CN101030165A - Magnetic disk space management and managing system - Google Patents
Magnetic disk space management and managing system Download PDFInfo
- Publication number
- CN101030165A CN101030165A CN 200610034112 CN200610034112A CN101030165A CN 101030165 A CN101030165 A CN 101030165A CN 200610034112 CN200610034112 CN 200610034112 CN 200610034112 A CN200610034112 A CN 200610034112A CN 101030165 A CN101030165 A CN 101030165A
- Authority
- CN
- China
- Prior art keywords
- disk
- binary tree
- disaster
- tree table
- space
- 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.)
- Granted
Links
Images
Abstract
A method for managing space of magnetic disc includes setting bit map binary tree table being used to record address and service state of space block as well as father-son relation between space blocks, utilizing said bit map binary tree table to carry out distribution and recovery on space block.
Description
Technical field
The invention belongs to computer realm, relate in particular to the method and system that disk space is managed.
Background technology
Partner (Buddy) algorithm is a kind of memory management algorithm under the Linux, can only distribute 2
nThe space piece of byte, n 〉=12 (promptly being 4kb at least).Relation between the piece of space can be represented by the binary tree of Fig. 1, except leaf, can tear half point open for every and be cleaved into two for distribution, and two adjacent fraternal free blocks of recovery can be merged into its father node-bigger free block.Because piece and address space have one-to-one relationship, so the position of its neighborhood of nodes determines and be unique that father node is also definite and unique.For example the address of 512MB piece must be 0,512M, 1024M, 1536M, wherein 0 with 512M be neighborhood of nodes, the address of its father node is 0.The advantage of buddy algorithm is because idle node can be merged into bigger free block very easily, so can not produce not recyclable fragment.
When utilizing buddy algorithm to carry out the memory headroom management, preserve distribution recovery information by two-part structure, these two parts all are internal storage structures.Referring to Fig. 1, idle queues is used for allocation space is convenient in free block taxonomic organization, when required length does not have, then searches bigger free block downwards to divide requisite space.The binary tree on the right has represented that taking of all nodes concerns and syntople, can search a situation that takies that reclaims its neighborhood of nodes of free block easily by it, is convenient to recurrence and upwards merges idle node, produces bigger free block-father node.
The data in magnetic disk file has two kinds usually, and a kind of is that form is fixed, and can not reclaim idle space in the time of can not dynamically changing or change, and does not possess the disk space management ability.Another kind has the general spatial managerial ability of distributing recovery.
As shown in Figure 2, the common way of general spatial management is that un-occupied space is distributed into the piece of fixed size and the piece of similar size, strings together the idle queues that becomes corresponding blocks length by chain type during recovery.But usually for better usage space, can support the distribution of some kinds of sizes, every kind of size all forms an idle queues, and idle queues is used up, and then distributes corresponding size from un-occupied space.Because node do not divide,,, can only preserve expression with chain pointer physically usually to the preservation (reaching certain rank) of some data stream such as length so very large node type can not be arranged.
In above-mentioned ordinary magnetic disc space management scheme, owing to consider the disk reading efficiency, can not merge the free block of recovery, so every kind of length that reclaims formation is all relevant with historical peak value,, can not effectively increase operation rate even the kind that size is supported is got carefully.Simultaneously, express,, when I/O (I/O), need the extra time of expense to be used for the magnetic head reorientation because it is discontinuous for the chain structure of some data stream.Operating system need read many garbages when bunch operating, or since node stride bunch, thereby more bunch of operating unit of expense reduced the I/O performance of disk.In addition,, can't support the disaster recovery of data, need additionally back up data because every position is all inconsistent.
Summary of the invention
One object of the present invention is to provide a kind of disk space management method, is intended to solve the free block in the disk management process that exists in the prior art and can't reclaims, the problem that the I/O performance is lower and the extra resource of needs is carried out disaster recovery.
Another object of the present invention is to provide a kind of disk space management system.
The present invention is achieved in that a kind of disk space management method, and described method comprises the steps:
A., bitmap binary tree table is set, is used for the set membership between address, user mode and the space piece of record space piece;
B. utilize described bitmap binary tree table that the space piece is distributed and reclaim.
Described step B further comprises:
When data input disk, search described bitmap binary tree table, distribute space piece with the described data of storage;
When the data output disk, reclaim corresponding free space piece;
Distribution or recovery state according to the space piece upgrade described bitmap binary tree table, keep the structure of described bitmap binary tree table consistent with the user mode of disk space.
Described method further comprises:
C. described bitmap binary tree table is mapped to internal memory, the disk operating of space management is converted to internal memory operation.
Described method further comprises:
D. when disaster takes place, recover described bitmap binary tree table.
Described step D further comprises:
D1., disk disaster sign and internal memory disaster sign are set;
D2. whether judge whether to take place disaster synchronously according to described disk disaster sign and internal memory disaster sign, when judging the generation disaster, the scanning disk space travels through described bitmap binary tree table according to the actual user mode of disk space, and described bitmap binary tree table is carried out topology update.
A kind of disk space management system, described system comprises:
The external interface module is used to receive the outside data of importing, and perhaps the data that disk is preserved are externally exported;
Data read-write module is used for the data that described external interface module receives are write disk, and perhaps the data that disk is preserved are exported by described external interface module;
Bitmap binary tree table administration module is used to be provided with bitmap binary tree table, and utilizes described bitmap binary tree table that the space piece is distributed and reclaim;
Described bitmap binary tree table is used for the set membership between address, user mode and the space piece of record space piece.
Described system further comprises:
The memory mirror mapping block is used for described bitmap binary tree table is mapped to internal memory, and the disk operating of space management is converted to internal memory operation.
Described system further comprises:
Disaster flag management module is used to be provided with disk disaster sign and internal memory disaster sign;
The disaster recovery administration module, be used for whether judging whether to take place disaster synchronously according to described disk disaster sign and internal memory disaster sign, when judging the generation disaster, the scanning disk space, actual user mode according to disk space travels through described bitmap binary tree table, and described bitmap binary tree table is carried out topology update.
When adopting the present invention to carry out disk space management, except that normal distribution recovery ability, has best magnetic disc i/o performance.Simultaneously, the space can merge fast, form big can not busy piece, the utilization factor height can not produce fragment.In addition, have the good data disaster recovery capability, and do not need occupation of extra resources.
Description of drawings
Fig. 1 is the binary tree structure synoptic diagram in partner (Buddy) algorithm;
Fig. 2 is the realization synoptic diagram of the ordinary magnetic disc space management that provides in the prior art;
Fig. 3 is the structural drawing of disk management file provided by the invention;
Fig. 4 is the structural representation of bitmap binary tree table provided by the invention;
Fig. 5 is the realization synoptic diagram of disk space division provided by the invention and recovery;
Fig. 6 is a bunch composition synoptic diagram in the present invention;
Fig. 7 is the realization flow figure of the disaster recovery in the initialization procedure provided by the invention;
Bitmap binary tree table when Fig. 8 is the dynamic input and output of disk runtime among the present invention upgrades synoptic diagram;
Fig. 9 is the composition structural drawing of disk management provided by the invention system.
Embodiment
In order to make purpose of the present invention, technical scheme and advantage clearer,, the present invention is further elaborated below in conjunction with drawings and Examples.Should be appreciated that specific embodiment described herein only in order to explanation the present invention, and be not used in qualification the present invention.
Management is applied to disk space management for memory headroom with buddy algorithm in the present invention, on the problem basis that its introducing disk space management of solution brings, utilize its characteristics that functional, safety is provided and the disk space management scheme of disaster recovery capability is arranged, because the site position is fixed, and wherein do not have chain pointer, do not produce construction error entirely because of pointer data is asynchronous.
In the present invention, at first the disk management file is written in the disk, Fig. 3 shows the structure of disk management file, comprises disaster sign, bitmap binary tree table and text, and disaster sign, bitmap binary tree table are Header (head) part of disk management file.Wherein, bitmap binary tree table is used for the size and the operating position of node (space piece) in the recording disc space, and when data write disk or data when deleted, corresponding node is assigned with or reclaims, and the structure of bitmap binary tree table will be upgraded.Text is the disk space that is used for distributing loading data.The disaster sign is used for identifying the disk management file and whether is in the disaster state, to judge whether to carry out disaster recovery.
The free time that the present invention uses the bitmap binary tree to show to express simultaneously in the buddy algorithm is reclaimed formation and binary tree structure, as shown in Figure 4.Relation between the subscript of bitmap has reflected the node set membership as binary tree, and with the seizure condition of 0 and 1 expression node of the value of bitmap, this bitmap binary tree table takies the preceding 64KB of disk space.By the bitmap subscript of bitmap binary tree table, can learn the position and the operating position of the neighborhood of nodes of certain node, also the pairing bitmap subscript of the node of certain length bound as can be known.Therefore, when data write disk and need allocation space, can be prior to searching idle node in this scope, if can not find then recurrence is searched free block in the bitmap subscript bound of major node more, produce required space piece with division.When needs reclaim idle node, can determine the seizure condition of neighborhood of nodes by bitmap, and recurrence upwards merges node, as shown in Figure 5.
If the table of the bitmap binary tree on the direct control disk, can be because magnetic disc i/o causes performance to descend, the present invention is mapped to internal memory with this bitmap binary tree table becomes a mirror image, because the disk management File mapping is synchronized to the operation of disk, handle by operating system in the time of can occurring in the disk management File mapping automatically and close, so only need operate memory mirror when using, be equivalent to pure internal memory operation, referring to Fig. 4.
Such scheme also has following advantage except the advantage with buddy algorithm no fragment on disk space management:
The first, every block space all is continuous, to the data of random length, only need head position of expense in theory, the highest I/O performance of physical disk level is arranged, in fact whether relevant continuously on physical disk with data, but minimum head position number of times is still arranged;
The second, the space divisional mode of buddy algorithm is identical with the scale-of-two of machine, so be easy to accomplish all bunch alignment of all pieces.Operating system is come management document with a bunch I/O minimum operation unit (NTFS's bunch is 4KB, and FAT32 is the multiple of 2KB, is called piece under the Linux, and default size is 1KB) usually, and bunch (piece) size is generally 2
nCan distribute several bunches to be used to deposit its data for file, again because the space piece in the buddy algorithm all is the multiple of these bunches (piece), do not influence a bunch alignment relation, so it can be asserted that, the start address of all pieces all is positioned at certain bunch reference position, therefore each operation all only act on minimum Operational System Control unit (bunch), as shown in Figure 6, will have the highest disk and Operational System Control performance like this.
In the present invention, owing to the management of showing to realize disk space by the bitmap binary tree, so the data security of bitmap binary tree table is necessary, comprises data security visit checking and data disaster recovery.
Can reserve some bytes in the node front portion of using when realizing data security visit checking, be used to preserve authorization information to be different from idle node.In when visit, can by authorization information determine this node state whether normally, whether can normally use, avoid distributing the node that has used or reclaim original just idle node.In the present invention, authorization information is also represented the size of this node except the user mode of expression node, and realization can produce unique secure access identifying code by information such as node size and node address.
In the present invention, except bitmap y-bend numerical table is the buffer memory character, all operations all acts on physical disk, so disaster only can have influence on this bitmap binary tree table, only may produce the incorrect of bitmap y-bend numerical table, so the data disaster recovery can only be shown to do at the bitmap binary tree.
For close the disaster that causes because of program exception, because bitmap y-bend numerical table has used the disk file mapping, operating system is kept the corresponding relation of mirror image and disk file, even program exception is closed, also can keep safeguarding the synchronous of bitmap binary tree table, therefore need not to recover for this kind disaster by operating system.
For the disaster that causes because of the collapse of system crash or disk physics, in one embodiment of the invention, solve in the following way:
At the disk management top of file disaster sign is set, during system start-up this disaster sign is loaded into internal memory together with bitmap binary tree table by the disk management File mapping, then by the disk management File mapping, the disaster sign of the disk management file on the disk is set to " disaster generation ", and the disaster sign of its memory mirror is set to " no disaster ", artificial generation is asynchronous, so just can whether recognize synchronously whether disaster takes place by both.When first kind of disaster took place, operating system can keep that both are synchronous, so the disaster sign of disk management file can be updated to " no disaster ", so can not produce erroneous judgement to this kind disaster.When second kind of disaster takes place, this moment, mirror image can't be returned disk synchronously, because during data security visit checking, the authorization information of node is except the state that shows node and whether use, also expressed its size, and every type the address of node is all fixed in the buddy algorithm, only needing the length 4K with minimum node is sweep spacing, disk space is carried out single pass, all nodes that just can binary tree traversal, (reference position of all nodes is bunch alignment all with the value of correcting the bitmap binary tree, and on the multiple position of 4K), upgrade the structure of bitmap binary tree table, make bitmap binary tree table distribute operating position to conform to, reach the purpose of disaster recovery with the disk space of reality.
Fig. 7 shows the realization flow of the disaster recovery in initialization procedure provided by the invention, and details are as follows:
In step S601, the disk management file load is arrived disk;
In step S602, judge that this disk management file is new file, if there is not this document on the disk, show that this document is new file, execution in step S603 is if existed this document, then execution in step S604 on the disk;
In step S603, the Header of this document is partly carried out initialization, the disaster sign is set, set up bitmap binary tree table;
In step S604, the Header part mapping to internal memory, is established mirror image;
In step S605, the disaster sign of chkdsk management document, if find to have taken place disaster according to the disaster sign, execution in step S606 then, otherwise execution in step S607;
In step S606, bitmap binary tree table is carried out aforesaid disaster recovery;
In step S607, the disaster sign is set, make that the disaster sign of the disaster sign of disk management file and memory mirror is asynchronous.
When the disk run duration, when the dynamic input and output of data take place, need the instant node information of adjusting, upgrade bitmap binary tree table, as shown in Figure 8:
When data stream is input to disk, need search bitmap binary tree table and distribute node, and the structure of bitmap binary tree table is upgraded, the header information of node is set simultaneously so that store data, comprise node size and node address, set up the node address index.
When needs use data in the node, for example data output or when deleted, program will be checked node address according to the node address index, find the corresponding node that needs to use data.After data were used, node became the free time, needed to upgrade bitmap binary tree table synchronously, with the user mode of real time record disk.
Fig. 9 shows the composition structure of disk space management provided by the invention system, describes easylier in order to make, content described above repeats no more.
Application program or operating system call external interface module 901, write or sense data to disk, data read-write module 902 writes disk with data or the data in the disk is read.
Bitmap binary tree table administration module 903 is responsible for maintenance update bitmap binary tree table, thereby make the bitmap binary tree table and the actual user mode of disk be consistent, when data write disk, search bitmap binary tree table, for writing, data distribute corresponding node, when data when disk is read, reclaim corresponding idle node.
Disaster flag management module 904 is used to be provided with the disaster sign of disk management file and memory mirror.The disaster sign that disaster recovery administration module 905 is provided with according to disaster flag management module 904 carries out disaster recovery to bitmap binary tree table when judging the generation disaster.
Memory mirror mapping block 906 is responsible for bitmap binary tree table and disaster tag maps are arrived internal memory, as the memory mirror operation, to improve the I/O performance of disk space management.
The above only is preferred embodiment of the present invention, not in order to restriction the present invention, all any modifications of being done within the spirit and principles in the present invention, is equal to and replaces and improvement etc., all should be included within protection scope of the present invention.
Claims (8)
1, a kind of disk space management method is characterized in that, described method comprises the steps:
A., bitmap binary tree table is set, is used for the set membership between address, user mode and the space piece of record space piece;
B. utilize described bitmap binary tree table that the space piece is distributed and reclaim.
2, disk space management method as claimed in claim 1 is characterized in that, described step B further comprises:
When data input disk, search described bitmap binary tree table, distribute space piece with the described data of storage;
When the data output disk, reclaim corresponding free space piece;
Distribution or recovery state according to the space piece upgrade described bitmap binary tree table, keep the structure of described bitmap binary tree table consistent with the user mode of disk space.
3, disk space management method as claimed in claim 1 is characterized in that, described method further comprises:
C. described bitmap binary tree table is mapped to internal memory, the disk operating of space management is converted to internal memory operation.
4, disk space management method as claimed in claim 3 is characterized in that, described method further comprises:
D. when disaster takes place, recover described bitmap binary tree table.
5 disk space management methods as claimed in claim 4 is characterized in that described step D further comprises:
D1., disk disaster sign and internal memory disaster sign are set;
D2. whether judge whether to take place disaster synchronously according to described disk disaster sign and internal memory disaster sign, when judging the generation disaster, the scanning disk space travels through described bitmap binary tree table according to the actual user mode of disk space, and described bitmap binary tree table is carried out topology update.
6, a kind of disk space management system is characterized in that described system comprises:
The external interface module is used to receive the outside data of importing, and perhaps the data that disk is preserved are externally exported;
Data read-write module is used for the data that described external interface module receives are write disk, and perhaps the data that disk is preserved are exported by described external interface module;
Bitmap binary tree table administration module is used to be provided with bitmap binary tree table, and utilizes described bitmap binary tree table that the space piece is distributed and reclaim;
Described bitmap binary tree table is used for the set membership between address, user mode and the space piece of record space piece.
7, disk space management as claimed in claim 6 system is characterized in that described system further comprises:
The memory mirror mapping block is used for described bitmap binary tree table is mapped to internal memory, and the disk operating of space management is converted to internal memory operation.
8, disk space management as claimed in claim 6 system is characterized in that described system further comprises:
Disaster flag management module is used to be provided with disk disaster sign and internal memory disaster sign;
The disaster recovery administration module, be used for whether judging whether to take place disaster synchronously according to described disk disaster sign and internal memory disaster sign, when judging the generation disaster, the scanning disk space, actual user mode according to disk space travels through described bitmap binary tree table, and described bitmap binary tree table is carried out topology update.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2006100341124A CN100456264C (en) | 2006-03-02 | 2006-03-02 | Magnetic disk space management and managing system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2006100341124A CN100456264C (en) | 2006-03-02 | 2006-03-02 | Magnetic disk space management and managing system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101030165A true CN101030165A (en) | 2007-09-05 |
| CN100456264C CN100456264C (en) | 2009-01-28 |
Family
ID=38715532
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2006100341124A Active CN100456264C (en) | 2006-03-02 | 2006-03-02 | Magnetic disk space management and managing system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100456264C (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102023821A (en) * | 2010-12-16 | 2011-04-20 | 成都市华为赛门铁克科技有限公司 | Disc space management method and system |
| CN102521114A (en) * | 2011-11-30 | 2012-06-27 | 华中科技大学 | File system log storage system under virtualization environment |
| CN102929979A (en) * | 2012-10-17 | 2013-02-13 | 华为技术有限公司 | Method and device for locating page memory space |
| CN101676906B (en) * | 2008-09-18 | 2013-06-05 | 中兴通讯股份有限公司 | Method of managing memory database space by using bitmap |
| CN103685600A (en) * | 2013-12-11 | 2014-03-26 | 河海大学 | Network address port translation pool managing method based on mixing of RB tree and bitmap |
| CN104133970A (en) * | 2014-08-06 | 2014-11-05 | 浪潮(北京)电子信息产业有限公司 | Data space management method and device |
| CN104461642A (en) * | 2014-12-16 | 2015-03-25 | 浪潮电子信息产业股份有限公司 | Linux system based 4K technology hard disk performance improvement setting and performance detection method |
| CN104572656A (en) * | 2013-10-12 | 2015-04-29 | 华为技术有限公司 | Method and device for recycling space of disk mirror image |
| CN111078920A (en) * | 2019-11-22 | 2020-04-28 | 北京云宽志业网络技术有限公司 | Data processing method, device, equipment and computer readable storage medium |
| CN112380004A (en) * | 2020-11-04 | 2021-02-19 | 成都佰维存储科技有限公司 | Memory management method and device, computer readable storage medium and electronic equipment |
| CN113805816A (en) * | 2021-09-27 | 2021-12-17 | 济南浪潮数据技术有限公司 | Disk space management method, device, equipment and storage medium |
| US11449437B2 (en) | 2017-06-28 | 2022-09-20 | Arm Limited | Invalidation of a target realm in a realm hierarchy |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5490274A (en) * | 1993-06-30 | 1996-02-06 | Microsoft Corporation | Modified buddy system for managing disk space |
| US5732402A (en) * | 1995-02-10 | 1998-03-24 | International Business Machines Corporation | System and method for data space management using buddy system space allocation |
| CN1258713C (en) * | 2003-11-14 | 2006-06-07 | 清华大学 | Data distribution dynamic mapping method based on magnetic disc characteristic |
| CN1295616C (en) * | 2004-12-31 | 2007-01-17 | 北京中星微电子有限公司 | Method for realizing write-in memory |
-
2006
- 2006-03-02 CN CNB2006100341124A patent/CN100456264C/en active Active
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101676906B (en) * | 2008-09-18 | 2013-06-05 | 中兴通讯股份有限公司 | Method of managing memory database space by using bitmap |
| CN102023821A (en) * | 2010-12-16 | 2011-04-20 | 成都市华为赛门铁克科技有限公司 | Disc space management method and system |
| CN102521114A (en) * | 2011-11-30 | 2012-06-27 | 华中科技大学 | File system log storage system under virtualization environment |
| CN102929979A (en) * | 2012-10-17 | 2013-02-13 | 华为技术有限公司 | Method and device for locating page memory space |
| CN104572656A (en) * | 2013-10-12 | 2015-04-29 | 华为技术有限公司 | Method and device for recycling space of disk mirror image |
| CN104572656B (en) * | 2013-10-12 | 2019-02-19 | 华为技术有限公司 | A kind of method and device of disk mirroring space reclamation |
| CN103685600B (en) * | 2013-12-11 | 2016-08-17 | 河海大学 | The network address port conversion pool management method mixed based on RB tree and bitmap |
| CN103685600A (en) * | 2013-12-11 | 2014-03-26 | 河海大学 | Network address port translation pool managing method based on mixing of RB tree and bitmap |
| CN104133970A (en) * | 2014-08-06 | 2014-11-05 | 浪潮(北京)电子信息产业有限公司 | Data space management method and device |
| CN104461642A (en) * | 2014-12-16 | 2015-03-25 | 浪潮电子信息产业股份有限公司 | Linux system based 4K technology hard disk performance improvement setting and performance detection method |
| US11449437B2 (en) | 2017-06-28 | 2022-09-20 | Arm Limited | Invalidation of a target realm in a realm hierarchy |
| TWI784016B (en) * | 2017-06-28 | 2022-11-21 | 英商Ａｒｍ股份有限公司 | Invalidation of a target realm in a realm hierarchy |
| CN111078920A (en) * | 2019-11-22 | 2020-04-28 | 北京云宽志业网络技术有限公司 | Data processing method, device, equipment and computer readable storage medium |
| CN111078920B (en) * | 2019-11-22 | 2021-09-03 | 北京云宽志业网络技术有限公司 | Data processing method, device, equipment and computer readable storage medium |
| CN112380004A (en) * | 2020-11-04 | 2021-02-19 | 成都佰维存储科技有限公司 | Memory management method and device, computer readable storage medium and electronic equipment |
| CN112380004B (en) * | 2020-11-04 | 2023-06-13 | 成都佰维存储科技有限公司 | Memory management method, memory management device, computer readable storage medium and electronic equipment |
| CN113805816A (en) * | 2021-09-27 | 2021-12-17 | 济南浪潮数据技术有限公司 | Disk space management method, device, equipment and storage medium |
| CN113805816B (en) * | 2021-09-27 | 2024-02-23 | 济南浪潮数据技术有限公司 | Disk space management method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100456264C (en) | 2009-01-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101030165A (en) | Magnetic disk space management and managing system | |
| CN103186350B (en) | The moving method of mixing storage system and hot spot data block | |
| US8037112B2 (en) | Efficient access of flash databases | |
| KR100389867B1 (en) | Flash memory management method | |
| KR101376937B1 (en) | Describing and querying discrete regions of flash storage | |
| CN102779096B (en) | Page, block and face-based three-dimensional flash memory address mapping method | |
| CN100426259C (en) | Virtual access method of storage document data | |
| CN105389264B (en) | The method and system of garbage collection in storage system | |
| CN103838853B (en) | Mixed file system based on different storage media | |
| JP2015512604A (en) | Cryptographic hash database | |
| US10402317B2 (en) | Data storage device and flash memory control method | |
| JP2014099163A (en) | Method, system, and computer program product for hybrid table implementation using buffer pool as permanent in-memory storage for memory-resident data | |
| CN1862476A (en) | Super large capacity virtual magnetic disk storage system | |
| CN110018966A (en) | A kind of memory, storage system, host and data manipulation, rubbish recovering method | |
| KR20110093035A (en) | Apparatus for flash address translation apparatus and method thereof | |
| CN110968269A (en) | SCM and SSD-based key value storage system and read-write request processing method | |
| CN1845093A (en) | Attribute extensible object file system | |
| US8239427B2 (en) | Disk layout method for object-based storage devices | |
| CN1908937A (en) | FLASH document management method and system in FLASH memory | |
| KR100907477B1 (en) | Apparatus and method for managing index of data stored in flash memory | |
| CN101241469B (en) | Method and device for storing and reading data in embedded system | |
| CN1200354C (en) | Automatic compressing/decompressing file system and its compressing algorithm | |
| CN111241090A (en) | Method and device for managing data index in storage system | |
| CN1430148A (en) | Method of storing and accessing embedded database | |
| KR101465426B1 (en) | Extent mapping flash device |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20190808 Address after: 518057 Nanshan District science and technology zone, Guangdong, Zhejiang Province, science and technology in the Tencent Building on the 1st floor of the 35 layer Co-patentee after: Tencent cloud computing (Beijing) limited liability company Patentee after: Tencent Technology (Shenzhen) Co., Ltd. Address before: 2, No. 518044, East 410, SEG Science Park, Zhenxing Road, Shenzhen, Guangdong, Futian District Patentee before: Tencent Technology (Shenzhen) Co., Ltd. |
|
| TR01 | Transfer of patent right |