CN103577470A - File system and method for improving performances of web server - Google Patents
File system and method for improving performances of web server Download PDFInfo
- Publication number
- CN103577470A CN103577470A CN201210275779.9A CN201210275779A CN103577470A CN 103577470 A CN103577470 A CN 103577470A CN 201210275779 A CN201210275779 A CN 201210275779A CN 103577470 A CN103577470 A CN 103577470A
- Authority
- CN
- China
- Prior art keywords
- file
- disk
- data
- file system
- web server
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
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
- G06F16/172—Caching, prefetching or hoarding of files
Abstract
The invention provides a file system and a method for improving performances of a web server. The file system comprises a disc space; the disc space comprises a disc metadata area and a disc data area; the disc metadata area comprise a file system supper block, a file directory entry bitmap, a volume information array and a file directory entry; the disc data area comprises a plurality of constant-capacity volumes, wherein each volume comprises a plurality of same-capacity file disc blocks, and each file disc block is used for storing small files in the web server. According to the file system and the method for improving the performances of the web server, which are disclosed by the invention, the disc space management is simplified by aiming at the application environment where the web server saves a great number of small files, so that the file data is continuously stored, and disc fragments are avoided; the disc operation times in file data reading and writing processes are reduced, the API interaction times of the file system are reduced, and the performances of the file system are effectively improved; according to a Hash chain table array, the file positioning process is simplified and accelerated; an IO operation of the file system utilizes an asynchronous mode, so that the problem that the application program is blocked when the IO operation is synchronized is eliminated.
Description
Technical field
The invention belongs to computer hardware management domain, relate to a kind of file system and method, particularly relate to a kind of file system and method that promotes web server performance.
Background technology
In operating system, file system is in a consequence very, and along with the continuous lifting of computer hardware performance (such as CPU, bus bandwidth), hard disk becomes system performance bottleneck place just day by day.And file system is used for data on hyperdisk, hyperdisk data how efficiently, the performance of giving full play to disk unit is the target that file system is pursued always.
Early such as ext2 file system by disk according to fixed size as 1024 bytes, 2048 bytes, 4096 bytes etc. are divided into piece, file data storing is in piece, cause like this file that surpasses block size to deposit continuously, the file that is less than disk block size may cause serious disk fragments problem.
ResierFs by Hans Reiser and the Namesys of team exploitation thereof is good at processing small documents applied environment in the majority.ResierFs does not distribute inode to some small documents, but by these small documents packings, leaves in same disk piecemeal.And ReiserFS is that balanced tree is very remarkable on search performance based on Fast-Balance tree (balanced tree) search.
Silicon Graphics is that the high-performance log file system XFS of IRIX operating system exploitation is good at the large file of processing.It provides the allocation scheme based on extent, variable block length allows user's physical block size when format as required, from 512 bytes to 64KB not etc., it can reduce disk fragments problem under small documents applied environment as far as possible, postpone distribution mechanism can be to greatest extent storing documents data continuously, and can reduce disk fragments problem carrying the high performance while.
Above-mentioned traditional file systems adopts various thought to be optimized file system performance, but because needs are preserved inode for file, the data structures such as dentry, therefore before reading and writing data, need to carry out repeatedly extra disk operating at every turn, so just cannot guarantee to reduce disk fragments when file data is deposited continuously on disk as far as possible.First traditional file systems needed the metadata information of locating file before reading file data, and this is to be also kept on disk conventionally.Like this, before reading valid data, need repeatedly IO to operate the memory location of locating file.For small documents, this can reduce file system performance exponentially.Although can carry out buffer memory to partial document metadata, because the metadata information of current file system is too complicated, buffer memory should be not excessive.Therefore, the main time loss of read-write process, in head position process, has so just produced impact to small documents IO performance.Also having traditional I/O operation is synchronous operation, in synchronous IO request implementation, can block application program, can affect like this performance issue of application program.
Summary of the invention
The shortcoming of prior art in view of the above, the object of the present invention is to provide a kind of file system and method that promotes web server performance, for solve that prior art large file cannot deposit continuously at disk, the problem that small documents data can bring disk fragments and IO performance and application program capacity to reduce.
For achieving the above object and other relevant objects, the invention provides a kind of file system that promotes web server performance, described file system comprises that described disk space comprises disk meta-data region and data in magnetic disk district; Described disk meta-data region comprises file system superblock, file directory item bitmap, volume information array and file directory item; Described data in magnetic disk district comprises a plurality of fixed-size volumes; Wherein, each volume comprises the file disk piece that a plurality of sizes are identical, continuous, and described file disk piece is for depositing the small documents of web server.
Preferably, the size of described volume is adjustable, from 32MB, rises to 256MB.
Preferably, the size of described file disk piece is adjustable, from 4KB by doubly rising to 128KB.
Preferably, described file directory item comprises filename, file disk piece number and file size, and described file directory item is cached in and in internal memory, is organized into Hash linked list array.
Preferably, described file system also comprises the application programming interfaces after simplification, and the application programming interfaces after described simplification comprise mount application programming interfaces, unmount application programming interfaces, read application programming interfaces, write application programming interfaces, delete application programming interfaces and flush application programming interfaces.
The invention provides a kind of method that promotes web server performance, it is characterized in that: described method comprises:
Disk space is divided into disk meta-data region and data in magnetic disk district;
Described disk meta-data region is divided into file system superblock, file directory item bitmap, label information array and file directory item;
Described data in magnetic disk zoning is divided into a plurality of fixed-size volumes; Wherein, each volume is divided into the identical continuous file disk piece of a plurality of sizes.
Preferably, described method also comprises the information of searching in meta-data region; The described performing step of searching the information in meta-data region comprises:
Q1, the information in simplified-file meta-data region, a log file name, file disk piece number and file size;
Q2, simplifies application programming interfaces;
Q3, is all cached in file directory item among internal memory;
Q4, carries out Hash by filename and searches, and described file directory necklace is entered in Hash linked list array.
Preferably, described method also comprises asynchronous execution application program.
Preferably, when described application program is read application program, the process of asynchronous execution read application program comprises:
R1, calculates Hash according to filename;
R2 searches the file directory item of this document in Hash linked list array by Hash;
R3, whether interpretation searches successful; If searched unsuccessfully, return, file does not exist in this document directory entry; If searched successfully, continue to carry out R4;
R4, obtains the information such as disk block number of file according to the file directory item finding, and is that this read request creates IO appealing structure, IO appealing structure described in initialization; The form of IO appealing structure is:
IO asks numbering;
The start address of the data buffer comprising in IO request;
The length of data buffer in IO request;
Treat the sector number of read;
Treat the sector number of read;
Request completion status;
R5, arranges IO required parameter, and file system is according to the parameter that is the disk block initialization IO request of file allocation, and described parameter comprises that the disk block number of read is converted into sector number by disk block number conversion sector number;
R6, adds created IO appealing structure in the controller of described disk drive to, is about to the IO request queue that IO request inserts and waits for IO thread process; Return to success status.
Preferably, when described application program is write application program, the process of asynchronous execution write application program comprises:
R1 ' is file data distribute data space to be written;
R2 ', whether judgement distributes described data space successful;
R3 ', distributes described disk file directory entry;
R4 ', whether judgement distributes described disk file directory entry successful; If finish using in file directory item space, return to failure; For file data to be written distributes disk file directory entry space, according to disk file directory entry space bitmap information, distribute; For memory file directory entry of this document data configuration, and memory file directory entry described in initialization; Described memory file directory entry is inserted in Hash linked list array; For write request creates and initialization IO appealing structure; The form of this IO appealing structure is:
IO asks numbering;
The start address of the data buffer comprising in IO request;
The length of data buffer in IO request;
Treat the sector number of write;
Treat the sector number of write;
Request completion status;
R5 ', arranges described IO required parameter, and file system is according to the parameter that is the disk block initialization IO request of file allocation, and described parameter comprises that the disk block number of write is converted into sector number by disk block number conversion sector number;
R6 ', adds the initialized IO appealing structure creating in Magnetic Disk Controller to, returns to success status.
As mentioned above, a kind of file system and method that promotes web server performance of the present invention, has following beneficial effect:
1, simplify disk space, reduce continuously disk fragments when storing documents data;
2, reduce the time that required IO number of operations in each reading and writing data process and magnetic head move, promoted the IO performance of file system;
3, the obstruction of having eliminated the lower application program of synchronous IO operation, adopts asynchronous execution, and the efficiency of application program is higher.
Accompanying drawing explanation
Disk space form schematic diagram after the file system that Fig. 1 is shown as lifting web server performance of the present invention is divided;
Fig. 2 is shown as Hash linked list array structural representation in the file system of lifting web server performance of the present invention;
Fig. 3 is shown as the process flow diagram of searching information in meta-data region in the method for lifting web server performance of the present invention;
Fig. 4 is shown as the function reading process flow diagram of IO request asynchronous execution in the method for lifting web server performance of the present invention;
What Fig. 5 was shown as IO request asynchronous execution in the method for lifting web server performance of the present invention writes function process flow diagram.
Element numbers explanation
1 disk space
11 disk meta-data region
12 data in magnetic disk districts
111 file system superblocks
112 file directory item bitmaps
113 volume information arrays
114 file directory items
121 volumes
1211 file disk pieces
Embodiment
Below, by specific instantiation explanation embodiments of the present invention, those skilled in the art can understand other advantages of the present invention and effect easily by the disclosed content of this instructions.The present invention can also be implemented or be applied by other different embodiment, and the every details in this instructions also can be based on different viewpoints and application, carries out various modifications or change not deviating under spirit of the present invention.
Refer to accompanying drawing.It should be noted that, the diagram providing in the present embodiment only illustrates basic conception of the present invention in a schematic way, satisfy and only show with assembly relevant in the present invention in graphic but not component count, shape and size drafting while implementing according to reality, during its actual enforcement, kenel, quantity and the ratio of each assembly can be a kind of random change, and its assembly layout kenel also may be more complicated.
Below in conjunction with embodiment and accompanying drawing, the present invention is described in detail.
Embodiment mono-
A kind of file system that promotes web server performance that the present embodiment provides, as shown in Figure 1, described file system comprises disk space 1.Described disk space 1 comprises disk meta-data region 11He data in magnetic disk district 12, and described disk meta-data region 11 comprises file system superblock 111, file directory item bitmap 112, volume information array 113, file directory item 114, wherein, described file system superblock 111 is for the information of description document system, described file directory item bitmap 112 is used for illustrating message bit pattern by bit value, described volume information array 113 is for describing volume information, described file directory item 114 refers to that the file of file system preserves the data structure of some information of file, the packets of information that described file directory item is preserved has contained < filename, file disk piece number, the mapping relations of file size >, described file directory item 114 is cached in to memory group and is made into Hash linked list array, as shown in Figure 2, described Hash linked list array comprises N Hash chained list, for example, Hash[0], Hash[1], Hash[N], each Hash chained list comprises file directory item (dentry), described file directory item (dentry) log file name (fname), the piece number of file disk piece (blk), and file size (fsize), the form of described file directory item is different from the complex format of traditional file inode.
Described data in magnetic disk district 12 comprises a plurality of volumes 121, and the size of these volumes 121 is adjustable, and for example, the size of described volume rises to 256MB from 32MB; Meanwhile, described volume 121 comprises the file disk piece (blk) 1211 that a plurality of sizes are identical, continuous; Wherein, described file disk piece (blk) is for depositing continuously the small documents of web server, and the size of described file disk piece (blk) is adjustable, for example, from 4KB, presses doubly and increases until 128KB.
Described file system also comprises the application programming interfaces (API) after simplification, and the application programming interfaces after described simplification comprise mount application programming interfaces, unmount application programming interfaces, read application programming interfaces, write application programming interfaces, delete application programming interfaces and flush application programming interfaces.Carry file system before described mount application programming interfaces are used for described file system; Described unmount application programming interfaces for described file system finish using after unloading of file system; Described read application programming interfaces are for reading the file data of described file system; Described write application programming interfaces are used for to disk data writing; Described delete application programming interfaces are used for deleting the specific file of described file system; Described flush application programming interfaces are for being synchronized to disk by internal memory apoplexy involving the solid organs data block, and described dirty data piece, for former data, refers to be modified, with the different data of former data.
File system of the present invention, can greatly promote the performance of web server.This system, for the applied environment simplified-file systems of preserving small documents in web server, is simplified disk space more, and this just can guarantee that small documents can deposit continuously; Meanwhile, the large I of the piece of file disk described in volume (blk) is set dynamically as required according to file size, farthest reduces disk fragments problem.
Embodiment bis-
The invention provides a kind of method that promotes web server performance, described method comprises:
Described disk meta-data region is divided into file system superblock 111, file directory item bitmap 112, volume information array 113, file directory item 114; Wherein, described file system superblock 111 has been described the information of file system; Described file directory item bitmap 112 illustrates message bit pattern by bit value; Described volume information array 113 is described volume information; Described file directory item 114 is preserved the data structure of the information of file, and described file directory item 114 is organized into Hash linked list array, according to filename locating file memory location on disk space in described Hash linked list array;
Described data in magnetic disk district 12 is divided into a plurality of fixed-size volumes 121, meanwhile, described volume 121 is divided into the identical continuous file disk piece (blk) 1211 of a plurality of sizes; Wherein, described file disk piece (blk) can be deposited the small documents in web server continuously, for example, deposits the small documents such as webpage.
File system is before reading and writing of files data, need to search the information in meta-data region, common described information is to be kept on disk space, so need repeatedly IO operation so that the memory location of locating file data before file reading data, so described method also comprises the information of searching in meta-data region, as shown in Figure 3, realizing this process specifically comprises:
Q1, simplifies the information in meta-data region, simplified-file directory entry namely, a log file name, file disk piece number, and file size;
Q2, simplify application programming interfaces (API), only provide mount application programming interfaces, umount application programming interfaces, read application programming interfaces, write application programming interfaces, delete application programming interfaces and flush application programming interfaces these essential application programming interfaces (API); Wherein,
Described mount application programming interfaces are carry file system before described file system is used;
Described umount application programming interfaces unloading of file system after described file system is finished using;
Described read application programming interfaces read the file data in described file system;
Described write application programming interfaces are to data writing in disk;
Described delete application programming interfaces are deleted specific file in described file system;
Described flush application programming interfaces are synchronized to internal memory apoplexy involving the solid organs data block on disk.
Q3, after described file system initialization, by the metadata of All Files, namely the file directory item in metadata is all cached among internal memory;
Q4, searches by filename is carried out to Hash, and described file directory necklace is entered in Hash linked list array.
By described method, before each IO operation, location and locating file positional information are just very simple, only need in internal memory, pass through Hash linked list array fast finding, without producing extra IO request;
Traditional I/O operation is synchronous execution, so before IO operation, application program can get clogged, so the method for lifting web server performance provided by the invention also comprises asynchronous execution IO request, first calls application programming interfaces in file system and carries out required application program;
Then, file system is responsible for creating and the corresponding IO appealing structure of initialization; Then, IO required parameter is set; Finally, the IO appealing structure of establishment is inserted among the controller of bottom disk drive, file system can be returned successfully to described application program, after this request completes, the controller of described disk drive completes by interrupt notification file system requests, file system and then notify described application requests to complete, waits for that described application program further processes.
When required application program is read application program, as shown in Figure 4, the process of asynchronous execution read application program comprises:
R1, calculates Hash according to filename (fname);
R2, the file directory name of searching this document by Hash in Hash linked list array;
R3, whether interpretation searches successful; If searched unsuccessfully, return, file does not exist at this document directory name; Meanwhile, if searched successfully, continue to carry out R4;
R4, obtains the information such as file disk piece number according to the file directory item finding, and is that this read request creates IO appealing structure, and IO appealing structure described in initialization; The form of described IO appealing structure is:
{
IO asks numbering;
The start address of the data buffer comprising in IO request;
The length of data buffer in IO request;
Treat the sector number of read;
Treat the sector number of read;
Request completion status;
}；
R5, arranges IO required parameter, and file system is according to the parameter that is the disk block initialization IO request of file allocation, and described parameter comprises that the disk block number of read is converted into sector number by disk block number conversion sector number;
R6, adds the IO appealing structure of establishment in the controller of described disk drive to, and IO thread process is waited in the IO request queue that the request of being about to is inserted; Return to success status.
Embodiment tri-
The invention provides a kind of method that promotes web server performance, described method comprises:
Described disk meta-data region is divided into file system superblock 111, file directory item bitmap 112, volume information array 113, file directory item 114; Wherein, described file system superblock 111 has been described the information of file system; Described file directory item bitmap 112 illustrates message bit pattern by bit value; Described volume information array 113 is described volume information; Described file directory item 114 is preserved the data structure of the information of file, and described file directory item 114 is organized into Hash linked list array, according to filename locating file memory location on disk space in described Hash linked list array;
Described data in magnetic disk district 12 is divided into a plurality of fixed-size volumes 121, meanwhile, described volume 121 is divided into the identical continuous file disk piece (blk) 1211 of a plurality of sizes; Wherein, described file disk piece (blk) can be deposited the small documents in web server continuously, for example, deposits the small documents such as webpage.
File system is before reading and writing of files data, need to search the information in meta-data region, common described information is to be kept on disk space, so need repeatedly IO operation so that the memory location of locating file data before file reading data, so described method also comprises the information of searching in meta-data region, realize this process and specifically comprise:
Q1, simplifies the information in meta-data region, simplified-file directory entry namely, a log file name, file disk piece number, and file size;
Q2, simplify application programming interfaces (API), only provide mount application programming interfaces, umount application programming interfaces, read application programming interfaces, write application programming interfaces, delete application programming interfaces and flush application programming interfaces these essential application programming interfaces (API); Wherein,
Described mount application programming interfaces are carry file system before described file system is used;
Described umount application programming interfaces unloading of file system after described file system is finished using;
Described read application programming interfaces read the file data in described file system;
Described write application programming interfaces are to data writing in disk;
Described delete application programming interfaces are deleted specific file in described file system;
Described flush application programming interfaces are synchronized to internal memory apoplexy involving the solid organs data block on disk.
Q3, after described file system initialization, by the metadata of All Files, namely the file directory item in metadata is all cached among internal memory;
Q4, searches by filename is carried out to Hash, and described file directory necklace is entered in Hash linked list array.
By described method, before each IO operation, location and locating file positional information are just very simple, only need in internal memory, pass through Hash linked list array fast finding, without producing extra IO request;
Traditional I/O operation is synchronous execution, so before IO operation, application program can get clogged, so the method for lifting web server performance provided by the invention also comprises asynchronous execution IO request, first calls application programming interfaces in file system and carries out required application program;
Then file system is responsible for creating corresponding IO appealing structure, and this IO appealing structure of initialization;
Then, IO required parameter is set;
Finally, this IO appealing structure is inserted among the controller of bottom disk drive, file system can be returned successfully to described application program, after this request completes, the controller of described disk drive completes by interrupt notification file system requests, file system and then notify described application requests to complete, waits for that described application program further processes.
When required application program is write application program, as shown in Figure 5, the process of asynchronous execution write application program comprises:
R1 ' is file data distribute data space to be written;
R2 ', whether judgement distributes described data space successful; In internal memory, search the current volume that whether has suitable block size, for example the volume of the file allocation 4KB block size of 3KB size if had, is a piece of this document distribution in described volume; If there is no described volume, from free time volume, obtain one, and its block size is decided to be to suitable size, for example, the file of 3KB is decided to be 4KB by described volume (volume) block size, and from described volume, distributes a piece to deposit this document, if there is no described idle volume, failure, disk space is full;
R3 ', distributes described disk file directory entry;
R4 ', whether judgement distributes described disk file directory entry successful; If finish using in file directory item space, return to failure; Meanwhile, for file data to be written distributes disk file directory entry space, according to disk file directory entry space bitmap information, distribute; For memory file directory entry of this document data configuration, and memory file directory entry described in initialization; Described memory file directory entry is inserted in Hash linked list array; For write request creates and initialization IO appealing structure, the form of described IO appealing structure is:
{
IO asks numbering;
The start address of the data buffer comprising in IO request;
The length of data buffer in IO request;
Treat the sector number of write;
Treat the sector number of write;
Request completion status;
}；
R5 ', arranges described IO required parameter, and file system is according to the parameter that is the disk block initialization IO request of file allocation, and described parameter comprises that the disk block number of write is converted into sector number by disk block number conversion sector number;
R6 ', adds the IO appealing structure of establishment in Magnetic Disk Controller to, is about to described initialized IO request and is inserted in I0 request queue; Return to success status.
Protection scope of the present invention is not limited to the sequencing of above listed step, as long as reduce disk fragments, IO performance, application program capacity problem when small documents is deposited continuously in can solving in prior art according to system of the present invention, all belongs to protection scope of the present invention.
The disk space management of web server File system design of the present invention, minimizing required IO the number of file reading data and IO request asynchronous execution can guarantee that file data has reduced disk fragments as much as possible when depositing continuously under small documents applied environment, postpone distribution mechanism file data is deposited to greatest extent continuously, can reduce disk fragments when improving performance; Secondly, remove the system calls such as calling open/close that takes place frequently in small documents IO read-write process, reduced required IO number in reading and writing data process, significantly reduce the APL of file system, promoted the performance of file system; The application program of finally having eliminated under synchronous IO pattern is blocked, and the efficiency of application program is higher.。
In sum, the present invention has effectively overcome various shortcoming of the prior art and tool high industrial utilization.
Above-described embodiment is illustrative principle of the present invention and effect thereof only, but not for limiting the present invention.Any person skilled in the art scholar all can, under spirit of the present invention and category, modify or change above-described embodiment.Therefore, such as in affiliated technical field, have and conventionally know that the knowledgeable, not departing from all equivalence modifications that complete under disclosed spirit and technological thought or changing, must be contained by claim of the present invention.
Claims (10)
1. a file system that promotes web server performance, is characterized in that: described file system comprises disk space; Described disk space comprises disk meta-data region and data in magnetic disk district;
Described disk meta-data region comprises file system superblock, file directory item bitmap, volume information array and file directory item;
Described data in magnetic disk district comprises a plurality of fixed-size volumes; Wherein, each volume comprises the file disk piece that a plurality of sizes are identical, continuous, and described file disk piece is for depositing the small documents of web server.
2. the file system of lifting web server performance according to claim 1, is characterized in that: the size of described volume is adjustable, from 32MB, rises to 256MB.
3. the file system of lifting web server performance according to claim 2, is characterized in that: the size of described file disk piece is adjustable, from 4KB by doubly rising to 128KB.
4. the file system of lifting web server performance according to claim 1, is characterized in that: described file directory item comprises filename, file disk piece number and file size, and described file directory item is cached in and in internal memory, is organized into Hash linked list array.
5. the file system of lifting web server performance according to claim 1, it is characterized in that: described file system also comprises the application programming interfaces after simplification, the application programming interfaces after described simplification comprise mount application programming interfaces, unmount application programming interfaces, read application programming interfaces, write application programming interfaces, delete application programming interfaces and flush application programming interfaces.
6. a method that promotes web server performance, is characterized in that: described method comprises:
Disk space is divided into disk meta-data region and data in magnetic disk district;
Described disk meta-data region is divided into file system superblock, file directory item bitmap, label information array and file directory item;
Described data in magnetic disk zoning is divided into a plurality of fixed-size volumes; Wherein, each volume is divided into the identical continuous file disk piece of a plurality of sizes.
7. the method for lifting web server performance according to claim 6, is characterized in that: described method also comprises the information of searching in meta-data region; The described performing step of searching the information in meta-data region comprises:
Q1, the information in simplified-file meta-data region, a log file name, file disk piece number and file size;
Q2, simplifies application programming interfaces;
Q3, is all cached in file directory item among internal memory;
Q4, carries out Hash by filename and searches, and described file directory necklace is entered in Hash linked list array.
8. the method for lifting web server performance according to claim 6, is characterized in that: described method also comprises asynchronous execution application program.
9. the method for lifting web server performance according to claim 8, is characterized in that: when described application program is read application program, the process of asynchronous execution read application program comprises:
R1, calculates Hash according to filename;
R2 searches the file directory item of this document in Hash linked list array by Hash;
R3, whether interpretation searches successful; If searched unsuccessfully, return, file does not exist in this document directory entry; If searched successfully, continue to carry out R4;
R4, obtains the information such as disk block number of file according to the file directory item finding, and is that this read request creates IO appealing structure, IO appealing structure described in initialization; The form of IO appealing structure is:
IO asks numbering;
The start address of the data buffer comprising in IO request;
The length of data buffer in IO request;
Treat the sector number of read;
Treat the sector number of read;
Request completion status;
R5, arranges IO required parameter, and file system is according to the parameter that is the disk block initialization IO request of file allocation, and described parameter comprises that the disk block number of read is converted into sector number by disk block number conversion sector number;
R6, adds created IO appealing structure in the controller of described disk drive to, is about to the IO request queue that IO request inserts and waits for IO thread process; Return to success status.
10. the method for lifting web server performance according to claim 8, is characterized in that: when described application program is write application program, the process of asynchronous execution write application program comprises:
R1 ' is file data distribute data space to be written;
R2 ', whether judgement distributes described data space successful;
R3 ', distributes described disk file directory entry;
R4 ', whether judgement distributes described disk file directory entry successful; If finish using in file directory item space, return to failure; For file data to be written distributes disk file directory entry space, according to disk file directory entry space bitmap information, distribute; For memory file directory entry of this document data configuration, and memory file directory entry described in initialization; Described memory file directory entry is inserted in Hash linked list array; For write request creates and initialization IO appealing structure; The form of this IO appealing structure is:
IO asks numbering;
The start address of the data buffer comprising in IO request;
The length of data buffer in IO request;
Treat the sector number of write;
Treat the sector number of write;
Request completion status;
R5 ', arranges described IO required parameter, and file system is according to the parameter that is the disk block initialization IO request of file allocation, and described parameter comprises that the disk block number of write is converted into sector number by disk block number conversion sector number;
R6 ', adds the initialized IO appealing structure creating in Magnetic Disk Controller to, returns to success status.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210275779.9A CN103577470B (en) | 2012-08-03 | 2012-08-03 | A kind of file system and method for lifting web server performance |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210275779.9A CN103577470B (en) | 2012-08-03 | 2012-08-03 | A kind of file system and method for lifting web server performance |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103577470A true CN103577470A (en) | 2014-02-12 |
| CN103577470B CN103577470B (en) | 2017-11-21 |
Family
ID=50049268
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210275779.9A Expired - Fee Related CN103577470B (en) | 2012-08-03 | 2012-08-03 | A kind of file system and method for lifting web server performance |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103577470B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105549903A (en) * | 2015-12-07 | 2016-05-04 | 上海斐讯数据通信技术有限公司 | Easily-transplanted file system and method for managing data of external memory by file system |
| CN105573678A (en) * | 2015-12-17 | 2016-05-11 | 深圳市华讯方舟软件技术有限公司 | PostgreSQL block |
| CN106126555A (en) * | 2016-06-16 | 2016-11-16 | 北京翼辉信息技术有限公司 | A kind of file management method and file system |
| CN107357869A (en) * | 2017-07-03 | 2017-11-17 | 张宇 | A kind of file operation method suitable for cloud storage |
| CN109144420A (en) * | 2018-08-22 | 2019-01-04 | 深圳忆联信息系统有限公司 | More particle self-adaptive initial method, apparatus, computer equipment and storage medium |
| CN111221472A (en) * | 2019-12-26 | 2020-06-02 | 天津中科曙光存储科技有限公司 | Multi-block allocation strategy optimization method and system for disk space allocation |
| CN111400247A (en) * | 2020-04-13 | 2020-07-10 | 杭州九州方园科技有限公司 | User behavior auditing method and file storage method |
| CN112800005A (en) * | 2021-01-22 | 2021-05-14 | 中孚安全技术有限公司 | Deep inspection method, system, terminal and storage medium for file system |
| CN113296700A (en) * | 2021-04-20 | 2021-08-24 | 新华三信息技术有限公司 | Storage space management method and device and server |
| US11106588B2 (en) | 2017-11-28 | 2021-08-31 | International Business Machines Corporation | Deferred method of allocating disk space for lightning segments |
| CN114546248A (en) * | 2020-11-26 | 2022-05-27 | 中移物联网有限公司 | Memory file system and processing method of storage block |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5129088A (en) * | 1987-11-30 | 1992-07-07 | International Business Machines Corporation | Data processing method to create virtual disks from non-contiguous groups of logically contiguous addressable blocks of direct access storage device |
| CN1471082A (en) * | 2002-07-25 | 2004-01-28 | 深圳易拓科技有限公司 | Variable zoning method for disc drive |
| CN1661568A (en) * | 2004-02-24 | 2005-08-31 | 中国科学院声学研究所 | File system in device of recording and playing back sounds and images under embedded type environment |
| CN101446976A (en) * | 2008-12-26 | 2009-06-03 | 中兴通讯股份有限公司 | File storage method in distributed file system |
| CN102012866A (en) * | 2009-09-07 | 2011-04-13 | 英业达股份有限公司 | A storage server and the method thereof for dynamically adjusting its storage space size |
| CN102096722A (en) * | 2011-03-21 | 2011-06-15 | 成都市华为赛门铁克科技有限公司 | File storage method and device |
-
2012
- 2012-08-03 CN CN201210275779.9A patent/CN103577470B/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5129088A (en) * | 1987-11-30 | 1992-07-07 | International Business Machines Corporation | Data processing method to create virtual disks from non-contiguous groups of logically contiguous addressable blocks of direct access storage device |
| CN1471082A (en) * | 2002-07-25 | 2004-01-28 | 深圳易拓科技有限公司 | Variable zoning method for disc drive |
| CN1661568A (en) * | 2004-02-24 | 2005-08-31 | 中国科学院声学研究所 | File system in device of recording and playing back sounds and images under embedded type environment |
| CN101446976A (en) * | 2008-12-26 | 2009-06-03 | 中兴通讯股份有限公司 | File storage method in distributed file system |
| CN102012866A (en) * | 2009-09-07 | 2011-04-13 | 英业达股份有限公司 | A storage server and the method thereof for dynamically adjusting its storage space size |
| CN102096722A (en) * | 2011-03-21 | 2011-06-15 | 成都市华为赛门铁克科技有限公司 | File storage method and device |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105549903B (en) * | 2015-12-07 | 2019-06-25 | 上海斐讯数据通信技术有限公司 | The file system for being easy to transplant and its management method to external memory data |
| CN105549903A (en) * | 2015-12-07 | 2016-05-04 | 上海斐讯数据通信技术有限公司 | Easily-transplanted file system and method for managing data of external memory by file system |
| CN105573678A (en) * | 2015-12-17 | 2016-05-11 | 深圳市华讯方舟软件技术有限公司 | PostgreSQL block |
| CN105573678B (en) * | 2015-12-17 | 2018-11-09 | 深圳市华讯方舟软件技术有限公司 | A kind of PostgreSQL blocks |
| CN106126555A (en) * | 2016-06-16 | 2016-11-16 | 北京翼辉信息技术有限公司 | A kind of file management method and file system |
| CN106126555B (en) * | 2016-06-16 | 2019-03-26 | 北京翼辉信息技术有限公司 | A kind of file management method and file system |
| CN107357869A (en) * | 2017-07-03 | 2017-11-17 | 张宇 | A kind of file operation method suitable for cloud storage |
| US11106588B2 (en) | 2017-11-28 | 2021-08-31 | International Business Machines Corporation | Deferred method of allocating disk space for lightning segments |
| CN109144420B (en) * | 2018-08-22 | 2021-07-13 | 深圳忆联信息系统有限公司 | Multi-particle self-adaptive initialization method and device, computer equipment and storage medium |
| CN109144420A (en) * | 2018-08-22 | 2019-01-04 | 深圳忆联信息系统有限公司 | More particle self-adaptive initial method, apparatus, computer equipment and storage medium |
| CN111221472A (en) * | 2019-12-26 | 2020-06-02 | 天津中科曙光存储科技有限公司 | Multi-block allocation strategy optimization method and system for disk space allocation |
| CN111221472B (en) * | 2019-12-26 | 2023-08-25 | 天津中科曙光存储科技有限公司 | Multi-block allocation strategy optimization method and system for disk space allocation |
| CN111400247A (en) * | 2020-04-13 | 2020-07-10 | 杭州九州方园科技有限公司 | User behavior auditing method and file storage method |
| CN111400247B (en) * | 2020-04-13 | 2023-08-01 | 杭州九州方园科技有限公司 | User behavior auditing method and file storage method |
| CN114546248A (en) * | 2020-11-26 | 2022-05-27 | 中移物联网有限公司 | Memory file system and processing method of storage block |
| CN112800005A (en) * | 2021-01-22 | 2021-05-14 | 中孚安全技术有限公司 | Deep inspection method, system, terminal and storage medium for file system |
| CN113296700A (en) * | 2021-04-20 | 2021-08-24 | 新华三信息技术有限公司 | Storage space management method and device and server |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103577470B (en) | 2017-11-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103577470A (en) | File system and method for improving performances of web server | |
| US8782324B1 (en) | Techniques for managing placement of extents based on a history of active extents | |
| CN110825748B (en) | High-performance and easily-expandable key value storage method by utilizing differentiated indexing mechanism | |
| CN105302744B (en) | The invalid data area of Cache | |
| CN102541985A (en) | Organization method of client directory cache in distributed file system | |
| US20190042571A1 (en) | Update-Insert for Key-Value Storage Interface | |
| US8364751B2 (en) | Automated client/server operation partitioning | |
| CN104881466B (en) | The processing of data fragmentation and the delet method of garbage files and device | |
| US20100088349A1 (en) | Virtual file system stack for data deduplication | |
| CN102023810B (en) | Method and device for writing data and redundant array of inexpensive disk | |
| US20120290595A1 (en) | Super-records | |
| CN100424699C (en) | Attribute extensible object file system | |
| CN105027122A (en) | Compression and deduplication layered driver | |
| CN105117417A (en) | Read-optimized memory database Trie tree index method | |
| CN103020315A (en) | Method for storing mass of small files on basis of master-slave distributed file system | |
| CN103186350A (en) | Hybrid storage system and hot spot data block migration method | |
| US9542401B1 (en) | Using extents of indirect blocks for file mapping of large files | |
| CN104021145A (en) | Mixed service concurrent access method and device | |
| CN102890722A (en) | Indexing method applied to time sequence historical database | |
| US11042328B2 (en) | Storage apparatus and method for autonomous space compaction | |
| CN103176754A (en) | Reading and storing method for massive amounts of small files | |
| WO2014047159A1 (en) | Write cache sorting | |
| CN102915340A (en) | Expanded B+ tree-based object file system | |
| CN102411632B (en) | Chain table-based memory database page type storage method | |
| CN103942301B (en) | Distributed file system oriented to access and application of multiple data types |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20171121 Termination date: 20200803 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |