CN101609430B - Memory Storage Space Management Method - Google Patents
Memory Storage Space Management Method Download PDFInfo
- Publication number
- CN101609430B CN101609430B CN 200810111461 CN200810111461A CN101609430B CN 101609430 B CN101609430 B CN 101609430B CN 200810111461 CN200810111461 CN 200810111461 CN 200810111461 A CN200810111461 A CN 200810111461A CN 101609430 B CN101609430 B CN 101609430B
- Authority
- CN
- China
- Prior art keywords
- storage device
- address
- file system
- storage space
- physical blocks
- 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.)
- Active
Links
Images
Abstract
A management method for memory storage space includes utilizing a resident program of loading operation system or utilizing controller in storage device to monitor storage space in storage device and resource distribution condition of file system, checking logic address of deleted invalid data and state of its correspondent entity block by using a logic/entity address mapping table, utilizing specific data deleting command to modify said mapping relation by controller, cancelling connection relation between said entity block and logic address, deleting said entity block to release stored space, storing checked position again finally and providing basis for next check.
Description
Technical field
The present invention relates to a kind of method for managing memory storage space, particularly utilize resident program monitoring storage area and file system to distribute the use situation, the entity memory region of usefulness not with in good time release busy.
Background technology
The relevant application system of general computing machine is often come accessing storing device by file system (file system), and the addressing method of file system normally the sector (Sector) take capacity as 512 bytes (Byte) as the logical block addresses (Logical Block Address, LBA) of unit as main.During storage device physical record data, then must carry out addressing with the physical address that can clearly define the deposit data position, such as: the addresses such as device (device), memory region, memory page.Requiring in application system to change between the physical address of the logical address of read/write and actual put data, is exactly logic/physical address translation function, that is Logical-to-Physical Address Translation.
Logic/physical blocks address corresponding relation wherein shows logical address (logical address, 11) as shown in Figure 1, and the address is by 0 to 2047999, the memory capacity of corresponding 512 bytes in each address (byte) in this example; Internal memory has more been distinguished a plurality of sections (segment), and segment 0 to segment 31 is arranged in this example, and logical address 0 to 63999 forms a section.In addition display logic block address (LBA, 12) and physical blocks address (PBA, 14) among the figure, between preserve the corresponding table of logic/entity (L2P Mapping Table, 13) correspondence with one.
In the upper example, each section is distinguished 256 logical block addresses (12), wherein keep 6 redundant blocks, in adding up to 256 physical blocks (physical block), corresponding 250 logical blocks (logical block), wherein keep 6 memory regions as redundant block (spareblock, free block), in order to record control data, for example: preserve the corresponding table of logic/entity (13), or be used for to replace faulty blocks (bad block), or when new data more, be used for recording more new data.Originally the physical blocks of save data is recorded to new position in Data Update, then replys after deletion action and includes available redundant block in.
From logic addressing space (Logical Address), add up to totally 32 sections (segment0～31), comprise 250 logical blocks in each section, each logical blocks corresponds to a physical blocks by corresponding table (13), and the data that can preserve 128KB; In each logical blocks, cooperate memory page (memory page) capacity of physical blocks, be divided into 64 continuous memory pages, each memory page can be held 2KB, contain 4 contiguous sectors (sector), that is each logical blocks contains the record space of continuous 256 sectors.
From entity addressing space (Physical Address), comprise 256 physical blocks in each section, each entity memory region comprises 64 memory pages, the data that can preserve 128KB; And can correspond to a logical blocks by corresponding table (13).The person that wherein do not define the corresponding relation is available redundant block, or because its bad characteristic is faulty blocks (bad block does not show among the figure), or as starting block (boot block) etc.
In the transformational relation of known logic addressing space shown in Figure 1 and physical address, each address in the logical address (11) represents a sector, can deposit user's data of 512 bytes, directly (each logical blocks capacity 128KB is equivalent to 256 sectors divided by unit block capacity, user's data of 128KB namely), getting its integral part is logical block addresses (Logical Block Address, 12), get its remainder divided by the quantity of unit memory page capacity representative sector.For instance, the memory page capacity 2KB of unit represents continuous 4 sectors, can calculate logical page address, as: logic addressing 100/256=0, corresponding logical block addresses (12) is 0, corresponding memory page address is the 25th memory page of 100/4=25; Logic addressing 256/256=1, corresponding logical block addresses (12) is 1, corresponding memory page address is 0.
So after application system is learnt logical block addresses (12) according to file system, namely need search above-mentioned logic/physical address corresponding table of conversion (13) and in table, detect the corresponding physical blocks of this logical blocks address (14).Be known as to save and deposit the corresponding table of conversion requisite space, generally correspondence table (13) is divided into a plurality of sections, each section is changed by a corresponding table (13).Each section is by a transformational relation of independently changing between the continuous logic addressing space of corresponding table record one and the continuous entity addressing space.For example: section 0 (Segment 0) record logic addressing 0～63999, LBA 0～249 add up to 250 logical blocks to be converted to the transformational relation that entity logic addressing 0～255 adds up between 256 entity memory regions in user's deposit data space of 32MB altogether; The rest may be inferred to section 31 (Segment 31) record logic addressing 1984000～2047999, and it also is that 250 logical blocks are converted to the transformational relation that entity logic addressing 7936～8191 adds up between 256 entity memory regions in user's deposit data space of 32MB altogether that LBA 7750～7999 adds up to.
The conversion that is addressed to the physical blocks addressing by logic is to adopt fixing continuous transformational relation, whereby by the logical blocks quantity of LBA divided by single section representative, can change easily and learn the sector number of depositing this logical blocks, and then loading obtains this actual physical blocks address (14) of depositing this logical blocks user data with the corresponding table of the conversion of searching this section.
Yet, general application system file system commonly used be for the magnetic Storage Media (for example: Winchester disk drive) designed, microsoft operation system (Windows for example
) commonly used FAT16, FAT32, NTFS etc., or Mac OS HFS, HFS+ etc. commonly used, when deleted file, it all directly directly removes the annotation of distributing in file configuration table (Allocation Table or Allocation Bit-Map), for example under the FAT file system, delete file operation is only deleted the file name of mark this document in root directory or the sub-directory, and the magnetic that this document takies the originally link structure (Cluster Chain) of gathering together is made zero.
And such operation all is to carry out the more write operation of new data for specific logic address (Logical Address) for known storage device, and can't learn according to this that data stored on the related entities address are invalid.For flash memory too, and more will cause when the renewal of carrying out valid data operates and carries out the censored mean operation, be subject to the impact that invalid data takies the entity memory region, produce and move the phenomenon of invalid data, and then the phenomenon that affects access speed and shorten the Storage Media life-span.
Summary of the invention
The technical problem to be solved in the present invention is, for may causing invalid data, the running that overcomes the known technology file system takies the problem of entity memory region, the present invention proposes a kind of method for managing memory storage space, resident program by the load operation system periodically or indicate the resource distribution situation of storage area and file system in the monitoring storage device according to the user, and when the logical address of the invalid data that is checked through deletion corresponds to a complete physical blocks, read first the corresponding table of logic/entity and correspond to the physical blocks address, and then produce the data delete instruction to the controller of storage device, then controller is carried out the step of the annexation between this physical blocks of cancellation and logical blocks, and then deletes this physical blocks to discharge the space that stores.The last storage again finished the position that checks file system, or then continues the inspection of the next position of execute file system according to above-mentioned program, discharges the space that invalid data occupies storage device.
The present invention then is applied to the flow process of above-mentioned deletion data the management of storage area, utilize the resident program of above-mentioned supervisory system internal memory and file system running, or utilize storage device controller directly to carry out monitoring, controller needs first identification contingency table and file system wherein before checking file system, if after meeting the kenel of support, be the hand inspection file system, according to the formal check logical blocks of contingency table, file system/page or leaf whether the entity memory region of unallocated use is arranged.
Then go to remove connection relationship between logical block addresses and physical blocks address by revising the corresponding table of L2P afterwards, and delete this physical blocks, the last position that this time checks that stores again so as to improving the access speed of storage device, and prolongs the term of life of Storage Media.
In other words, the invention provides a kind of method for managing memory storage space, described method includes:
The invalid data of monitoring storage area in the storage device and file system distributes situation;
The capacity of the physical blocks that corresponds to when the logical address of invalid data is during for a default unit capacity of this storage device;
Revise the corresponding table of this logic/physical address, to cancel the annexation between this physical blocks and logical address.
The present invention can discharge the shared storage area of invalid data, moves the situation of invalid data in the physical blocks when reducing Data Update and carrying out deletion, to prolong the term of life of memory storage.
Description of drawings
Fig. 1 is the logical and physical blocks address corresponding relation figure of known technology;
Fig. 2 utilizes resident program to check that file system discharges the operating process of available entity memory region;
Fig. 3 finishes to check operation system program afterwards;
Fig. 4 is logical address content synoptic diagram;
Fig. 5 is that the logical address of invalid data corresponds to the physical blocks synoptic diagram;
Fig. 6 is that the logical address block of invalid data corresponds to the physical blocks synoptic diagram;
Fig. 7 is the process flow diagram of deletion data in the method for managing memory storage space of the present invention;
Fig. 8 is method for managing memory storage space process flow diagram of the present invention.
[main element description of reference numerals]
Logical address 11 logical block addresses 12
Corresponding table 13 physical blocks of L2P address 14
Embodiment
Under the running of known file system, the operation of deleted file is usually only deleted the file name of mark this document in the catalogue, and the magnetic that this document takies the originally link structure of gathering together made zero, but not the entity memory space that actual deletion takies, but this mode will make executing data upgrade and write the fashionable state that can't learn the data that the related entities address is stored.
For avoiding or improving the problems referred to above, the present invention proposes a kind of method for managing memory storage space, mainly be to carry out a resident program by application system, can check in proper order that storage area that storage device provides and file system distribute the situation of using, and in the space of finding unallocated use, and when the capacity of this continuous space is the deletion unit capacity of storage device distribution use, assigns exclusive instruction notification storage device controller and discharge the entity memory region of distributing to this logical blocks originally.
Above-mentioned resident program is automatically to load after the application system start, the resource of monitoring at any time between storage area and file system is distributed, the entity memory region of usefulness not with in good time release busy, and utilize operating process that this resident program checks that file system discharges available entity memory region as shown in Figure 2.
After system's (such as computer system) starts (step S201), namely load a resident program (step S203) for supervisory system internal memory and file system running, can can check in general running that file system discharges available entity memory region in the storage device, comprise the file mark that checks under file configuration table (Allocation Table), each catalogue, or the gather together state of link structure (Cluster Chain) of each magnetic.This resident program will check the record whether front subsystem is arranged when starting, if start-up system is carried out resident program for the first time, then the load default value checks file system; If not start-up system is carried out resident program for the first time, then load the position of finishing last time and check file system with continuation.
Then system begins the built-in or plug-in storage device of detecting system (step S205), to confirm whether the controller on the storage device has the function (step S207) of supporting to discharge the entity memory region, be end operation (step S209) if do not support.Support this function if having after the detecting, then then capture the parameter (step S211) of storage device, comprise the information etc. of size, manufacturer or device of number, the memory page of memory page in the block size, block of storage device.
Then, (finished because of previous check process by next position in the file system, then carry out the inspection of next position) begin to check (step S213), and according to the formal check logical blocks of contingency table (partition table), file system/page or leaf whether the entity memory region (step S215) of unallocated use (free) is arranged, if the entity memory region without any unallocated use, then store this inspection record (step S219), such as the position that checks.
And if have the entity memory region of unallocated use, and when the capacity that continuous space is arranged is the deletion unit capacity of storage device distribution use, then assign instruction release and wherein be disabled the space (step S217) that data take, and store equally this inspection record (step S219).Namely finish the inspection operation and finish, this resident program then namely according to setting, checks the operation whether entity of unallocated use memory region is arranged within a certain period of time next time.
Fig. 3 is presented to finish and last time checks operation system program afterwards, above-mentioned resident program will carry out general judgment rule, comprise through the delay of certain hour or the instruction of operation suspension (step S301), do you judge that this computer system closes? (step S303), work will be closed, end operation (step S309) then, if do not close, does this resident program then then judge whether to continue operation? (step S305), if do not continue operation, then finish (step S309), if continue then to check operation (step S307) next time.
According to above-mentioned flow process, method for managing memory storage space of the present invention will be referred to the addressing method in the file system, the corresponding table of logic/entity (L2P mapping table) that comprises wherein logical address (LA) of utilization, physical blocks address (PBA), counterlogic block address (LBA) and physical blocks address, wherein, as shown in Figure 4 logical address content synoptic diagram.
Each address in the logical address represents a sector, part except the zone that corresponds to reservation, other mainly includes the address of system region (system area) 41 and 43 two sectors, data field (data area), it wherein is the sector that correspondence comprises contingency table, ROM-BIOS parameter block (BIOS Parameter Block, BPB), file system (such as FAT) etc. in the system region; The sector of corresponding storage data is then recorded in the data field, data be can gather together (cluster is assumed to be the size of 4K) come addressing for unit.The present invention utilizes inspection logical blocks/page or leaf to judge whether that the method for unallocated use need to consider the size of invalid data on the data field of logical address, and the corresponding relation of logical and physical address.
The logical address that please refer to again invalid data shown in Figure 5 corresponds to the physical blocks synoptic diagram, comprising the valid data 51,55 on the physical blocks, invalid data 53.In this example, be 2 gather together (cluster) according to invalid data (53) size in the INFORMATION DISCOVERY logical address data district of file system, the size that 8K is arranged, its logical address corresponds to a physical blocks address through above-mentioned L2P conversion table, namely occupy 4 memory pages (i.e. memory page page n among the figure, page n+1, page n+2, page n+3, each memory page 2K) space, 4 memory pages in the presentation-entity block are invalid data, but other memory page still is valid data in this physical blocks.In addition, take flash memory as example, the unit of wherein deleting data is block, and the size of invalid data may be less than physical blocks size in the storage device, so can not will also have the physical blocks deletion of valid data on the logical address.
Then, the logical address block of consulting again invalid data shown in Figure 6 corresponds to the physical blocks synoptic diagram, show three parts among the figure, include logical block addresses (LBA), the corresponding table of logic/entity (L2P) and physical blocks address (PBA), the corresponding table of this logic/entity namely is used for counterlogic block address and physical blocks address.
Invalid data (61 among the figure in the display logic address date district, the oblique line part) size has 32 to gather together, be total to the size of 128K, but this section logical block addresses corresponds to the physical blocks address through the corresponding table conversion of L2P, be assigned to two physical blocks 63 in the storage device, 65, in this example, two physical blocks 63, include respectively valid data and invalid data (oblique line part) in 65, since the least unit of deletion data be block (as in the FAT file system for 128K), the therefore above-mentioned physical blocks that has simultaneously valid data and invalid data of unsuppressible-suppression still.
Said method demonstrates, if invalid data and valid data are arranged in same physical blocks (block), then still can't delete data, so discharge the condition in the space of physical blocks, must be on checking logical address outside the size of invalid data, also need consider the physical blocks address that it is corresponding.
The flow process of deletion data as shown in Figure 7 in the method for managing memory storage space provided by the present invention, it mainly is the resource distribution situation of utilizing above-mentioned resident program periodically or indicating storage area and file system in the monitoring storage device according to the user, comprise the file mark that checks under a file configuration table, each catalogue, or the gather together state of link structure of each magnetic, distribute situation with the storage area of learning this storage device and the resource of file system.
When the logical address of the invalid data that deletion is arranged corresponds to a complete physical blocks, the capacity that continuous space is namely arranged be storage device distribute to use the deletion unit capacity time (step S701), represent that this block can be deleted, namely produce an exclusive data delete instruction to the controller (step S703) of storage device, after controller receives this data delete instruction, just carry out correlation step, comprise the annexation between this physical blocks of cancellation and logical blocks, and delete this physical blocks to discharge the space that stores, such as following detailed content.
Then, storage device controller is then when receiving the data delete instruction that application system execution resident program is assigned, check the specified logical blocks of this instruction (step S705), and read logic/entity correspondence and show (step S707), so as to corresponding to the physical blocks address, and judged whether the allocation entity memory region, and it still preserves deleted invalid data (step S709), if the specified logical blocks of order parameter does not have the allocation entity memory region, this EO then, and can wait until when the logical address that next time is checked through invalid data corresponds to a physical blocks, carry out next time flow process; If the specified logical blocks of above-mentioned order parameter has the allocation entity memory region, then controller is removed the connection relationship (step S711) of this logical blocks and this physical blocks by upgrading/revise the corresponding table of logic/entity.
Controller can be revised the record of this logical blocks correspondent entity block relation in above-mentioned corresponding table, to represent not allocation entity memory region of this logical blocks, also active data do not preserved in record, and this entity memory region is deleted (step S713).And, store and finish the position (step S715) that checks file system, or then continue the inspection of the next position of execute file system according to above-mentioned program, occupy the space of storage device to discharge invalid data.
Flow process can discharge the shared storage area of invalid data whereby, moves the situation of invalid data in the physical blocks when reducing Data Update and carrying out deletion, to prolong the term of life of memory storage.
Afterwards, the present invention more is applied to the memory storage space management process with the flow process of deletion data shown in Figure 7, step as shown in Figure 8.
After system starts, namely load the resident program that the present invention is used for supervisory system internal memory and file system running, this resident program distributes the operation of behaviour in service in order to check file system, but except this resident program, also can be automatically performed by the storage device controller of system and carry out the identification inspection, but the operation of carrying out identification at this storage device then must be limited to the file system that adopts when application system uses this storage device to format, in other words, this document system is that the controller in the storage device can identification use the interval whether to record the file system of valid data (comprise whether being allocated for the log file data, and whether file being deleted).
So, be connected to the storage device (step S801) of system in startup after, whether the record that last time checked is arranged, if having, then the load default value checks file system with checking; If not start for the first time, then can load the position of finishing last time.Afterwards, controller wherein needs first identification contingency table (step S803) before checking file system, identify after the configuration of storage device, identification file system (step S805) wherein again, and determine whether the file system (step S807) that controller is supported, if be not the file system (no) that controller is supported, then end operation; If but confirm as the file system format (being) of support, then then capture the parameter of storage device, comprise the information etc. of size, manufacturer or device of number, the memory page of memory page in the block size, block of storage device.Then begin again to check the use state of file system, begin (step S809) by an inferior position, comprise the file mark that checks under file configuration table, each catalogue, or/and the gather together state of link structure of each magnetic.Wherein may store check result last time because of before this time check process, so must be begun to check by next memory address.
The entity memory region (step S811) that then according to formal check logical blocks and the central memory page of contingency table, file system whether unallocated use is arranged, if the entity memory region (no) without any unallocated use, then to step S821, store this inspection record, such as the position that checks; If logical blocks/page or leaf has the entity memory region (being) of unallocated use, then read the corresponding table of logic/entity (step S813), so as to counterlogic block address and physical blocks address, judge whether the logical blocks allocation entity memory region of deleted document, it still preserves deleted invalid data (step S815).
In the determining step of step S815, if specified logical blocks does not have the allocation entity memory region, then carry out step S821 storing step and EO, until check process carries out next time flow process when finding to have the entity memory region of unallocated use.
But the logical blocks of judging deleted document in the step of step S815 still has the allocation entity memory region, if have and account for a whole physical blocks, then go to remove connection relationship (step S817) between logical block addresses and physical blocks address by revising the corresponding table of L2P, and delete this physical blocks (step S819).
At last, step will then store the position (step S821) that this time checks, and can reexamine the next position (step S809) of file system after the delay of a period of time (step S823), continue the memory management flow process of next stage.
Among the present invention, described deletion unit capacity is not limited to the deletion unit capacity of memory body itself, also can be the default capacity of user's self-defining.
In sum, the present invention is a kind of method for managing memory storage space, can utilize resident program monitoring storage area and the file system of load operation system to distribute the use situation, by the corresponding table of logical blocks in the revised file system with physical blocks, release busy and the entity memory region of usefulness not, can improve the access speed of storage device, and prolong the term of life of Storage Media.
The above only is better possible embodiments of the present invention, is not so limits to interest field of the present invention, so the equivalent structure that all utilizations instructions of the present invention and accompanying drawing content are done changes, all in like manner is contained within the scope of the invention hereby explanation.
Claims (8)
1. method for managing memory storage space is characterized in that described method includes:
The invalid data of monitoring storage area in the storage device and file system distributes situation;
The capacity of the physical blocks that corresponds to when the logical address of invalid data is during for a default unit capacity of this storage device;
Revise the corresponding table of logic/physical address, to cancel the annexation between this physical blocks and this logical address.
2. method for managing memory storage space as claimed in claim 1 is characterized in that, the invalid data of monitoring storage area in this storage device and file system more comprises before distributing situation:
The resident program of load operation system;
Check a file configuration table; And
Check the file mark under each catalogue, or the gather together state of link structure of each magnetic.
3. method for managing memory storage space as claimed in claim 1 is characterized in that, utilizes the controller of this storage device to carry out each step of this method for managing memory storage space.
4. method for managing memory storage space as claimed in claim 1, it is characterized in that the step of the corresponding table of this logic of described modification/physical address before, judge whether this logical address has the allocation entity block;
If this logical address has the allocation entity block, then controller is carried out the step of the corresponding table of this modification logic/physical address, cancelling the corresponding relation of this logical address and this physical blocks, and deletes this physical blocks to discharge the interval;
If this logical address does not have allocation entity block, then EO.
5. method for managing memory storage space as claimed in claim 2 is characterized in that, stores at last the position of finishing this time inspection.
6. method for managing memory storage space as claimed in claim 3 is characterized in that, the invalid data of monitoring storage area in this storage device and file system more comprises before distributing situation:
Start this storage device;
Judge whether this document system is the file system that this controller is supported;
If be not the file system that this controller is supported, then end operation;
If confirm as the file system format of support, then capture the step of storage device parameter.
7. method for managing memory storage space as claimed in claim 6 is characterized in that described storage device parameter comprises the number of memory page in the block size, block of this storage device, the size of memory page.
8. method for managing memory storage space as claimed in claim 6 is characterized in that, after starting this storage device, judgement comprises:
If start for the first time, then load a default value and check this document system;
If not start for the first time, then load the position of finishing last time and check this document system with continuation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200810111461 CN101609430B (en) | 2008-06-19 | 2008-06-19 | Memory Storage Space Management Method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200810111461 CN101609430B (en) | 2008-06-19 | 2008-06-19 | Memory Storage Space Management Method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101609430A CN101609430A (en) | 2009-12-23 |
| CN101609430B true CN101609430B (en) | 2013-03-20 |
Family
ID=41483189
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200810111461 Active CN101609430B (en) | 2008-06-19 | 2008-06-19 | Memory Storage Space Management Method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101609430B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102262592A (en) * | 2010-05-28 | 2011-11-30 | 多利吉科技股份有限公司 | Method for improving read-write speed of non-volatile memory device |
| CN102279881A (en) * | 2011-07-28 | 2011-12-14 | 南京中兴新软件有限责任公司 | Junk data clearing method and file access server |
| US9606813B2 (en) * | 2012-03-30 | 2017-03-28 | Canon Kabushiki Kaisha | Information processing apparatus, application activation method, and program |
| CN109508991A (en) * | 2018-10-16 | 2019-03-22 | 深圳市圆世科技有限责任公司 | A kind of edge collaboration method based on block chain |
| CN112699067A (en) * | 2021-01-04 | 2021-04-23 | 瑞芯微电子股份有限公司 | Instruction addressing method and device |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1124378A (en) * | 1994-04-19 | 1996-06-12 | 株式会社日立制作所 | Data processor |
| US5784698A (en) * | 1995-12-05 | 1998-07-21 | International Business Machines Corporation | Dynamic memory allocation that enalbes efficient use of buffer pool memory segments |
-
2008
- 2008-06-19 CN CN 200810111461 patent/CN101609430B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1124378A (en) * | 1994-04-19 | 1996-06-12 | 株式会社日立制作所 | Data processor |
| US5784698A (en) * | 1995-12-05 | 1998-07-21 | International Business Machines Corporation | Dynamic memory allocation that enalbes efficient use of buffer pool memory segments |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101609430A (en) | 2009-12-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR100823171B1 (en) | Computer system having a partitioned flash translation layer and flash translation layer partition method thereof | |
| JP5186367B2 (en) | Memory migration system and method | |
| US9672158B2 (en) | Asymmetric memory migration in hybrid main memory | |
| US8370602B2 (en) | Method for memory space management | |
| EP2791806B1 (en) | Working set swapping using a sequentially ordered swap file | |
| US9405476B2 (en) | Systems and methods for a file-level cache | |
| JP4749255B2 (en) | Storage system control device having multiple types of storage devices | |
| CN100501678C (en) | Controlling memory access in a multi-booting system | |
| JP5081498B2 (en) | Computer system and control method thereof | |
| US20150039837A1 (en) | System and method for tiered caching and storage allocation | |
| CN107209726B (en) | Flushing in a file system | |
| JP2009043030A (en) | Storage system | |
| US8244955B2 (en) | Storage system and its controlling method | |
| CN100498705C (en) | System and method of storing user data in a partition file or using a partition file containing user data | |
| EP2972747B1 (en) | Data storage, file and volume system providing mutliple tiers | |
| CN111194438B (en) | Extending SSD permanence | |
| KR20090046567A (en) | Solid state disk and operating method thereof | |
| CN107908571B (en) | Data writing method, flash memory device and storage equipment | |
| CN101609430B (en) | Memory Storage Space Management Method | |
| CN115617503A (en) | System and method for load balancing in heterogeneous memory systems | |
| KR101392062B1 (en) | Fast speed computer system power-on ＆ power-off method | |
| WO2013023090A2 (en) | Systems and methods for a file-level cache | |
| JP7125964B2 (en) | Computer system and management method | |
| JP2008299559A (en) | Storage system and data transfer method for storage system | |
| JP6100750B2 (en) | Device with memory and controller and device with data storage 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 |