CN111694765A - Mobile application feature-oriented multi-granularity space management method for nonvolatile memory file system - Google Patents
Mobile application feature-oriented multi-granularity space management method for nonvolatile memory file system Download PDFInfo
- Publication number
- CN111694765A CN111694765A CN202010505367.4A CN202010505367A CN111694765A CN 111694765 A CN111694765 A CN 111694765A CN 202010505367 A CN202010505367 A CN 202010505367A CN 111694765 A CN111694765 A CN 111694765A
- Authority
- CN
- China
- Prior art keywords
- space
- block
- mobile application
- nvm
- linked list
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000007726 management method Methods 0.000 title claims abstract description 32
- 230000015654 memory Effects 0.000 title claims abstract description 29
- 238000000034 method Methods 0.000 claims abstract description 23
- 238000011084 recovery Methods 0.000 claims abstract description 9
- 238000002791 soaking Methods 0.000 claims abstract description 7
- 235000019580 granularity Nutrition 0.000 claims abstract 10
- 238000004064 recycling Methods 0.000 claims description 8
- 238000009434 installation Methods 0.000 claims description 5
- 238000007781 pre-processing Methods 0.000 claims description 4
- 239000012634 fragment Substances 0.000 abstract description 4
- 238000005299 abrasion Methods 0.000 abstract 1
- 230000008447 perception Effects 0.000 abstract 1
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000002085 persistent effect Effects 0.000 description 2
- 238000001816 cooling Methods 0.000 description 1
- 238000013467 fragmentation Methods 0.000 description 1
- 238000006062 fragmentation reaction Methods 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 235000003642 hunger Nutrition 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 230000037351 starvation Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F12/00—Accessing, addressing or allocating within memory systems or architectures
- G06F12/02—Addressing or allocation; Relocation
- G06F12/0223—User address space allocation, e.g. contiguous or non contiguous base addressing
- G06F12/023—Free address space management
- G06F12/0238—Memory management in non-volatile memory, e.g. resistive RAM or ferroelectric memory
- G06F12/0246—Memory management in non-volatile memory, e.g. resistive RAM or ferroelectric memory in block erasable memory, e.g. flash memory
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F12/00—Accessing, addressing or allocating within memory systems or architectures
- G06F12/02—Addressing or allocation; Relocation
- G06F12/0223—User address space allocation, e.g. contiguous or non contiguous base addressing
- G06F12/023—Free address space management
- G06F12/0253—Garbage collection, i.e. reclamation of unreferenced memory
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/18—File system types
- G06F16/188—Virtual file systems
Abstract
The invention discloses a multi-granularity space management method in a non-volatile memory file system facing mobile application characteristics. Managing the free storage space in the NVM dynamically according to the data access mode of the mobile application. The method provides three linked list data structures with different granularity management NVM spaces according to a mobile application data access mode, and selects blocks with proper granularity for files to expand the space size of the files through the idle linked list management of the three different granularities. In addition, according to the soaking degree space recovery method, the types of the linked lists to be inserted into the idle blocks are adjusted by using a combined splitting mode for the release blocks, so that the wear balance effect of the NVM under a mobile application scene is achieved. Compared with the traditional PMFS fixed block size and no-moving data feature perception space allocation management method, the method disclosed by the invention can effectively promote the writing performance of a file system, reduce fragments in a nonvolatile storage medium and balance the abrasion degree of the medium, so that the space utilization rate and the durability of the NVM are improved.
Description
Technical Field
The invention relates to the technical field of storage, in particular to a multi-granularity space management method of a non-volatile memory file system facing mobile application characteristics.
Technical Field
In recent years, with the development of smart mobile devices, the dramatic increase in the number and complexity of mobile applications has posed a significant challenge to conventional block-based storage systems, while also limiting the increase in mobile device performance. The emerging non-volatile memory (NVM) has the advantages of near DRAM access speed, high density and byte addressing capability, and is considered to be the next generation of persistent memory-level memory. However, NVMs also suffer from low write endurance, which can result in localized permanent physical damage to the NVM if a local write operation is overheated. NVM is applied to different memory levels, e.g. as cache, replacing DRAM, constituting heterogeneous memory. For the heterogeneous memory architecture, the NVM and the DRAM are both installed on a memory bus and can be accessed directly in units of bytes through a Load/Store instruction of the CPU. Many new memory file systems are designed to manage NVM to persist file data at the memory level.
PMFS is a set of persistent memory based file management systems. Unlike the traditional file system, the PMFS directly copies file data between the user program Buffer and the NVM, and avoids the overhead of the traditional file data that passes through an I/O storage stack and two data copies (the block device copies to the Page cache and then to the user program Buffer). In addition, the PMFS is compatible with POSIX access standard, and the existing application program can directly access the file data in the PMFS without any modification.
The advent of new non-volatile memories has shortened the storage hierarchy. The NVM-based storage system hierarchy includes a user application layer, a Virtual File System (VFS) layer, a file system layer, and a storage medium. As shown in fig. 1, for a write operation, a user submits a file system write request to the VFS layer of the operating system, and the VFS provides a uniform abstract interface for the underlying file system to support the use of different file systems, so that the VFS passes the write request to the PMFS, and the PMFS copies data from the user state Buffer into the NVM area of the corresponding file.
According to statistics of mobile application I/O access modes, random write operation accounts for 75% in write operation, wherein file write requests with the granularity smaller than 4KB account for 64%, and large files such as installation packages, audio and video files and the like are only written once. The I/O access features of mobile applications are significantly different from desktop applications. Therefore, for mobile devices, spatial management of PMFS has certain limitations on block size. After receiving a write command sent by the VFS, the PMFS allocates space according to pages with the size of 4KB, adds Inode corresponding to the pages into a constructed B-tree when allocating data pages or directory pages each time, updates the state of the Inode in a free table, and submits modified records in a log. If the specified file is deleted to recycle the space, the corresponding file Inode is found first, and when no more directory Inode points to the deleted file Inode, the PMFS deletes the file Inode and sets the Inode to be deleted. Once all the controls are deleted by the handle of the deleted Inode, the VFS will issue a request for space reclamation to the PMFS. However, for small files, such as 1kB files, allocating 4kB blocks results in 75% waste of space, while for photo, audio, video, etc. files are typically several MB, even hundreds of MB in size, and these files are typically written once. For these files, allocating space for the file using fixed-size blocks of 4KB requires multiple calls to the space management unit, limiting the performance of file writes. In a mobile device usage scenario, space management in PMFS has the following drawbacks:
a) native page allocation uses fixed-size blocks (4KB), and mobile device data features are mainly small file reads and writes, where a file write ratio of less than 4KB is 64%, which inevitably results in large internal fragmentation of NVM under mobile device usage environment and insufficient utilization of storage medium space.
b) The NVM has a limited write time, and native PMFS does not consider wear leveling, even in a mobile application scenario. The data characteristics of the mobile terminal are that large files (such as videos and installation packages) are often subjected to write operation once, and only read operation is performed subsequently; small files are written and erased at high frequency, which easily causes excessive writing in local areas of the NVM. The file read-write characteristics of the mobile terminal will cause extreme separation of the NVM cold-write area and the hot-write area, and finally the NVM will be permanently physically damaged because the hot-write area exceeds the write frequency limit.
PMFS currently supports only the x86 system architecture, with no version for the mobile architecture, nor for the mobile device data features.
Disclosure of Invention
The invention provides a mobile application characteristic-oriented multi-granularity space management method for a nonvolatile memory file system, which solves the problems of a large amount of internal fragments generated by an NVM (non-volatile memory) in a mobile equipment use scene and the use durability.
The invention provides the following scheme:
a multi-granularity space management method for a non-volatile memory file system facing mobile application features that a linked list with proper granularity is dynamically selected according to the characteristics of the I/O access mode of the mobile application to allocate space, the three linked list free space management method is used for managing, and blocks with proper granularity are selected for a file requested to be written to perform writing operation.
A multi-granularity space management method for a non-volatile memory file system facing mobile application features is characterized in that according to the heat-equalizing space recovery method, the types of linked lists to be inserted into idle blocks are selected by using a combined splitting mode for release blocks, and therefore wear balance of an NVM (non-volatile memory) in a mobile application scene is achieved.
Further, the method for dynamic space management includes the following operations:
a) and preprocessing each write request sent by the mobile application, and confirming the size of a written file.
b) And dynamically selecting a proper linked list space for the file request according to the I/O size of the file request so as to allocate a free storage block with proper granularity.
Further, the method for managing the free space of the three linked lists comprises the following operations:
a) managing an idle area of the NVM storage medium by a chained three-linked list;
b) the granularity of the memory unit of the NVM storage medium of the mobile equipment is divided into three classes, namely 8KB, 4KB and 2 KB;
c)8KB represents the cold zone, which is mainly used by large files (such as installation packages, video screens and the like);
d)4KB represents a hot zone, primarily used by smaller files (e.g., pictures in chat tools, emoticons, etc.);
e)2KB represents the hot zone, which is mainly used by small files (such as short messages, chatting words and the like);
further, the soaking degree space recovery method comprises the following operations:
1) preprocessing a file release request of a mobile equipment NVM storage medium, confirming the size of a released file, and then selecting different space recycling strategies.
2)8KB block group recycling strategy
a) If the file size is larger than 8KB, the freed 8KB block group is split into a proportion (e.g., 1:1) of 4KB and 2KB block groups.
b) Adding a 4KB block to the tail of a 4KB linked list managed by a three linked list;
c) adding a 2KB block to the tail of a 2KB linked list managed by a three linked list;
3)4KB and 2KB Block group reclamation strategies
a) Adding the just released 4KB block group or 2KB block group into a temporary list;
b) combining the 2KB block and the 4KB block into an 8KB block in a temporary list;
c) adding the generated 8KB block group to the tail of the 8KB linked list managed by the three linked lists;
4) the unbalanced block group correction strategy is to make a statistical correction to the block numbers of 8KB, 4KB and 2KB at intervals, if the number of a certain block is found to be obviously unbalanced, the number of different blocks is adjusted to make the number ratio of the three different blocks approximately return to a preset state.
Therefore, the invention provides a mobile application feature-oriented multi-granularity space management method for a nonvolatile memory file system. By utilizing a dynamic space management method, in the I/O operation issued by the mobile application, data blocks with proper granularity are dynamically allocated and selected for writing operation to the writing requests of files with different sizes, so that internal fragments generated by writing of the granularity (4KB) of the fixed block of the NVM are greatly reduced, and the space utilization rate of a storage medium is improved; by the soaking degree space recovery method, the cold writing block group of the NVM storage body is disassembled and redistributed to the hot writing area for heating, and the hot block group is recombined to the cold writing area for cooling, so that the overall writing loss balance of the storage body medium is realized, the possibility of local permanent physical damage of the equipment caused by local over-thermal writing is reduced, and the overall durability of the NVM is improved.
Description of the drawings:
FIG. 1 is a diagram of NVM-based storage system architecture
FIG. 2 is a management flow of dynamic space allocation of a file write request of a PMFS to an upper application according to an embodiment of the present invention
FIG. 3 is a three-linked list free storage space management scheme according to an embodiment of the present invention
FIG. 4 shows the heat uniformity space recovery method of the present invention
FIG. 5 is a diagram illustrating an unbalanced block group correction method according to the present invention
Detailed description of the invention
In order to make the objects, technical solutions and advantages of the present invention more apparent to those skilled in the art, the following will further describe embodiments of the present invention with reference to the accompanying drawings and examples, which are not intended to limit the embodiments of the present invention.
Example (b):
the invention provides a mobile application feature-oriented multi-granularity space management method for a nonvolatile memory file system. The method has universality in the memory file system based on the mobile equipment, and can be used in different memory file systems based on the mobile equipment.
The following is a detailed description of the multi-granularity space management method of the nonvolatile memory file system.
Hereinafter, taking the write operation of the PMFS file system as an example, how to improve the space utilization and endurance of the NVM storage device is analyzed.
Firstly, space is dynamically allocated to a write request issued by an upper layer application. As shown in fig. 2, the VFS layer interface of the upper layer issues a write content request, the data feature recognition model module performs model recognition on the write request, and the write area distributor dynamically distributes a destination block conforming to the data characteristics from the empty space managed by the three linked list according to the recognition result; if the size of the requested file is identified to be smaller than 2KB, extracting blocks from a 2KB free space managed by the three linked list and distributing the blocks to the file; if the file size of the request is identified to be between 2KB and 4KB, allocating a 4KB free block managed by the three linked list to the file; if the identified requested file size is greater than 4KB, then the 8KB free block of triple-linked list management is allocated for the file.
As shown in fig. 3, the blocks in the NVM storage medium are divided into a data area for storing data to be stored and a pointer area for pointing to the block address of the next area. The data areas are all free areas, and the pointers point to the next free block address. The idle blocks in the NVM are managed in a form of a three-linked list, a pointer of an 8KB head node points to a head node of 4KB, and a pointer of a head node of 4KB points to a head node of 2 KB; counters are arranged in head nodes of 8KB, 4KB and 2KB to record the number of blocks administered by the counters; the 8KB head node also points to the first 8KB free block, the pointer of the free block points to the next 8KB free block address, and the free blocks are managed in a chain mode; similarly, the head nodes of 4KB and 2KB point to the first free block managed by the head nodes of the self, and the free blocks are managed in a chain mode. According to the data characteristics of the mobile terminal, the 8KB block is a cold writing area, most of the stored files are files such as a video screen, an installation package and the like, the files are large, and the updating frequency is lowest; the 4KB block is a medium writing area, such as an emoticon and a picture in a chat tool, a temporary cache map in a browser and the like, and the updating frequency is high; 2KB block query hot writing areas, such as characters, voice and the like in a chat tool, and the writing frequency is high;
when the file is released, the space management module is called to trigger the soaking degree space recovery method. The method improves the writing times of the cold writing area by disassembling the cold writing area and redistributing the cold writing area to the hotter writing area and the hot writing area; the hot area and the hot area are recombined into the cold writing area to reduce the writing times of the hot writing area, so that the wear of the NVM reaches a relatively balanced state on the whole, and the durability of the NVM storage medium is improved on the whole. The following describes the soaking space recovery method in detail:
as shown in fig. 4, the soaking degree space recycling method is performed with different recycling strategies for 8KB block groups and 4KB and 2KB block groups. The released 8KB block group is characterized by continuous space and large-scale release, so the cold-writing partition reallocation principle is used for splitting the 8KB block group into a 4KB block group and a 2KB block group, and the generated block groups are respectively managed by a 4KB linked list and a 2KB linked list which are managed by a triple linked list space. The released 2KB and 4KB blocks are characterized by discontinuous space and discrete release, so the released blocks are added into a temporary linked list queue, the blocks are recombined into 8KB blocks, then the 8KB blocks are placed in a finished product area of a recombination result, and the 8KB linked list managed by the space of the three linked lists is used for taking over management in time; the remaining blocks that are less than 8KB and not successfully reassembled continue to wait for new blocks to be released and added to the temporary linked list queue for the next reassembly.
As shown in fig. 5, the block group modification strategy is to avoid starvation of other block groups due to unbalance of the number of a certain block group when the system runs the dynamic space allocation method and the heat uniformity space recycling method. We refer to a certain block group with a large number as block number imbalance; at the beginning of system operation, the block group proportion of 8KB, 4KB and 2KB is set to be 1:2:2 according to the characteristics of data at a mobile terminal and the proportion of different file sizes, wherein the proportion is a suggested proportion and can be dynamically adjusted according to actual conditions; under the guidance ratio of 1:2:2, when the number of 4KB block groups is unbalanced, the corrector merges 1)4KB blocks into 8KB blocks and 2)4KB blocks into 2KB blocks, and adds the 2KB blocks into corresponding three-linked-list jurisdictions; when the 8KB block group is unbalanced, the 8KB block is mainly split into 4KB and 2KB blocks and is respectively added into the domain under the 4KB and 2KB three-linked list; when the block group of 2KB is unbalanced, the blocks are combined into blocks of 4KB and 8KB, and the blocks are added into a corresponding linked list; the goal of the block group correction strategy is to blur-correct the 8KB, 4KB, 2KB block group number ratio to approximately 1:2: 2;
based on the above description, the present invention performs space dynamic allocation management according to the data characteristics of the file written by the mobile device, that is, selects the target data block with a suitable size for writing operation, thereby greatly reducing the generation of internal fragments caused by the management of the free space of the original fixed block (4KB), and improving the space utilization of the NVM medium. According to the invention, through the heat-equalizing space recovery method, the next writing area of the released block is adjusted, and the wear imbalance of the NVM in a mobile application scene is balanced. The medium space utilization rate and the durability of the NVM under the use scene of the mobile equipment are improved on the whole.
The above-described embodiments are not intended to limit the present invention, and any modifications, equivalent substitutions, improvements, etc. made within the principle of the present invention should be included in the scope of the present invention.
Claims (5)
1. A multi-granularity space management method for a non-volatile memory file system facing mobile application features that a linked list with proper granularity is dynamically selected according to mobile application data access mode to allocate space, that is, a linked list data structure for managing NVM space with three granularities is provided according to mobile application data access mode, and blocks with proper granularity are selected for files to expand the space size of the files through the management of idle linked lists with three different granularities.
2. A multi-granularity space management method for a non-volatile memory file system facing mobile application features is characterized in that according to the heat-equalizing space recovery method, the types of linked lists to be inserted into idle blocks are adjusted by using a combined splitting mode for release blocks, and therefore wear balance of an NVM (non-volatile memory) in a mobile application scene is achieved.
3. A method for implementing the triple-linked list free space management of claim 1, characterized by:
a) the idle storage unit of the mobile equipment NVM storage medium is divided into three granularities according to the data characteristics of mobile application, 8KB, 4KB and 2 KB;
b)8KB represents the cold zone, which is mainly used by large files (such as installation packages, video screens and the like);
c)4KB represents a hot zone, primarily used by smaller files (e.g., pictures in chat tools, emoticons, etc.);
d)2KB represents the hot zone, which is mainly used by small files (such as short messages, chatting words and the like);
e) blocks of three granularities in the free storage space are managed by a three-linked list.
4. A method of implementing the dynamic space management of claim 1, wherein:
a) and preprocessing each I/O request sent by the mobile application, and confirming the size of the I/O request written into the file.
b) And dynamically selecting a proper linked list space for the file request according to the I/O size of the file request so as to allocate a free storage block with proper granularity.
5. The soaking degree space recycling method implemented according to claim 2 is used for adjusting the next writing area of the released block, balancing the wear imbalance of the NVM in a mobile application scene, and improving the durability of the NVM cut-off. The method is characterized by comprising the following operations: 1) preprocessing a file release request, confirming the size of the released file, and then selecting a different space recycling strategy 2)2)8KB block group recycling strategy
a) If the file size is larger than 8KB, splitting the block released by the 8KB block group into a certain proportion (such as 1:1) of 4KB and 2KB block groups;
b) adding a 4KB block to the tail of a 4KB linked list managed by a three linked list;
c) adding a 2KB block to the tail of a 2KB linked list managed by a three linked list;
3)4KB and 2KB Block group reclamation strategies
a) Adding the just released 4KB block group or 2KB block group into a temporary list;
b) combining the 2KB block and the 4KB block into an 8KB block in a temporary list;
c) adding the generated 8KB block group to the tail of the 8KB linked list managed by the three linked lists;
4) the unbalanced block group periodically corrects the strategy, the statistical correction is performed on the block numbers of 8KB, 4KB and 2KB at intervals, and if the number of a certain block is obviously unbalanced, the number of different blocks is adjusted to restore the number proportion of three different blocks to a preset state.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010505367.4A CN111694765A (en) | 2020-06-05 | 2020-06-05 | Mobile application feature-oriented multi-granularity space management method for nonvolatile memory file system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010505367.4A CN111694765A (en) | 2020-06-05 | 2020-06-05 | Mobile application feature-oriented multi-granularity space management method for nonvolatile memory file system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111694765A true CN111694765A (en) | 2020-09-22 |
Family
ID=72479451
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010505367.4A Pending CN111694765A (en) | 2020-06-05 | 2020-06-05 | Mobile application feature-oriented multi-granularity space management method for nonvolatile memory file system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111694765A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113297409A (en) * | 2021-06-30 | 2021-08-24 | 杭州海康威视数字技术股份有限公司 | Image searching method and device, electronic equipment and storage medium |
| CN114064588A (en) * | 2021-11-24 | 2022-02-18 | 建信金融科技有限责任公司 | Storage space scheduling method and system |
| CN114265562A (en) * | 2021-12-27 | 2022-04-01 | 北京国腾创新科技有限公司 | File storage method and system based on flash memory |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101183337A (en) * | 2007-12-12 | 2008-05-21 | 中兴通讯股份有限公司 | Space management techniques based on NAND FLASH mobile terminal storage medium |
| CN105184187A (en) * | 2015-08-18 | 2015-12-23 | 重庆大学 | Encrypting method and decrypting method of computer memorizer |
| CN106168883A (en) * | 2016-05-05 | 2016-11-30 | 诸葛晴凤 | A kind of efficient data tissue and access method |
| CN107562806A (en) * | 2017-08-08 | 2018-01-09 | 上海交通大学 | Mix the adaptive perception accelerated method and system of memory file system |
| CN109918317A (en) * | 2019-03-01 | 2019-06-21 | 重庆大学 | It is a kind of based on abrasion perception NVM item between abrasion equilibrium method |
| CN110287127A (en) * | 2019-05-14 | 2019-09-27 | 江苏大学 | A kind of Nonvolatile memory management method and system that more granularity multicores are expansible |
| CN110688345A (en) * | 2019-09-26 | 2020-01-14 | 重庆大学 | Multi-granularity structured space management mechanism of memory file system |
| CN110716694A (en) * | 2018-07-11 | 2020-01-21 | 三星电子株式会社 | Device and system for storage management |
| CN111159058A (en) * | 2019-12-27 | 2020-05-15 | 深圳大普微电子科技有限公司 | Wear leveling method and device and nonvolatile storage equipment |
-
2020
- 2020-06-05 CN CN202010505367.4A patent/CN111694765A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101183337A (en) * | 2007-12-12 | 2008-05-21 | 中兴通讯股份有限公司 | Space management techniques based on NAND FLASH mobile terminal storage medium |
| CN105184187A (en) * | 2015-08-18 | 2015-12-23 | 重庆大学 | Encrypting method and decrypting method of computer memorizer |
| CN106168883A (en) * | 2016-05-05 | 2016-11-30 | 诸葛晴凤 | A kind of efficient data tissue and access method |
| CN107562806A (en) * | 2017-08-08 | 2018-01-09 | 上海交通大学 | Mix the adaptive perception accelerated method and system of memory file system |
| CN110716694A (en) * | 2018-07-11 | 2020-01-21 | 三星电子株式会社 | Device and system for storage management |
| CN109918317A (en) * | 2019-03-01 | 2019-06-21 | 重庆大学 | It is a kind of based on abrasion perception NVM item between abrasion equilibrium method |
| CN110287127A (en) * | 2019-05-14 | 2019-09-27 | 江苏大学 | A kind of Nonvolatile memory management method and system that more granularity multicores are expansible |
| CN110688345A (en) * | 2019-09-26 | 2020-01-14 | 重庆大学 | Multi-granularity structured space management mechanism of memory file system |
| CN111159058A (en) * | 2019-12-27 | 2020-05-15 | 深圳大普微电子科技有限公司 | Wear leveling method and device and nonvolatile storage equipment |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113297409A (en) * | 2021-06-30 | 2021-08-24 | 杭州海康威视数字技术股份有限公司 | Image searching method and device, electronic equipment and storage medium |
| CN114064588A (en) * | 2021-11-24 | 2022-02-18 | 建信金融科技有限责任公司 | Storage space scheduling method and system |
| CN114265562A (en) * | 2021-12-27 | 2022-04-01 | 北京国腾创新科技有限公司 | File storage method and system based on flash memory |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR100874702B1 (en) | Device Drivers and Methods for Efficiently Managing Flash Memory File Systems | |
| JP3507132B2 (en) | Storage device using flash memory and storage control method thereof | |
| US6968424B1 (en) | Method and system for transparent compressed memory paging in a computer system | |
| CN101997918B (en) | Method for allocating mass storage resources according to needs in heterogeneous SAN (Storage Area Network) environment | |
| CN111694765A (en) | Mobile application feature-oriented multi-granularity space management method for nonvolatile memory file system | |
| US9792227B2 (en) | Heterogeneous unified memory | |
| CN102063406B (en) | Network shared Cache for multi-core processor and directory control method thereof | |
| US8694563B1 (en) | Space recovery for thin-provisioned storage volumes | |
| US9612975B2 (en) | Page cache device and method for efficient mapping | |
| JP2019079113A (en) | Storage device, data management method, and data management program | |
| US11144464B2 (en) | Information processing device, access controller, information processing method, and computer program for issuing access requests from a processor to a sub-processor | |
| CN103838853A (en) | Mixed file system based on different storage media | |
| CN114372007A (en) | Memory system and control method for controlling nonvolatile memory | |
| CN112000287A (en) | IO request processing device, method, equipment and readable storage medium | |
| CN110688345A (en) | Multi-granularity structured space management mechanism of memory file system | |
| US7434026B2 (en) | Disk array device and virtual volume management method using a logical table and a physical table | |
| CN111061652B (en) | Nonvolatile memory management method and system based on MPI-IO middleware | |
| CN110109677B (en) | Dynamic object cache pool allocation method | |
| CN111338569A (en) | Object storage back-end optimization method based on direct mapping | |
| JP6584529B2 (en) | Method and apparatus for accessing a file and storage system | |
| CN106407409A (en) | A virtual file system based on DAS architecture storage servers and a file management method thereof | |
| JP6254986B2 (en) | Information processing apparatus, access controller, and information processing method | |
| CN114518962A (en) | Memory management method and device | |
| CN113448722A (en) | Mapping method of process memory and instance processing method based on serverless architecture | |
| JP2017068805A (en) | Information processing apparatus, processor, and information processing method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200922 |