CN106326384A - File storage method suitable for high-speed mass storage based on FPGA (Field Programmable Gate Array) - Google Patents
File storage method suitable for high-speed mass storage based on FPGA (Field Programmable Gate Array) Download PDFInfo
- Publication number
- CN106326384A CN106326384A CN201610675496.1A CN201610675496A CN106326384A CN 106326384 A CN106326384 A CN 106326384A CN 201610675496 A CN201610675496 A CN 201610675496A CN 106326384 A CN106326384 A CN 106326384A
- Authority
- CN
- China
- Prior art keywords
- file
- bunch
- data
- fat
- dir
- 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
Classifications
-
- 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
- G06F16/17—Details of further file system functions
- G06F16/1727—Details of free space management performed by the file system
-
- 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
- G06F16/13—File access structures, e.g. distributed indices
Abstract
The invention relates to a file storage method suitable for high-speed mass storage based on FPGA (Field Programmable Gate Array). The file storage method comprises the following steps of equally dividing the whole storage capacity into C clusters, wherein each cluster contains S sectors, serial numbers of all the clusters are different, and bit width of the serial numbers is 16bit; managing file data by managing a cluster chain table FAT (File Allocation Table); sequentially combining data contained in the clusters corresponding to all the cluster chain tables to form a file; recording basic information of each file by managing a DIR (catalog data structure); recording basic information of a hardware file system by managing FSINFO (File System Information), including a magic number mark 0 x 55AA55AA for identifying the file system to be effective, the number of clusters contained in the file system as well as the number of sectors occupied by each cluster; saving the FAT, the DIR and the FSINFO in a 0th cluster in sequence. According to the file storage method suitable for the high-speed mass storage based on the FPGA, disclosed by the invention, the storage and management capabilities of mass data in an FPGA system in a CPU (Central Processing Unit)-free software execution environment are greatly improved, and application of a mass storage medium in the FPGA system is facilitated.
Description
Technical field
The present invention relates to data file technical field of memory, be specifically related to a kind of high speed magnanimity that is applicable to based on FPGA and deposit
The file memory method of storage.
Background technology
File system is the method for clear and definite storage device (disk, solid state hard disc etc.) storage data and data structure,
Data are managed in the form of a file, the method organizing data the most on a storage device.File system is responsible for setting storage
Standby space is organized and distributes, and is responsible for file storage and retrieves the file being stored in.Specifically, it is responsible for using
Family is set up, is stored in, reads, revises, deletes, is retrieved file etc..
But above-mentioned file system is integrated in an operating system, by counting usually used as the part of operating system software function
Calculation machine software (instruction) perform data management order, as the FAT file system in Windows system, new technology file system,
ExFAT file system, the ext file system operated in linux system, SWAP file system etc..But without operating system
In embedded system, particularly with in the FPGA system performing environment without CPU software, above-mentioned file system is due to its complete function
Complicated, by reasons such as CPU software execution, and FPGA hardware system need to be not suitable with, thus limited by application.
FPGA system is except being widely used in high-speed data process field simultaneously, and magnanimity high-speed data processing management function is also
There is extensive application demand.In tradition application, FPGA only achieves simple data management (as memory space is divided into fixed size region
For storing data), or no data management function (as directly by whole memory space sequential storage, again during storage data from
Section start covers legacy data).This method cannot realize effective management of file-type in similar computer system, as data are big
Little arbitrarily, data file memory space utilization rate large number of, high etc..
Summary of the invention
Present invention solves the technical problem that and be: for realizing magnanimity in the FPGA hardware system performing environment without CPU software
Data storage management, for the existing business of (but not limited to)/industry hard-disc storage standard (with sector as storage cell, every sector
512 bytes), it is provided that a kind of file memory method being applicable to high speed mass memory based on FPGA.
In order to solve above-mentioned technical problem, technical scheme is specific as follows:
A kind of file memory method being applicable to high speed mass memory based on FPGA, comprises the following steps:
Whole memory capacity is bisected into C bunch, and each bunch comprises S sector;Each bunch has different numberings, numbering position
Wide 16bit;
CAP=512 × C × S
Wherein C is 65533 to the maximum, and S is 16383 to the maximum;
File control data is carried out by a management bunch chained list FAT;Bunch data sequential combination comprised that each bunch of chained list is corresponding
Become a file;
The essential information of each file is recorded by administrative directory data structure DIR;
The essential information of this Hard disk file system is recorded by management FSINFO, the most unreal including mark file system
Count bunch number and each bunch of sector number taken indicating that 0x55AA55AA, this document system contain;
FAT, DIR and FSINFO are sequentially saved in the 0th bunch;
This document storage method has provided a user with following file operation interface: formats, create/write file, deletion
File, reading file, files listed;
Specifically:
Format: when using new hard disk or file system corruption, need to format hard disk, including by all DIR mesh
Record item empties, and shows do not have file;All for all FAT array members are labeled as 0, show all bunches unoccupied;
Create/write file: first search for idle DIR directory entry, simultaneously scan for empty bunch of the FAT next one;If without idle mesh
Record item or empty bunch of FAT, then create file failure;If containing the file identical with given file ID in catalogue, first deleting should
File;Otherwise according to creating fileinfo filling directory entry file ID and the time of establishment, and in new bunch, write data;Current new
Search again for next empty bunch after bunch filling up data, the most repeatedly, update the file size in DIR directory entry simultaneously;When search sky
Time bunch failed, showing that hard-disk capacity is exhausted, it is impossible to be stored in new data, file write is forcibly stopped;
Delete file: first search for DIR catalogue, obtain the starting cluster numbering of specified file ID, according to this linked list head, should
Chained list all the elements are labeled as empty bunch, complete the deletion of file;Respective file item in catalogue is removed simultaneously;If not searching finger
Determine file ID and then delete file failure;
Read file: in FAT, travel through chained list according to the file linked list head that catalogue obtains, read the data of each bunch, by
Walk whole file data to read;
List All Files: travel through DIR directory entry, if current directory item non-NULL, then export this directory entry;If current directory
For sky, then skip this directory entry;So operation is until last directory entry;
List specified file: search for DIR directory entry, if file ID is identical with given ID in current directory item, then export
This directory entry.
In technique scheme, in each bunch of multiple sectors comprised, each sector comprises 512 bytes.
In technique scheme, DIR maintains the essential information of file, for the Array for structural body of regular length;Including
File ID, establishment time, starting cluster and document size information.
In technique scheme, FAT is one-dimension array, and each member represents the next one bunch numbering of corresponding bunch, numbering
Bit wide 16Bit.
In technique scheme, after system electrification or reset, FAT and the DIR table data preserved in a hard disk are loaded into
In FPGA internal RAM, data are verified by loading procedure, if data check mistake occurs, then use another one backup
Tables of data;If Backup Data table still mistake, then file system crash, need to reinitialize file system;Loading terminate after etc.
Treat user instruction.
In technique scheme, when creating/write file, empty bunch searching algorithm is: one bunch of pointer of system maintenance,
It is initialized to 1 when powering on, is used for identifying current cluster;This pointer can be updated according to FAT when reading file;Write or establishment file
Time search for next nearest sky bunch backward from current pointer for storing data, and be marked in FAT table.
The present invention has a following beneficial effect:
The file memory method being applicable to high speed mass memory based on FPGA of the present invention, substantially increases without CPU soft
Part performs the FPGA system mass data storage operating capability of environment, facilitates large-capacity storage media answering in FPGA system
With.
Accompanying drawing explanation
With detailed description of the invention, the present invention is described in further detail below in conjunction with the accompanying drawings.
Fig. 1 be possess 30 bunches file distribution schematic diagram;
Fig. 2 is 128GB capacity HDDs, 32K bunch, safeguards that FAT and the DIR data structure of the file system of 1024 files is shown
It is intended to;
Fig. 3 is SFAT16 file system control flow schematic diagram;
Fig. 4 is SFAT16 file system function module frame chart schematic diagram.
Detailed description of the invention
A kind of file memory method being applicable to high speed mass memory based on FPGA, this document storage method does not support son
Catalogue, file identification uses 32Bit data as unique ID.
For convenience of management, whole memory capacity CAP being bisected into C bunch, each bunch comprises S sector.Typically, each
Sector comprises 512 bytes.Each bunch has different numberings, and (file system the most of the present invention is referred to as numbering bit wide 16bit
SFAT16)。
CAP=512 × C × S
Wherein C is 65533 to the maximum, and S is 16383 to the maximum.
File system, by management bunch chained list (i.e. file allocation table FAT), carrys out file control data.Each bunch of chained list correspondence
Bunch data sequential combination comprised become a file, bunch numbering specified by bunch chained list corresponding to file is not necessarily
Continuously.Fig. 1 show the file allocation table schematic diagram possessing 30 bunches, and bunch numbering specified by bunch chained list that file is corresponding does not connects
Continuous.
File system also needs to manage a directory data structure (DIR) simultaneously, is used for recording the basic letter of each file
Breath.
Additionally need to record the essential information (FSINFO) of this Hard disk file system, effective including mark file system
Bunch number that magic number mark 0x55AA55AA, this document system contain and each bunch of sector number taken.
FSINFO, FAT and DIR are sequentially saved in the 0th bunch, and the 0th bunch is not used in storage file data.
DIR maintains the essential information of file, for the Array for structural body of regular length.Including file ID, establishment time, rise
The information such as beginning bunch, file size.Wherein file ID is 4 byte datas, 0 represent this directory entry for sky, can be used for creating new file;
The establishment time is 4 byte datas;Starting cluster is this document linked list head place bunch numbering, is hereafter discussed in detail.File size is divided
For take complete bunch number, remaining data take complete sector (512 byte) number and again remaining data take 4 words
The number of joint.This representation facilitates data statistics of variables.Each DIR directory entry takies 16 bytes, as safeguarding
The file system of 1K file, this catalogue takies 16KB memory space.
FAT is one-dimension array, and each member represents the next one bunch numbering of corresponding bunch, numbering bit wide 16Bit.FAT array
If member 0, show that corresponding bunch is untapped bunch, can be selected for storing data when empty bunch of search.If-1, then should
Bunch it is last bunch of presents bunch chained list, the end of file.As a example by Fig. 1, this FAT containing 30 bunches, is wherein labeled as 0 altogether
For empty bunch, other blocks are belonging respectively to bunch chained list 1 and 2, represent 2 file sets.Wherein bunch chained list 1,2,3,4,7,12,16,20,
29 (ending clusters-1) constitute file 1;5,6,26,8 (ending cluster-1) configuration file 2.
FAT and DIR is stored in the 0th bunch, causes whole file system crash for preventing FAT and DIR from damaging, right
FAT and DIR has carried out data check and Double redundancy backup.FAT and DIR table in a hard disk will be preserved after system electrification or reset
Data are loaded in FPGA internal RAM, in loading procedure verify data, if there is data check mistake, then use another
An outer Backup Data table;If Backup Data table still mistake, then file system crash, need to reinitialize file system, i.e.
Format.Loading waits user instruction after terminating.
SFAT16 file system has provided a user with following file operation interface: formats, create (write) file, deletion
File, reading file, files listed.
A) format: when using new hard disk or file system corruption, need to format hard disk.Format and i.e. initialize
The process of file system, empties including by all DIR directory entries, shows do not have file;By all for all FAT array member's labellings
Be 0, show all bunches unoccupied.After formatting completes, hard disk no data file;
B) create/write file: first search for idle DIR directory entry, simultaneously scan for empty bunch of the FAT next one.If without the free time
Directory entry or empty bunch of FAT, then create file failure;If containing the file identical with given file ID in catalogue, first deleting
This document;Otherwise according to creating fileinfo filling directory entry file ID and the time of establishment, and in new bunch, write data.Currently
New bunch fill up data after search again for next empty bunch, the most repeatedly, update the file size in DIR directory entry simultaneously.Work as search
When sky bunch is failed, showing that hard-disk capacity is exhausted, it is impossible to be stored in new data, file write is forcibly stopped.The search of its hollow bunch is calculated
Method is: one bunch of pointer of system maintenance, is initialized to 1, is used for identifying current cluster when powering on.Can be according to FAT during reading file
Update this pointer;Write or search for next nearest sky bunch when creating file backward from current pointer for storing data, and
It is marked in FAT table.This searching method reduces the most occupied bunch of searched probability, improves search hit rate and speed
Degree.
C) delete file: first search DIR catalogue, obtain starting cluster numbering (linked list head) of specified file ID, according to this
These chained list all the elements are labeled as empty bunch (0), complete the deletion of file by linked list head.Simultaneously by respective file Xiang Qing in catalogue
Remove.If not searching specified file ID, delete file failure.Due to file not by real " erasing ", therefore delete file
Simple to operate rapidly;
D) file is read: in FAT, travel through chained list according to the file linked list head that catalogue obtains, read the data of each bunch,
It is gradually completing whole file data to read;
E) list All Files: traversal DIR directory entry, if current directory item non-NULL, then export this directory entry (i.e. file letter
Breath, including file ID, file size, creates the time);If current directory is empty, then skip this directory entry;So operation is until
Later directory entry;
F) specified file is listed: search DIR directory entry, if file ID is identical with given ID in current directory item, the most defeated
Go out this directory entry (i.e. fileinfo, including file ID, file size, creates the time).
Illustrate below in conjunction with the accompanying drawings:
Following G, M, K unit converts according to equation below:
1G=1024 × 1M
1M=1024 × 1K
1K=1024
As a example by the hard disk of 128GB, 256M sector, each sector 512B altogether.
Memory space be bisected into 4MB (8K sector) size bunch, 32K bunch altogether.And each member of FAT is 16bit
(2B), then FAT takes up room 64KB.
If file system can safeguard that 1024 files, each file directory item take 16B, then DIR takes up room 16KB.
This file system the 0th cluster data structure is as shown in Figure 2.
SFAT16 file system control flow is as shown in Figure 3.
According to above-mentioned workflow, SFAT16 file system function Module Division is following sections:
A) major state machine: for main flow control module, respond user file operation requests, other modules are assisted
Adjust, it is achieved control flow shown in Fig. 3;
B) FAT control module: response host state machine instruction, completes the search of FAT, labelling etc.;
C) DIR control module: response host state machine instruction, completes the search of DIR, rewriting etc.;
D) hard disk health status control module: periodically send to hard disk and read S.M.A.R.T. request, obtain hard disk healthy
State;
E) hard disk read control module: response host state machine instruction, completes hard disk read operation instruction and controls;
F) control module write by hard disk: response host state machine instructs, and completes hard disk write operation instruction and controls;
G) hard disk instruction switch: according to different operating, the hard disk operation above-mentioned disparate modules sent hard disk is asked
Switch over, formed unified to hard disk instruction interface.
SFAT16 file system function module frame chart is as shown in Figure 4.Wherein for improving system data read-write speed, it is right to increase
The support (RAID0) of disk array, makes data transfer rate be doubled and redoubled.
Obviously, above-described embodiment is only for clearly demonstrating example, and not restriction to embodiment.Right
For those of ordinary skill in the field, can also make on the basis of the above description other multi-form change or
Variation.Here without also cannot all of embodiment be given exhaustive.And the obvious change thus extended out or
Change among still in the protection domain of the invention.
Claims (6)
1. the file memory method being applicable to high speed mass memory based on FPGA, it is characterised in that comprise the following steps:
Whole memory capacity is bisected into C bunch, and each bunch comprises S sector;Each bunch has different numberings, numbering bit wide
16bit;
CAP=512 × C × S
Wherein C is 65533 to the maximum, and S is 16383 to the maximum;
File control data is carried out by a management bunch chained list FAT;Bunch data sequential combination comprised that each bunch of chained list is corresponding
It it is a file;
The essential information of each file is recorded by administrative directory data structure DIR;
The essential information of this Hard disk file system is recorded, including mark file system effective magic number mark by management FSINFO
Bunch number that will 0x55AA55AA, this document system contain and each bunch of sector number taken;
FAT, DIR and FSINFO are sequentially saved in the 0th bunch;
This document storage method has provided a user with following file operation interface: format, create/write file, delete file,
Read file, files listed;
Specifically:
Format: when using new hard disk or file system corruption, need to format hard disk, including by all DIR directory entries
Empty, show there is no file;All for all FAT array members are labeled as 0, show all bunches unoccupied;
Create/write file: first search for idle DIR directory entry, simultaneously scan for empty bunch of the FAT next one;If without idle directory entry
Or empty bunch of FAT, then create file failure;If containing the file identical with given file ID in catalogue, first deleting this document;
Otherwise according to creating fileinfo filling directory entry file ID and the time of establishment, and in new bunch, write data;Fill out for the newest bunch
Search again for next empty bunch after full data, the most repeatedly, update the file size in DIR directory entry simultaneously;When empty bunch of search is lost
When losing, showing that hard-disk capacity is exhausted, it is impossible to be stored in new data, file write is forcibly stopped;
Delete file: first search DIR catalogue, the starting cluster obtaining specified file ID is numbered, according to this linked list head, by this chained list
All the elements are labeled as empty bunch, complete the deletion of file;Respective file item in catalogue is removed simultaneously;If not searching appointment literary composition
Part ID then deletes file failure;
Read file: in FAT, travel through chained list according to the file linked list head that catalogue obtains, read the data of each bunch, the completeest
Whole file data is become to read;
List All Files: travel through DIR directory entry, if current directory item non-NULL, then export this directory entry;If current directory is
Sky, then skip this directory entry;So operation is until last directory entry;
List specified file: search for DIR directory entry, if file ID is identical with given ID in current directory item, then export this mesh
Record item.
The file memory method being applicable to high speed mass memory based on FPGA the most according to claim 1, its feature exists
In, in each bunch of multiple sectors comprised, each sector comprises 512 bytes.
The file memory method being applicable to high speed mass memory based on FPGA the most according to claim 1, its feature exists
In, DIR maintains the essential information of file, for the Array for structural body of regular length;Including file ID, establishment time, starting cluster
And document size information.
The file memory method being applicable to high speed mass memory based on FPGA the most according to claim 1, its feature exists
In, FAT is one-dimension array, and each member represents the next one bunch numbering of corresponding bunch, numbering bit wide 16Bit.
The file memory method being applicable to high speed mass memory based on FPGA the most according to claim 1, its feature exists
In, after system electrification or reset, FAT and the DIR table data preserved in a hard disk are loaded in FPGA internal RAM, loading procedure
In data are verified, if data check mistake occurs, then use another one Backup Data table;If Backup Data table is the most wrong
By mistake, then file system crash, need to reinitialize file system;Loading waits user instruction after terminating.
The file memory method being applicable to high speed mass memory based on FPGA the most according to claim 1, its feature exists
In,
When creating/write file, empty bunch searching algorithm is: one bunch of pointer of system maintenance, is initialized to 1, uses when powering on
In mark current cluster;This pointer can be updated according to FAT when reading file;Search for backward from current pointer when write or establishment file
Next nearest sky bunch is used for storing data, and is marked in FAT table.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610675496.1A CN106326384A (en) | 2016-08-16 | 2016-08-16 | File storage method suitable for high-speed mass storage based on FPGA (Field Programmable Gate Array) |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610675496.1A CN106326384A (en) | 2016-08-16 | 2016-08-16 | File storage method suitable for high-speed mass storage based on FPGA (Field Programmable Gate Array) |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106326384A true CN106326384A (en) | 2017-01-11 |
Family
ID=57740043
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610675496.1A Pending CN106326384A (en) | 2016-08-16 | 2016-08-16 | File storage method suitable for high-speed mass storage based on FPGA (Field Programmable Gate Array) |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106326384A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110543452A (en) * | 2019-08-07 | 2019-12-06 | 浙江大华技术股份有限公司 | data acquisition method and equipment |
CN111356094A (en) * | 2020-01-07 | 2020-06-30 | 北京简约纳电子有限公司 | SIM short message processing method and device |
CN111400259A (en) * | 2020-03-24 | 2020-07-10 | 中孚信息股份有限公司 | Directory content traversal method |
CN111427513A (en) * | 2020-03-25 | 2020-07-17 | 上海威固信息技术股份有限公司 | Method for improving storage performance of high-speed signal acquisition system |
WO2021248406A1 (en) * | 2020-06-11 | 2021-12-16 | 华为技术有限公司 | Method and apparatus for repairing file system |
CN115826884A (en) * | 2023-02-16 | 2023-03-21 | 湖南源科创新科技有限公司 | Lightweight high-speed file system facing embedded heterogeneous platform and processing method thereof |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1928869A (en) * | 2006-09-19 | 2007-03-14 | 珠海金山软件股份有限公司 | Method for deleting files in FAT roll |
CN101000614A (en) * | 2006-01-11 | 2007-07-18 | 富士通株式会社 | File delete method and file open method |
CN101042696A (en) * | 2006-03-23 | 2007-09-26 | 北京握奇数据系统有限公司 | Smart card storage system and managing method for file creation in the system |
CN101051317A (en) * | 2007-05-11 | 2007-10-10 | 北京中星微电子有限公司 | FAT file system and its processing method |
CN102426586A (en) * | 2011-10-25 | 2012-04-25 | 厦门雅迅网络股份有限公司 | FAT (file allocation table) file storage method in embedded system |
-
2016
- 2016-08-16 CN CN201610675496.1A patent/CN106326384A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101000614A (en) * | 2006-01-11 | 2007-07-18 | 富士通株式会社 | File delete method and file open method |
CN101042696A (en) * | 2006-03-23 | 2007-09-26 | 北京握奇数据系统有限公司 | Smart card storage system and managing method for file creation in the system |
CN1928869A (en) * | 2006-09-19 | 2007-03-14 | 珠海金山软件股份有限公司 | Method for deleting files in FAT roll |
CN101051317A (en) * | 2007-05-11 | 2007-10-10 | 北京中星微电子有限公司 | FAT file system and its processing method |
CN102426586A (en) * | 2011-10-25 | 2012-04-25 | 厦门雅迅网络股份有限公司 | FAT (file allocation table) file storage method in embedded system |
Non-Patent Citations (4)
Title |
---|
刘睿强: "《FPGA应用技术及实践》", 30 June 2016 * |
唐先余: "《实用DOS技术》", 31 October 1989 * |
张鹤峰: "《数字化办公管理与应用》", 31 March 2013 * |
李剑勇: "《数据恢复实训教程》", 28 February 2014 * |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110543452A (en) * | 2019-08-07 | 2019-12-06 | 浙江大华技术股份有限公司 | data acquisition method and equipment |
CN111356094A (en) * | 2020-01-07 | 2020-06-30 | 北京简约纳电子有限公司 | SIM short message processing method and device |
CN111356094B (en) * | 2020-01-07 | 2021-09-03 | 北京简约纳电子有限公司 | SIM short message processing method and device |
CN111400259A (en) * | 2020-03-24 | 2020-07-10 | 中孚信息股份有限公司 | Directory content traversal method |
CN111400259B (en) * | 2020-03-24 | 2023-04-21 | 中孚信息股份有限公司 | Method for traversing directory contents |
CN111427513A (en) * | 2020-03-25 | 2020-07-17 | 上海威固信息技术股份有限公司 | Method for improving storage performance of high-speed signal acquisition system |
CN111427513B (en) * | 2020-03-25 | 2023-02-28 | 上海威固信息技术股份有限公司 | Method for improving storage performance of high-speed signal acquisition system |
WO2021248406A1 (en) * | 2020-06-11 | 2021-12-16 | 华为技术有限公司 | Method and apparatus for repairing file system |
CN115826884A (en) * | 2023-02-16 | 2023-03-21 | 湖南源科创新科技有限公司 | Lightweight high-speed file system facing embedded heterogeneous platform and processing method thereof |
CN115826884B (en) * | 2023-02-16 | 2023-04-18 | 湖南源科创新科技有限公司 | Lightweight high-speed file system facing embedded heterogeneous platform and processing method thereof |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106326384A (en) | File storage method suitable for high-speed mass storage based on FPGA (Field Programmable Gate Array) | |
US8909887B1 (en) | Selective defragmentation based on IO hot spots | |
US9703717B2 (en) | Computer system and control method | |
US9274723B2 (en) | Storage apparatus and its control method | |
US8135907B2 (en) | Method and system for managing wear-level aware file systems | |
US7574560B2 (en) | Methods, systems, and computer program products for dynamic mapping of logical units in a redundant array of inexpensive disks (RAID) environment | |
US7584229B2 (en) | Method and system for priority-based allocation in a storage pool | |
CN103514249B (en) | A kind of automatic data reduction and system and storage device | |
US10521345B2 (en) | Managing input/output operations for shingled magnetic recording in a storage system | |
CN107526689B (en) | Read cache management | |
KR20110039417A (en) | Apparatus, system, and method for efficient mapping of virtual and physical addresses | |
KR20110048486A (en) | Apparatus, system, and method for coordinating storage requests in a multi-processor/multi-thread environment | |
JP2008015769A (en) | Storage system and writing distribution method | |
JP2013011919A (en) | Storage apparatus and snapshot control method of the same | |
TW201214270A (en) | Preserving an existing volume map in re-initializing a data storage volume | |
US8478933B2 (en) | Systems and methods for performing deduplicated data processing on tape | |
CN102779096A (en) | Page, block and face-based three-dimensional flash memory address mapping method | |
US11461287B2 (en) | Managing a file system within multiple LUNS while different LUN level policies are applied to the LUNS | |
WO2024077860A1 (en) | Metadata management method and apparatus, and computer device and storage medium | |
US20180307426A1 (en) | Storage apparatus and storage control method | |
CN109992204A (en) | Date storage method and device | |
CN101997919B (en) | Storage resource management method and device | |
CN105611979B (en) | Have hierarchical storage system, storage control and the medium of test area | |
US20220404987A1 (en) | Storage system, storage control device, and storage control method | |
US7437528B1 (en) | Gang blocks |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170111 |
|
WD01 | Invention patent application deemed withdrawn after publication |