CN101241476B - Dummy storage system and method - Google Patents

Dummy storage system and method Download PDF

Info

Publication number
CN101241476B
CN101241476B CN2008100571892A CN200810057189A CN101241476B CN 101241476 B CN101241476 B CN 101241476B CN 2008100571892 A CN2008100571892 A CN 2008100571892A CN 200810057189 A CN200810057189 A CN 200810057189A CN 101241476 B CN101241476 B CN 101241476B
Authority
CN
China
Prior art keywords
data
resource
physical disk
mapping
virtual
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
Application number
CN2008100571892A
Other languages
Chinese (zh)
Other versions
CN101241476A (en
Inventor
朱旭东
柯剑
那文武
许金萍
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Institute of Computing Technology of CAS
Original Assignee
Institute of Computing Technology of CAS
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Institute of Computing Technology of CAS filed Critical Institute of Computing Technology of CAS
Priority to CN2008100571892A priority Critical patent/CN101241476B/en
Publication of CN101241476A publication Critical patent/CN101241476A/en
Application granted granted Critical
Publication of CN101241476B publication Critical patent/CN101241476B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The present invention provides a layered virtual storage system and method, the system includes a data logic module, a resource allocation/mapping module, a resource configuration/mapping module and a loading balance module, limiting logical semantic relation between the user data and the storage data, maintaining mapping from equipment address to resource pool address, organizing physical disk to virtual resource type having different attributes, providing a memory resource, detecting loading condition of the physical disk under the operation condition, processing adjustment at any times, easy expanding data logical semantic makes memory system quickly meet new application mode via applying the invention, users only give performance, credibility and space needs of the virtual disk, does not need manage division of the disk space, expand of space, mapping of resource.

Description

A kind of virtual storage system and method
Technical field
The present invention relates to technical field of information storage, more specifically, the present invention relates to a kind of virtual storage system and method.
Background technology
Along with the continuous development of virtual memory technique, network store system mainly provides manageability, high available, high performance data, services by virtual memory technique to the outside.Traditional virtual store is a plurality of physical disks in the memory device to be organized into a plurality of virtual disks offer user's use, each virtual disk can comprise the part or all of storage space of a plurality of physical disks, and can be defined as different types, and as RAID0, RAID0+1 etc.And the user can the defining virtual disk capacity and type, and can as using physical disk, use virtual disk, do not need the physical storage structure of being concerned about that disk is concrete, make the capacity that can expand virtual disk easily.
But traditional virtual memory structure is just corresponding one by one with amount of physical memory with the virtual disk storage space, and the distribution of amount of physical memory and being configured in when virtual disk defines is finished, and has following problem inevitably:
The management of data and the management of space resources are lumped together, the several data semanteme causes different virtual memory structures, need be the virtual memory structure of the semantic design specialized of every kind of virtual store, be difficult to general mutually between these storage organizations, cause the lifting of system complexity, increased cost and the difficulty of system development, maintenance and management;
The utilization rate of virtual disk is unbalanced, cause operating factor of memory space low, physical resource just distributes when virtual disk is created, virtual disk seldom can use these physical resources fully, when system does not have available physical resources, can't create new virtual disk, and idle physical resource is effectively used not;
After one class resource uses, need arteface and add the resource of the type, simultaneously, construct a class resource and need take a large amount of physical disks, be difficult between the dissimilar resources exchange, often have a class resources left, and can't distribute the situation of another kind of resource, the efficient and the dirigibility of storage space configuration are all very low;
The user will be bundled in the physical disk that this virtual disk shines upon to the visit of certain virtual disk, when its load is overweight, can't be the physical disk of load transfer to other low loads, and can there be load balance problem in memory device inside.
Summary of the invention
The extendability to using that has storage system now is poor, the storage resources service efficiency low and the defectives such as laod unbalance of physical store resource in order to overcome, and the present invention proposes a kind of virtual storage system and method.
According to an aspect of the present invention, provide a kind of virtual storage system, comprising: Data Service Model, resources allocation/mapping block and resource structure/mapping block;
Wherein, described Data Service Model is used to define the logical semantics relation between user data and the storage data, and the logical semantics of data and the storage semanteme of data are changed;
Wherein, described resources allocation/mapping block receives the storage data write request of described Data Service Model, manage the memory resource pool of a plurality of different attributes, safeguard the mapping of virtual unit address to the resource pool address, the mapping relations of establishment from the logical block to the Resource Block, and the mapping relations of described establishment are sent to resource structure/mapping block;
Wherein, described resource structure/mapping block managing physical disk is organized into the virtual resource form with different attribute with physical disk, according to described mapping relationship searching resource pool, provides storage resources to described resources allocation/mapping block.
Wherein, described system further comprises: the load balance module, described load balance module is responsible for detecting the loading condition of physical disk in system's operational process, carry out remapping of data migtation and resource, perhaps make in the resource structure/mapping block construction process afterwards, preferentially select the section of the little physical disk of load.
Wherein, described user data is to the visible data of user, and described storage data are the data that are stored on the physical medium.
Wherein, described Data Service Model comprises: the data operation parts, described data operation parts are changed mutually to input and output requesting users data with from the storage data of described resources allocation/mapping block according to the operational semantics rule.
Wherein, described Data Service Model comprises: the data map parts, described data map parts are set up the mapping of user data and some storage data manipulation sequences, and are mutual with described resources allocation/mapping block.
Wherein, described Data Service Model comprises: the data operation parts, and described data operation parts are changed mutually to input and output requesting users data with from the storage data of described resources allocation/mapping block according to the operational semantics rule;
The data map parts, described data map parts are set up the mapping of user data and some storage data manipulation sequences, and are mutual with described resources allocation/mapping block;
Described data operation parts are connected with described data map parts, to the data stream of input, with the output of the data operation parts input as the data map parts; To the data stream of output, with the output of data map parts input as the data operation parts.
Wherein, described resource structure/mapping block monitoring resource pool spatial variations is selected the physical disk section in good time, adds resource pool.
Wherein, described resource structure/mapping block is selected suitable physical disk section according to the type of resource and the state of physical disk, is configured to the resource of described type, joins resource pool; Wherein, the state of described physical disk comprises the size of physical disk free space and the request number that described resource per second arrives.
Wherein, described load balance module is selected the free segment of the little physical disk of load, and the data migtation of the frequent access section of the physical disk that load is big is revised the resource pool mapping table of the section of described physical disk to described free segment.
According to a further aspect in the invention, provide a kind of virtual memory method, comprising:
Logical semantics between step 10), definition user data and the storage data concerns, the logical semantics of request of data is converted to the storage semanteme of request of data;
Step 20), according to virtual address to the mapping table of resource pool address, carry out address translation to being converted to the semantic request of data of storage, safeguard the mapping of virtual unit address to the resource pool address, create the mapping relations from the logical block to the Resource Block;
Step 30), according to the resource pool address to the mapping table of physical disk address, the request of data that receives is carried out address translation, physical disk is organized into the virtual resource form with different attribute, storage resources is provided.
Wherein, described method further comprises:
Step 40), in system's operational process, detect the loading condition of physical disk, carry out remapping of data migtation and resource, perhaps make in the resource structure/mapping block construction process afterwards, preferentially select the section of the little physical disk of load.
Wherein, step 30) further comprise: according to the type of resource and the state of physical disk, select suitable physical disk section, be configured to the resource of described type, join resource pool; Wherein, the state of described physical disk comprises the size of physical disk free space and the request number that described resource per second arrives.
Wherein, step 40) further comprise: select the free segment of the little physical disk of load, the data migtation of the frequent access section of the disk that load is big is revised the resource pool mapping table at the section place of described physical disk to described free segment.
By using the present invention, being convenient to the expanded data logical semantics can make storage system satisfy new application model rapidly, the user only need simply provide the needed performance of virtual disk, reliability and space requirement, the division of Unsupervised disk space, the expansion in space, resource mapping; The virtual resource pond of different attribute is transparent fully to the user; And the present invention has good expandability and managerial, reduces secondary development cost and handling cost, improves the result of use of storage system simultaneously greatly.
Description of drawings
Fig. 1 is a hierarchical virtual memory system architecture synoptic diagram;
Fig. 2 is that the hierarchical virtual memory system data flow direction concerns synoptic diagram.
Embodiment
Below in conjunction with the drawings and specific embodiments, a kind of virtual storage system provided by the invention and method are described further.
Fig. 1 illustrates a kind of structure of virtual storage system according to an embodiment of the invention, in the present embodiment, virtual storage system comprises Data Service Model, resources allocation/mapping block, resource structure/mapping block and load balance module, and user's request inserts the Data Service Model of the virtual storage system of layering by the calibrated bolck equipment interface.
Data Service Model is used to define the logical semantics relation between user data and the storage data, and for example snapshot semanteme, version rely on semanteme, many copies semanteme, consistance semanteme, encryption is semantic, compression is semantic and other multiple operational semanticss.Alleged user data is to the visible data of user in the present embodiment, and the storage data are the data that are stored on the physical medium.Input and output (I/O) request of upper layer application to user data that be input as of Data Service Model is output as one or more input and output sequences to the storage data, and Data Service Model is divided into data operation parts and data map parts.
The data operation parts are changed mutually the storage data on I/O requesting users data and the storage medium according to the operational semantics rule.The operational semantics rule is by keeper's function setting according to equipment when creating virtual unit.For example in encryption device, operation rule is set at the AES cryptographic algorithm, when the data operation parts receive write request, user data is calculated by the AES cryptographic algorithm, obtains storing data, sends to resources allocation/mapping block; The data operation parts are read the storage data from resources allocation/mapping block when receiving read request, decipher by the inverse operation of cryptographic algorithm and obtain user data, return to the user.Again for example, the operation rule of data compression device is set at certain compression algorithm, and the data operation parts compress the user data of write request with compression algorithm, obtains storing data; The data operation parts obtain user data with the inverse operation decompress(ion) of the storage data use compression algorithm of read request, return to the user.
The data map parts are mapped as user data requests the sequence of operation of some storage data.The rule of mapping is by keeper's function setting according to equipment when creating virtual unit.For example a kind of simple mapping ruler of Snapshot Devices relies on and copy-on-write for reading: the data map parts write down the dependence equipment of reading of each Snapshot Devices, and safeguard a bitmap (bitmap) for each Snapshot Devices and represent the update mode of data block, each represents a data block, 1 expression is upgraded, and 0 expression is not upgraded.The data map parts find the updated space of data block correspondence when receiving user's read request, if 1 just sends to resources allocation/mapping block, the 0 dependence equipment of reading that is transmitted to this equipment.The data map parts find the updated space of data block correspondence when receiving user's write request, be updated if updated space is 1 this data block of expression, then directly send to resources allocation/mapping block.If being 0 this data block of explanation, updated space was not updated, then produce one and read the request of corresponding data piece from reading dependence equipment, data with the data of reading and write request merge then, produce a new write request, putting updated space is 1, promptly produces to comprise a read request, a write request and a sequence of operation of revising updated space.
Data operation parts and data map parts can be formed Data Service Model separately, also can mutual superposition form Data Service Model, data stream to input, with the output of data operation parts input as the data map parts, to the data stream of output, with the output of data map parts input as the data operation parts.The Snapshot Devices of an encryption for example, its Data Service Model just by data operation parts and the stack of data map parts, to write request, is encrypted with the data operation parts earlier, then data encrypted is sent to the data map parts, finish the copy-on-write operation; To read request, obtain storing data by the data map parts according to data dependence relation earlier, by the data operation parts data are decrypted then, return to application program at last.
Resources allocation/mapping block, the memory resource pool of managing a plurality of different attributes, the attribute of memory resource pool comprises peak bandwidth, reliability class and unit price etc.Every kind of resource in the pond all shows as the RAID of linearity (linear) or certain type configuration.For example the peak bandwidth of linear is the bandwidth of single physical disk, does not have redundance, and reliability class is minimum, and unit price is the unit price of physical disk.The reliability of 3*RAID0 is identical with linear with unit price, and peak bandwidth then is approximately 3 times of linear.The peak bandwidth of 4*RAID5 is identical with 3*RAID0, and reliability and unit price are all than 3*RAID0 height.The peak bandwidth of 5*RAID5 is higher than 4*RAID5, and reliability and price are then between 3*RAID0 and 4*RAID5.
Resources allocation/mapping block carries out the distribution of resource between a plurality of virtual pools, safeguard the mapping table of a virtual unit address to the resource pool address, and each in the table all comprises a LBA (Logical Block Addressing) and a Resource Block address.When a write request arrives resources allocation/mapping block,, then write request is mapped as request, and sends to resource structure/mapping block Resource Block if in mapping table, find corresponding request item.If do not find corresponding item, module is according to the Resource Properties demand of application settings, selection can be satisfied the resource pool of these attribute demands, in these resource pools, select idling-resource quantity resource pool more and that price is cheap as far as possible to distribute, and the mapping relations of establishment from the logical block to the Resource Block, join in the mapping table, then write request is mapped as request, and sends to resource structure/mapping block Resource Block.When a read request arrives this module,, then read request is mapped as request, and sends to resource structure/mapping block Resource Block if in mapping table, find corresponding item.If do not find corresponding item, then return complete 0 data to Data Service Model.
The IO request of the logical space that is input as Data Service Model output of resources allocation/mapping block is output as the IO request of resource pool address space, is transmitted to resource structure/mapping block.
Resource structure/mapping block managing physical disk, and physical disk is organized into the virtual resource form with different attribute, provide storage resources to resources allocation/mapping block.Resource structure/mapping block is with the space section of being divided into of physical disk, and section is the least unit of resource structure, and the size of section is set when system creation, and for example the physical space of 1MB is set at 1 section.The attribute and the type of the storage resources that structure obtains are introduced in resources allocation/mapping block.Each resource pool is safeguarded the mapping table of a resource pool space to physical disk space.Resource structure/mapping block is provided with the monitoring function interface in each resource pool, when resources allocation/mapping block Resources allocation from resource pool was given virtual unit, this monitoring function interface will be called.If the free space that monitor-interface is found this resource pool is during less than the threshold values set, notice resource structure/mapping block carries out resource and constructs.Resource structure/mapping block is selected suitable physical disk section according to the type of this resource and the situation of physical disk, is configured to the resource of this type, joins in the resource pool.The state of physical disk comprises the size of free space and the request number that per second arrives.For example the free space of monitor-interface discovery 3*RAID5 resource pool is just notified resource structure/mapping block less than 200MB, and resource structure/mapping block is checked all physical disks, selects 3 free spaces bigger, and per second arrives the less disk of request number.On three disks, respectively get one section, use the RAID5 algorithm construction to become a RAID5 band, and these mapping relations of taking 3 disk sections to are joined in the mapping table of 3*RAID5 resource pool.
Resource structure/mapping block is receiving the I/O request that resources allocation/mapping block sends, and the mapping table according to the address search resource pool of I/O request obtains corresponding physical disk section.Linear algorithm or RAID algorithm according to this resource pool sends to the physical disk section with request then.
The load balance module is responsible in system's operational process, detects the loading condition of physical disk, writes down the arrival request number of each physical disk and physical disk section.Reach request number and volume of transmitted data obviously greater than other disk if find the per second of certain physical disk, then carry out following adjustment: 1. remapping of data migtation and resource: the load balance module is selected the free segment on the little physical disk of load, the data migtation of the frequent access section on the disk that load is big is on this free segment, revise the resource pool mapping table at this section place, make the mapping of pointing to former disk section point to new disk section; 2. the suggestion of resource structure strategy is provided to resource structure/mapping block, makes in the resource structure/mapping block construction process afterwards, preferentially select the section of the little physical disk of load.
The data I flow process of system: as shown in Figure 2, the user only need to select the logical semantics of data and the storage resources attribute that needs, and the interface that uses the virtual unit system to provide is created the deletion virtual disk when using the virtual storage system of layering; In the read-write process, carry out the explanation of logical semantics, the logical semantics of data is converted to the storage semanteme of data, and is transmitted to resources allocation/mapping block by Data Service Model; Resources allocation/mapping block carries out address translation to the mapping table of resource pool address to the request that receives according to virtual address, sends to resource structure/mapping block then; Resource structure/mapping block carries out address translation to the mapping table of physical disk address to the request that receives according to the resource pool address, according to the RAID or the linear semanteme of this resource request is sent to physical disk, finishes the I/O operation by physical disk; If read operation, the data of reading return to upper layer module successively, return to application after handling through Data Service Model.When the resource quantity in the resource pool of certain attribute reduces to a certain degree, the resource constructing module will use physical disk to construct the resource of the type, put into resource pool, to guarantee demands of applications.Simultaneously, the space of load balance module monitors physical disk and the load state of performance are by data migtation with revise the tactful mode of structure and carry out load balance between many physical disks.
The present invention separates the data logical semantics of virtual memory facilities with storage is semantic.The hierarchical virtual device systems provides several data logical semanticses such as snapshot, versions of data, data encryption, data compression and many copies.Along with the development of application technology, new data storage form can constantly produce, and the virtual storage system of layering also allows the new data logical semantics of User Defined, need not to relate to complicated data storage and resource management problem.
The hierarchical virtual device systems provides the uniform resource management semanteme, resource management is divided into resources allocation, resource structure and load balance function, improve the space utilization efficient of resource, resources idle between the virtual unit of having avoided causing because of fixed allocation, resources idle between the different attribute of avoiding causing because of the resource structure provides the load balance in the memory device.
It should be noted that at last, above embodiment is only in order to illustrate that technical scheme of the present invention is not intended to limit, and on using, can extend to other modification, variation, application and embodiment, think that simultaneously all such modifications, variation, application, embodiment are within the spirit and scope of the present invention.

Claims (11)

1. a virtual storage system comprises: Data Service Model, resources allocation/mapping block, resource structure/mapping block and load balance module;
Wherein, described Data Service Model is used to define the logical semantics relation between user data and the storage data, and the logical semantics of data and the storage semanteme of data are changed, and user data requests is mapped as the sequence of operation of some storage data;
Wherein, described resources allocation/mapping block receives the storage data write request of described Data Service Model, manage the memory resource pool of a plurality of different attributes, safeguard the mapping of virtual unit address to the storage resources pool address, the mapping relations of establishment from the logical block to the Resource Block, and the mapping relations of described establishment are sent to resource structure/mapping block, the attribute of described memory resource pool comprises peak bandwidth, reliability class, unit price;
Wherein, described resource structure/mapping block managing physical disk is organized into the virtual resource form with different attribute with physical disk, according to described mapping relationship searching memory resource pool, provides storage resources to described resources allocation/mapping block; Resource structure/mapping block is provided with the monitoring function interface in each memory resource pool, when resources allocation/mapping block Resources allocation from memory resource pool is given virtual unit, this monitoring function interface will be called, if the free space of memory resource pool of knowing certain generic attribute is less than preset threshold, described resource structure/mapping block is according to the type of this resource and the situation of physical disk, select the physical disk section, dynamically construct the resource of this generic attribute, put into memory resource pool;
Load balance module, described load balance module are responsible for detecting physical disk in system's operational process loading condition carries out remapping of data migtation and resource; The load balance module also is used for providing to described resource structure/mapping block the suggestion of resource structure strategy, makes in the described resource structure/mapping block construction process afterwards, preferentially selects the section of the little physical disk of load.
2. the virtual storage system of claim 1, wherein, described user data is to the visible data of user, described storage data are the data that are stored on the physical medium.
3. the virtual storage system of claim 1, wherein, described Data Service Model comprises:
The data operation parts, described data operation parts are changed mutually to input and output requesting users data with from the storage data of described resources allocation/mapping block according to the operational semantics rule.
4. the virtual storage system of claim 1, wherein, described Data Service Model comprises:
The data map parts, described data map parts are set up the mapping of user data and some storage data manipulation sequences, and are mutual with described resources allocation/mapping block.
5. the virtual storage system of claim 1, wherein, described Data Service Model comprises:
The data operation parts, described data operation parts are changed mutually to input and output requesting users data with from the storage data of described resources allocation/mapping block according to the operational semantics rule;
The data map parts, described data map parts are set up the mapping of user data and some storage data manipulation sequences, and are mutual with described resources allocation/mapping block;
Described data operation parts are connected with described data map parts, to the data stream of input, with the output of the data operation parts input as the data map parts; To the data stream of output, with the output of data map parts input as the data operation parts.
6. the virtual storage system of claim 1, wherein, described resource structure/mapping block monitoring memory resource pool spatial variations is selected the physical disk section in good time, adds memory resource pool.
7. the virtual storage system of claim 1, wherein, described resource structure/mapping block is according to the type of resource and the state of physical disk, selects the physical disk section that is fit to, is configured to the resource of described type, joins memory resource pool; Wherein, the state of described physical disk comprises the size of physical disk free space and the request number that described resource per second arrives.
8. the virtual storage system of claim 1, wherein, described load balance module is selected the free segment of the little physical disk of load, and the data migtation of the frequent access section of the physical disk that load is big is revised the memory resource pool mapping table of the section of described physical disk to described free segment.
9. virtual memory method comprises:
Step 10), definition user data and the storage logical semantics relation between the data is converted to the storage semanteme of request of data with the logical semantics of request of data, user data requests is mapped as the sequence of operation of some storage data;
Step 20), physical disk is organized into virtual resource form with different attribute, according to the mapping table of virtual resource address to the storage resources pool address, carry out address translation to being converted to the semantic request of data of storage, safeguard the mapping of virtual unit address to the storage resources pool address, the mapping relations of establishment from the logical block to the Resource Block, the attribute of described memory resource pool comprises peak bandwidth, reliability class, unit price;
Step 30), according to the storage resources pool address to the mapping table of physical disk address, the request of data that receives is carried out address translation, physical disk is organized into the virtual resource form with different attribute, storage resources is provided; Resource structure/mapping block is provided with the monitoring function interface in each memory resource pool, when resources allocation/mapping block Resources allocation from memory resource pool is given virtual unit, this monitoring function interface will be called, if the free space of the memory resource pool of certain generic attribute is during less than preset threshold, described resource structure/mapping block is according to the type of this resource and the situation of physical disk, select the physical disk section, dynamically construct the resource of this generic attribute, put into memory resource pool;
Step 40), in system's operational process, detect the loading condition of physical disk, carry out remapping of data migtation and resource, and provide the suggestion of resource structure strategy to described resource structure/mapping block, make in the described resource structure/mapping block construction process afterwards, preferentially select the section of the little physical disk of load.
10. the method for claim 9, wherein, step 30) further comprise: according to the type of resource and the state of physical disk, select the physical disk section that is fit to, be configured to the resource of described type, join memory resource pool; Wherein, the state of described physical disk comprises the size of physical disk free space and the request number that described resource per second arrives.
11. the method for claim 9, wherein, step 40) further comprise: select the free segment of the little physical disk of load, the data migtation of the frequent access section of the disk that load is big is revised the memory resource pool mapping table at the section place of described physical disk to described free segment.
CN2008100571892A 2008-01-30 2008-01-30 Dummy storage system and method Expired - Fee Related CN101241476B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2008100571892A CN101241476B (en) 2008-01-30 2008-01-30 Dummy storage system and method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2008100571892A CN101241476B (en) 2008-01-30 2008-01-30 Dummy storage system and method

Publications (2)

Publication Number Publication Date
CN101241476A CN101241476A (en) 2008-08-13
CN101241476B true CN101241476B (en) 2010-12-08

Family

ID=39933015

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2008100571892A Expired - Fee Related CN101241476B (en) 2008-01-30 2008-01-30 Dummy storage system and method

Country Status (1)

Country Link
CN (1) CN101241476B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9424383B2 (en) 2011-04-11 2016-08-23 Nvidia Corporation Design, layout, and manufacturing techniques for multivariant integrated circuits

Families Citing this family (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101599978B (en) * 2009-07-17 2012-09-26 杭州华三通信技术有限公司 Method and equipment for allocating resources
CN101957729B (en) * 2010-09-27 2014-07-02 中兴通讯股份有限公司 Logical block transformation method and method and device compatible with reading and writing of user based on same
US20120110259A1 (en) * 2010-10-27 2012-05-03 Enmotus Inc. Tiered data storage system with data management and method of operation thereof
US9477597B2 (en) 2011-03-25 2016-10-25 Nvidia Corporation Techniques for different memory depths on different partitions
US9529712B2 (en) 2011-07-26 2016-12-27 Nvidia Corporation Techniques for balancing accesses to memory having different memory types
ES2608378T3 (en) 2011-12-29 2017-04-10 Huawei Technologies Co., Ltd. Cloud computing system and procedure to manage associated storage resources
CN102411524A (en) * 2011-12-30 2012-04-11 云海创想信息技术(天津)有限公司 Snapshot volume data copying method
CN104238955B (en) * 2013-06-20 2018-12-25 杭州迪普科技股份有限公司 A kind of device and method of storage resource virtualization distribution according to need
US9755987B2 (en) * 2014-02-05 2017-09-05 Futurewei Technologies, Inc. Virtual resource mapping mechanisms
CN105446813B (en) * 2014-05-30 2018-12-14 华为技术有限公司 A kind of method and device of resource allocation
CN104536909B (en) * 2014-12-09 2018-01-23 华为技术有限公司 A kind of memory management method, memory management unit and storage device
CN105988727B (en) * 2015-02-25 2021-04-16 中兴通讯股份有限公司 Storage method and storage device based on RAID
US10454845B2 (en) * 2015-04-22 2019-10-22 ColorTokens, Inc. Object memory management unit
CN105389129A (en) * 2015-11-11 2016-03-09 北京汉柏科技有限公司 Virtual cloud disk implementation method in cloud computing system
KR101754348B1 (en) * 2016-06-17 2017-07-06 고려대학교 산학협력단 Analyzing system for managing information storage table control method thereof
CN107168657B (en) * 2017-06-15 2020-05-26 深圳市云舒网络技术有限公司 Virtual disk hierarchical cache design method based on distributed block storage
CN110874257B (en) * 2018-08-31 2024-04-02 阿里巴巴集团控股有限公司 Method and device for configuring container on server
CN109324762A (en) * 2018-10-11 2019-02-12 郑州云海信息技术有限公司 A kind of construction method of storage system, device, equipment and storage medium
KR20200099882A (en) * 2019-02-15 2020-08-25 에스케이하이닉스 주식회사 Memory controller and operating method thereof
CN110058822B (en) * 2019-04-26 2022-06-24 北京计算机技术及应用研究所 Transverse expansion method for disk array
CN111708606A (en) * 2020-06-16 2020-09-25 杨海朝 Virtual computer storage service system and use method thereof
CN116701243B (en) * 2023-08-03 2023-10-20 烟台东方威思顿电气有限公司 Storage space allocation management method under bare computer

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9424383B2 (en) 2011-04-11 2016-08-23 Nvidia Corporation Design, layout, and manufacturing techniques for multivariant integrated circuits

Also Published As

Publication number Publication date
CN101241476A (en) 2008-08-13

Similar Documents

Publication Publication Date Title
CN101241476B (en) Dummy storage system and method
CN101997918B (en) Method for allocating mass storage resources according to needs in heterogeneous SAN (Storage Area Network) environment
US12079245B2 (en) System and method for supporting large queries in a multidimensional database environment
US10446174B2 (en) File system for shingled magnetic recording (SMR)
US7574560B2 (en) Methods, systems, and computer program products for dynamic mapping of logical units in a redundant array of inexpensive disks (RAID) environment
CN101814045B (en) Data organization method for backup services
CN102255962B (en) Distributive storage method, device and system
US8095728B2 (en) Method and system for power aware I/O scheduling
JP2017091548A (en) Dispersion storage system, and dispersion storage method, and dispersion facility
US7496586B1 (en) Method and apparatus for compressing data in a file system
CN103384877A (en) Storage system comprising flash memory, and storage control method
CN105027122A (en) Compression and deduplication layered driver
CN107016100A (en) A kind of metadata management method based on Nonvolatile memory file system
CN101315628A (en) Internal memory database system and method and device for implementing internal memory data base
CN102123176B (en) Space distribution and management method and device for network storage system
US20130097399A1 (en) Interface for management of data movement in a thin provisioned storage system
CN106570113B (en) Mass vector slice data cloud storage method and system
CN104272242B (en) Create encryption memory bank
CN104054071A (en) Method for accessing storage device and storage device
CN102945275B (en) File defragmentation method, device and equipment
CN108304142A (en) A kind of data managing method and device
CN103823641B (en) The virtual volume system of a kind of on-line rapid estimation and its implementation
JP2024525170A (en) Data compression method and device
US7424574B1 (en) Method and apparatus for dynamic striping
WO2024197789A1 (en) Fine-grained file system and file reading and writing method

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20101208

Termination date: 20220130

CF01 Termination of patent right due to non-payment of annual fee