术股份公司 | Method for hierarchically caching read and write data in storage cluster |
| CN112685337A (en) * | 2021-01-15 | 2021-04-20 | 浪潮云信息技术股份公司 | Method for hierarchically caching read and write data in storage cluster |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103605483A (en) | Feature processing method for block-level data in hierarchical storage system | |
| US9753987B1 (en) | Identifying groups of similar data portions | |
| US9952803B1 (en) | Techniques for automated evaluation and moment of data between storage tiers | |
| US10339455B1 (en) | Techniques for determining workload skew | |
| CN103631538B (en) | Cold and hot data identification threshold value calculation, device and system | |
| US8868797B1 (en) | Techniques for automated discovery of storage devices and their performance characteristics | |
| US8838931B1 (en) | Techniques for automated discovery and performing storage optimizations on a component external to a data storage system | |
| US10671431B1 (en) | Extent group workload forecasts | |
| US9785353B1 (en) | Techniques for automated evaluation and movement of data between storage tiers for thin devices | |
| US9940024B1 (en) | Techniques for determining workload skew | |
| US8521986B2 (en) | Allocating storage memory based on future file size or use estimates | |
| Kim et al. | HybridStore: A cost-efficient, high-performance storage system combining SSDs and HDDs | |
| US9665630B1 (en) | Techniques for providing storage hints for use in connection with data movement optimizations | |
| US9026765B1 (en) | Performing write operations in a multi-tiered storage environment | |
| CN103605615B (en) | Block-level-data-based directional allocation method for hierarchical storage | |
| US10338825B2 (en) | Managing SSD wear rate in hybrid storage arrays | |
| US8862837B1 (en) | Techniques for automated data compression and decompression | |
| US9367262B2 (en) | Assigning a weighting to host quality of service indicators | |
| US9946465B1 (en) | Adaptive learning techniques for determining expected service levels | |
| CN102981971B (en) | A kind of phase transition storage loss equalizing method of quick response | |
| CN104714753A (en) | Data access and storage method and device | |
| CN104699424A (en) | Page hot degree based heterogeneous memory management method | |
| CN102857560A (en) | Multi-service application orientated cloud storage data distribution method | |
| CN109710184A (en) | Hierarchical hybrid storage method and system for tile record disk perception | |
| CN109582649A (en) | A kind of metadata storing method, device, equipment and readable storage medium storing program for executing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20140226 |