CN101576912A - System and reading and writing method for realizing asynchronous input and output interface of distributed file system - Google Patents
System and reading and writing method for realizing asynchronous input and output interface of distributed file system Download PDFInfo
- Publication number
- CN101576912A CN101576912A CNA2009100858651A CN200910085865A CN101576912A CN 101576912 A CN101576912 A CN 101576912A CN A2009100858651 A CNA2009100858651 A CN A2009100858651A CN 200910085865 A CN200910085865 A CN 200910085865A CN 101576912 A CN101576912 A CN 101576912A
- Authority
- CN
- China
- Prior art keywords
- aio
- controll block
- asynchronous
- write
- read
- 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
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a system and a reading and writing method for realizing asynchronous input and output interface of a distributed file system. The system for realizing the asynchronous input and output interface of the distributed file system comprises a dynamic link library which is embedded in an application progress address space for running and provides the asynchronous input and output IO interface invoked by the application process, a memory sharing module which is connected with the dynamic link library in a communication way; the memory sharing module comprises at least one asynchronous input and output AIO control block; and the memory sharing module realizes the submission and completion of the asynchronous reading and writing requirement by the AIO control block. The method realizes asynchronous IO interface on the distributed file system by the memory sharing mechanism, solves the performance problem of the POSIX standard asynchronous IO interface, and improves the large throughout reading and writing performance of the distributed file system.
Description
Technical field
The present invention relates generally to data storage and visit field, is meant a kind of system and asynchronous read and write method that realizes the asynchronous IO interface of distributed file system especially.
Background technology
Distributed file system is widely used in professional occasions such as IPTV (Internet Protocol Television, IPTV), streaming media server.Ordinary business practice uses synchronous block type IO (Input/Output, the I/O) interface of distributed file system to carry out the read-write of file.When carrying out a large number of services operation, a thread need be handled a plurality of service request, therefore need read while write a plurality of files, and the IO interface of block type is easy to cause once reading and writing overtime influence read-write repeatedly synchronously, in order to address this problem, need distributed file system that the ability of asynchronous IO is provided.
The asynchronous IO of POSIX (Portable Operating System Interface, portable operating system interface) definition realizes by the built-in function of user's space, therefore for the IO of bottom, still is in fact synchronous mode.When a process is initiated the IO request by asynchronous IO mode aio_read or aio_write, current process of built-in function clone, allow its read of subprocess triggering synchronous and write operation then, and parent process is returned from aio_read or aio_write immediately, other code in the continuation executive routine.The problem that also has performance when therefore realizing asynchronous IO by the built-in function of POSIX definition.
Summary of the invention
The present invention proposes a kind of system and asynchronous method and asynchronous write method read that realizes the asynchronous IO interface of distributed file system, can improve the performance of the big handling capacity read-write of distributed file system.
Technical scheme of the present invention is achieved in that
A kind of system that realizes the asynchronous IO interface of distributed file system comprises:
Dynamic link library, described dynamic link library are embedded into operation in the application process address space, offer the asynchronous input and output IO interface that described application process is called;
The shared drive module, described shared drive module and described dynamic link library communicate to connect, described shared drive module comprises at least one asynchronous input and output AIO controll block, and described shared drive module realizes the submission of asynchronous read and write request by described AIO controll block and finishes.
Preferably, also comprise:
The agent process module, described agent process module is connected with described shared drive module communication, and described agent process module is used to handle the request of described application process.
Preferably, also comprise:
Data memory module, described data memory module is connected with described agent process module communication, and described data memory module is used to store data.
Preferably, described shared drive module also comprises:
At least one metadata cache piece, described metadata cache piece is corresponding one by one with described AIO controll block, described metadata cache piece is used for the exchanges data between described application process and the described agent process module, and same metadata cache piece can be used simultaneously by a plurality of application processes.
Preferably, described shared drive module comprises that the AIO request queue and the AIO that are used to deposit described AIO controll block finish formation, and the described AIO controll block of described shared drive module is finished the asynchronism of formation control read-write by described AIO request queue and AIO.
A kind of reading/writing method comprises step:
Application process is called the asynchronous input and output IO interface of dynamic link library and is submitted the asynchronous read and write request to, submits the parameter information that reads and writes data to;
Dynamic link library is resolved described parameter information, according to described parameter information application AIO controll block and the described AIO controll block of initialization, the asynchronous read and write request of related described AIO controll block and described application process;
Described AIO controll block is finished the asynchronous read and write that data between described application process and the agent process module are finished in formation by AIO request queue and AIO.
Preferably, described asynchronous read and write request is asynchronous read request, and described AIO controll block is finished formation by AIO request queue and AIO and finished the asynchronous read and write of data between described application process and the agent process module and specifically comprise:
Described AIO controll block is put into described AIO request queue;
Described agent process module is obtained described AIO controll block from described AIO request queue, information according to described AIO controll block reads corresponding data in the metadata cache piece of described AIO controll block correspondence, and described AIO controll block is put into described AIO finish formation;
Described dynamic link library scans described AIO finishes formation, obtains described AIO controll block, and data is copied to the buffer zone of described application process from the metadata cache piece of described AIO controll block correspondence.
Preferably, described asynchronous read and write request is the asynchronous write request, and described AIO controll block is finished formation by AIO request queue and AIO and finished the asynchronous read and write of data between described application process and the agent process module and specifically comprise:
The buffer zone of data to be written from described application process copied to the metadata cache piece of described AIO controll block correspondence, described AIO controll block is put into described AIO request queue;
Described agent process module is obtained described AIO controll block from described AIO request queue, information according to described AIO controll block is written to the data in the metadata cache piece of described AIO controll block correspondence in the corresponding data memory module, after write operation is finished described AIO controll block is put into described AIO and finishes formation.
Preferably, described AIO controll block being put into described AIO finishes and also comprises step after the formation:
Described dynamic link library scans described AIO finishes formation, takes out described AIO controll block.
Preferably, also comprise step:
Described dynamic link library call call back function or transmission information notify the request of described application process asynchronous read and write to finish.
Preferably, described parameter information comprises the buffer pointer information of positional information, size information and store data of filec descriptor, the read/write of read/write file.
Preferably, described application AIO controll block and the described AIO controll block of initialization are specially:
Positional information, size information according to read/write are calculated the page or leaf of wanting read/write, for AIO controll block of each page or leaf application, fill in the file description of read/write in the AIO controll block, the positional information and the size information of read/write.
The present invention realizes asynchronous IO interface by using shared drive mechanism in distributed file system, remedied the defective of synchronous IO interface by the AIO controll block in the shared drive, solve the performance issue of POSIX standard asynchronous IO interface simultaneously, improved the performance of the big handling capacity read-write of distributed file system.
Description of drawings
In order to be illustrated more clearly in the technical scheme of the embodiment of the invention, the accompanying drawing of required use is done to introduce simply in will describing embodiment below, apparently, accompanying drawing in describing below only is some embodiments of the present invention, for those of ordinary skills, under the prerequisite of not paying creative work, can also obtain other accompanying drawing according to these accompanying drawings.
Fig. 1 is a kind of structural representation of realizing the system of the asynchronous IO interface of distributed file system of the present invention;
Fig. 2 is the schematic flow sheet of asynchronous read and write method among the present invention;
Fig. 3 is the asynchronous schematic flow sheet of reading method among the present invention;
Fig. 4 is the schematic flow sheet of asynchronous write method among the present invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the invention, the technical scheme in the embodiment of the invention is clearly and completely described, obviously, described embodiment only is the present invention's part embodiment, rather than whole embodiment.Based on the embodiment among the present invention, those of ordinary skills belong to the scope of protection of the invention not making the every other embodiment that is obtained under the creative work prerequisite.
Performance issue for the standard asynchronous IO that solves POSIX definition, the mode that the present invention adopts shared drive communicates and data transmission between asynchronous IO interface and file access agent process (agent process of distributed file system reading and writing of files, be responsible for the position of locating file content and read and write data).Asynchronous IO interface offers application process in the mode of dynamic link library and calls, and realizes asynchronous reading/writing method fully with the shared drive queue mechanism between dynamic link library and agent process.
The shared drive queue mechanism is the inter-process communication mechanisms of Unix/Linux, a physical memory can be mapped to the address space of a plurality of processes, is used for exchange and shared data between process.In the present invention, shared drive is used for the mutual of message and data between dynamic link library and the agent process.
In shared drive, control the submission and the complete process of an asynchronous read and write request with asynchronous input and output AIO (Asynchronous Input/Ouput) controll block.When the application process calling interface carries out an asynchronous read and write, apply for one or more AIO controll blocks, after this asynchronous read and write is finished, AIO controll block is discharged, for supporting the read-write operation of big handling capacity, can be when the dynamic link library initialization the abundant AIO controll block (can reach several thousand) of predistribution.
Mainly guarantee the asynchronism read and write with two formations in shared drive, one is the AIO request queue, and one is that AIO finishes formation.When asynchronous read and write begins, apply for an AIO controll block, the described AIO controll block of initialization also puts it into the AIO request queue, returns application process then.AIO controll block is therefrom taken out in agent process scan A IO request queue, and the data of reading in the disk write data to disk to the shared drive buffer zone or from the shared drive buffer zone.After the process of agency is finished reading and writing data, AIO controll block is put into AIO finish formation.In dynamic link library, start a plurality of threads, be blocked in AIO and finish in the formation.Each thread is all regularly finished from AIO and is taken out AIO controll block in the formation, calls call back function then or sends the read-write of signalisation application process and finish.
Wherein the shared drive buffer zone is used to deposit the ephemeral data of read-write, when the process of agency reads, deposit data at the shared drive buffer zone so that the thread of dynamic link library is obtained these data.Management for convenience, the size of each buffer zone is fixed as one page (as 128KB), and corresponding one by one with AIO controll block.When distributing AIO controll block, its corresponding buffer region also is assigned with; In like manner, when discharging AIO controll block, buffer zone also is released.
With reference to Fig. 1, show a kind of structural representation of realizing the system of the asynchronous IO interface of distributed file system of the present invention.Described system comprises: dynamic link library 150, shared drive module 120, agent process module 130 and data memory module 140.
Described dynamic link library 150 is embedded into operation in the application process address space, and offers the asynchronous input and output IO interface 110 that described application process is called.Described asynchronous IO interface 110 offers application process in the mode of dynamic link library 150 and calls, and is used to provide the access interface of file system.
Described dynamic link library 150, described shared drive module 120, described agent process module 130 and described data memory module 140 sequenced communications connect.Described dynamic link library 150 is embedded into operation in the application process address space, its name is very similar with the standard file system interface, only need to use macro substitution and recompilate link and can normally use, in each interface function inside, all realize the purpose of access file system by visit to shared drive.
Described shared drive module 120 comprises at least one asynchronous input and output AIO controll block 160, and described shared drive module 120 realizes the submission of asynchronous read and write request by described AIO controll block 160 and finishes.Described shared drive module 120 is utilized the inter-process communication mechanisms of Unix/Linux operating system, a physical memory can be mapped to the address space of a plurality of processes, is used for exchange and shared related data between process.Described AIO controll block 160 and metadata cache piece 170 are corresponding one by one, and when distributing AIO controll block 160, its corresponding metadata cache piece 170 also is assigned with; In like manner, when discharging AIO controll block 160, corresponding metadata cache piece 170 also is released.Described metadata cache piece 170 is divided with fixed size, such as PAGE_SIZE, representative value is 128KB, the same with AIO controll block 160, also can be nearly thousands of, application process and agent process module 130 are by these metadata cache piece 170 swap datas, and same metadata cache piece 170 can be used by a plurality of application processes simultaneously according to certain rule, each metadata cache piece 170 corresponding coding.
Described agent process module 130 is used to handle the request of described application process, it obtains the request of application process by shared drive module 120, and make corresponding processing, such as the metadata of obtaining file, file data is read in described shared drive module 120 or the data in the described shared drive module 120 are write file, and the result after it is handled notifies application process by described shared drive module 120.
Described data memory module 140 is used to store related data.
Described shared drive module 120 guarantees the asynchronism read and write by two formations, and one is the AIO request queue, and one is that AIO finishes formation.When asynchronous read and write begins, apply for an AIO controll block 160, some information of initialization also put it into the AIO request queue.AIO controll block 160 is therefrom taken out in the 130 scan A IO request queues of agent process module, reads in corresponding data and writes data to data memory module 140 to shared drive metadata cache piece 170 or from shared drive metadata cache piece 170.When acting on behalf of after process module 130 finishes reading and writing data, AIO controll block 160 is put into AIO finish formation.In dynamic link library 150, start a plurality of threads, be blocked in AIO and finish in the formation.Each thread is all regularly finished from AIO and is taken out AIO controll block 160 in the formation, according to the information that wherein writes down, calls the read-write of call back function or transmission information notice application process and finishes.
The course of work and the principle of work of described system will make a more detailed description among two method embodiment in the back.
The present invention realizes asynchronous IO interface by using shared drive mechanism in distributed file system, remedied the defective of synchronous IO interface by the AIO controll block in the shared drive, solve the performance issue of POSIX standard asynchronous IO interface simultaneously, improved the performance of the big handling capacity read-write of distributed file system.
With reference to Fig. 2, the schematic flow sheet for the asynchronous read and write method of the asynchronous IO interface of a kind of distributed file system of the present invention comprises step:
Step S201, application process are called the asynchronous input and output IO interface of dynamic link library and are submitted the asynchronous read and write request to, submit the parameter information that reads and writes data to.
Step S202, dynamic link library are resolved described parameter information, according to described parameter information application AIO controll block and the described AIO controll block of initialization, the asynchronous read and write request of related described AIO controll block and described application process.
Step S203, described AIO controll block are finished the asynchronous read and write that data between described application process and the agent process module are finished in formation by AIO request queue and AIO.
More than the detailed description of each step will in the described embodiment of Fig. 3, Fig. 4, be described, do not repeat them here, get final product with reference to the description of relevant portion.
With reference to Fig. 3, the asynchronous schematic flow sheet of reading method for the asynchronous IO interface of a kind of distributed file system of the present invention comprises step:
Step S301, application process are called asynchronous input and output IO interface and are submitted asynchronous read request to, submit the parameter information of reading of data to.
Application process is submitted asynchronous read request to, brings the parameter informations such as buffer pointer of the filec descriptor (obtaining in File Open) that reads file, the position of reading, size and store data into;
Step S302, dynamic link library are resolved described parameter information, application AIO controll block and the described AIO controll block of initialization, the asynchronous read request of related described AIO controll block and described application process.
Resolve read request, be converted into the page or leaf (one page or multipage) that will read according to the position of reading, size, for AIO controll block of each page or leaf application, fill in information such as the file description that reads in the AIO controll block, the position of reading, size, and related read request and each AIO controll block.
Step S303, described AIO controll block is put into the AIO request queue.
Each AIO controll block chain is gone in the AIO request queue of shared drive module.
Step S304, return asynchronous read operation to application process and call success.
Step S305, agent process module are obtained described AIO controll block from described AIO request queue.
Step S306, according to the corresponding data in the information reading of data memory module of described AIO controll block in the metadata cache piece of described AIO controll block correspondence.
Step S307, described AIO controll block is put into AIO finish formation.
Step S308, described dynamic link library scan described AIO finishes formation, obtains described AIO controll block, and data is copied to the buffer zone of described application process from the metadata cache piece of described AIO controll block correspondence.
AIO in the dynamic link library checks that thread is finished from AIO and takes out AIO controll block in the formation, with the copying data in the metadata cache piece of AIO controll block correspondence in the buffer zone of application process.
Step S309, judge whether the AIO controll block of the asynchronous read request correspondence of application process is all finished, otherwise continue step S308.
The AIO controll block of the asynchronous read request correspondence of step S310, application process is all finished, and then calls the read operation of call back function (being begun just to configure by application process) or transmission information notice application process and finishes.
With reference to Fig. 4, the schematic flow sheet for the asynchronous write method of the asynchronous IO interface of a kind of distributed file system of the present invention comprises step:
Step S401, application process are called asynchronous input and output IO interface and are submitted the asynchronous write request to, submit the parameter information of write data to.
Application process is submitted the asynchronous write request to, brings the parameter informations such as buffer pointer of the filec descriptor (obtaining in File Open) of written document, the position that writes, size and store data into.
Step S402, dynamic link library are resolved described parameter information, application AIO controll block and the described AIO controll block of initialization.
Resolve write request,,, fill in information such as the file description that writes in the AIO controll block, the position that writes, size for AIO controll block of each page or leaf application according to the page or leaf (one page or multipage) that the position that writes, big or small conversion will write.
Step S403, copy the metadata cache piece of described AIO controll block correspondence data to be written the asynchronous write request of related described AIO controll block and described application process to from the buffer zone of application process.
Step S404, described AIO controll block is put into the AIO request queue.
Each AIO controll block chain is gone into AIO request queue to the shared drive module.
Step S405, return asynchronous write operation calls success to application process.
Step S406, agent process module are obtained described AIO controll block from described AIO request queue.
Step S407, the data in the metadata cache piece of described AIO controll block correspondence are written in the corresponding data memory module according to the information of described AIO controll block.
Find the data memory module of data correspondence, the data in the AIO controll block metadata cache piece are written in the corresponding data memory module.
After step S408, write operation are finished described AIO controll block is put into AIO and finish formation.
Step S409, described dynamic link library scan described AIO finishes formation, takes out described AIO controll block.
AIO inspection thread in the dynamic link library is finished from AIO and is taken out AIO controll block in the formation.
Step S410, judge whether the AIO controll block of the asynchronous write request correspondence of application process is all finished, otherwise continue step S409.
The AIO controll block of the asynchronous write request correspondence of step S411, application process is all finished, and then calls call back function (being begun just to configure by application process) or transmission information notice application process write operation and finishes.
One of ordinary skill in the art will appreciate that; in each method embodiment of the present invention; the sequence number of described each step can not be used to limit the sequencing of each step; for those of ordinary skills; under the prerequisite of not paying creative work, the priority of each step is changed also within protection scope of the present invention.
The above only is preferred embodiment of the present invention, and is in order to restriction the present invention, within the spirit and principles in the present invention not all, any modification of being done, is equal to replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (12)
1. a system that realizes the asynchronous IO interface of distributed file system is characterized in that, comprising:
Dynamic link library, described dynamic link library are embedded into operation in the application process address space, offer the asynchronous input and output IO interface that described application process is called;
The shared drive module, described shared drive module and described dynamic link library communicate to connect, described shared drive module comprises at least one asynchronous input and output AIO controll block, and described shared drive module realizes the submission of asynchronous read and write request by described AIO controll block and finishes.
2. the system of the asynchronous IO interface of realization distributed file system according to claim 1 is characterized in that, also comprises:
The agent process module, described agent process module is connected with described shared drive module communication, and described agent process module is used to handle the request of described application process.
3. the system of the asynchronous IO interface of realization distributed file system according to claim 2 is characterized in that, also comprises:
Data memory module, described data memory module is connected with described agent process module communication, and described data memory module is used to store data.
4. the system of the asynchronous IO interface of realization distributed file system according to claim 3 is characterized in that, described shared drive module also comprises:
At least one metadata cache piece, described metadata cache piece is corresponding one by one with described AIO controll block, described metadata cache piece is used for the exchanges data between described application process and the described agent process module, and same metadata cache piece can be used simultaneously by a plurality of application processes.
5. according to the system of the asynchronous IO interface of each described realization distributed file system of claim 1 to 4, it is characterized in that:
Described shared drive module comprises that the AIO request queue and the AIO that are used to deposit described AIO controll block finish formation, and the described AIO controll block of described shared drive module is finished the asynchronism of formation control read-write by described AIO request queue and AIO.
6. a reading/writing method is characterized in that, comprises step:
Application process is called the asynchronous input and output IO interface of dynamic link library and is submitted the asynchronous read and write request to, submits the parameter information that reads and writes data to;
Dynamic link library is resolved described parameter information, according to described parameter information application AIO controll block and the described AIO controll block of initialization, the asynchronous read and write request of related described AIO controll block and described application process;
Described AIO controll block is finished the asynchronous read and write that data between described application process and the agent process module are finished in formation by AIO request queue and AIO.
7. reading/writing method according to claim 6, it is characterized in that, described asynchronous read and write request is asynchronous read request, and described AIO controll block is finished formation by AIO request queue and AIO and finished the asynchronous read and write of data between described application process and the agent process module and specifically comprise:
Described AIO controll block is put into described AIO request queue;
Described agent process module is obtained described AIO controll block from described AIO request queue, information according to described AIO controll block reads corresponding data in the metadata cache piece of described AIO controll block correspondence, and described AIO controll block is put into described AIO finish formation;
Described dynamic link library scans described AIO finishes formation, obtains described AIO controll block, and data is copied to the buffer zone of described application process from the metadata cache piece of described AIO controll block correspondence.
8. reading/writing method according to claim 6, it is characterized in that, described asynchronous read and write request is the asynchronous write request, and described AIO controll block is finished formation by AIO request queue and AIO and finished the asynchronous read and write of data between described application process and the agent process module and specifically comprise:
The buffer zone of data to be written from described application process copied to the metadata cache piece of described AIO controll block correspondence, described AIO controll block is put into described AIO request queue;
Described agent process module is obtained described AIO controll block from described AIO request queue, information according to described AIO controll block is written to the data in the metadata cache piece of described AIO controll block correspondence in the corresponding data memory module, after write operation is finished described AIO controll block is put into described AIO and finishes formation.
9. reading/writing method according to claim 8 is characterized in that, described AIO controll block is put into described AIO finish and also comprise step after the formation:
Described dynamic link library scans described AIO finishes formation, takes out described AIO controll block.
10. reading/writing method according to claim 6 is characterized in that, also comprises step:
Described dynamic link library call call back function or transmission information notify the request of described application process asynchronous read and write to finish.
11. reading/writing method according to claim 6 is characterized in that:
Described parameter information comprises the buffer pointer information of positional information, size information and store data of filec descriptor, the read/write of read/write file.
12. reading/writing method according to claim 6 is characterized in that, described application AIO controll block and the described AIO controll block of initialization are specially:
Positional information, size information according to read/write are calculated the page or leaf of wanting read/write, for AIO controll block of each page or leaf application, fill in the file description of read/write in the AIO controll block, the positional information and the size information of read/write.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2009100858651A CN101576912A (en) | 2009-06-03 | 2009-06-03 | System and reading and writing method for realizing asynchronous input and output interface of distributed file system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2009100858651A CN101576912A (en) | 2009-06-03 | 2009-06-03 | System and reading and writing method for realizing asynchronous input and output interface of distributed file system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101576912A true CN101576912A (en) | 2009-11-11 |
Family
ID=41271845
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2009100858651A Pending CN101576912A (en) | 2009-06-03 | 2009-06-03 | System and reading and writing method for realizing asynchronous input and output interface of distributed file system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101576912A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101937465A (en) * | 2010-09-10 | 2011-01-05 | 中兴通讯股份有限公司 | Access method of distributed file system and upper file system thereof |
| WO2011147073A1 (en) * | 2010-05-24 | 2011-12-01 | 中兴通讯股份有限公司 | Data processing method and device in distributed file system |
| CN102097122B (en) * | 2009-12-10 | 2013-03-27 | 上海华虹集成电路有限责任公司 | NAND flash controller circuit of multi-channel shared data cache region |
| CN103324491A (en) * | 2012-03-20 | 2013-09-25 | 腾讯科技（深圳）有限公司 | Method and device for loading data |
| CN105843693A (en) * | 2016-03-22 | 2016-08-10 | 同济大学 | High-speed maglev transportation simulation oriented memory sharing method |
| WO2017028719A1 (en) * | 2015-08-19 | 2017-02-23 | 阿里巴巴集团控股有限公司 | Metadata output method, client side, and metadata server |
| CN108121594A (en) * | 2016-11-29 | 2018-06-05 | 阿里巴巴集团控股有限公司 | A kind of process management method and device |
| CN108153637A (en) * | 2016-12-05 | 2018-06-12 | 武汉比亚迪汽车有限公司 | Detect the method and device of EEPROM read-write capabilities |
| CN108737397A (en) * | 2018-05-09 | 2018-11-02 | 烽火通信科技股份有限公司 | A kind of method for realizing data interaction between business and protocol stack in router |
-
2009
- 2009-06-03 CN CNA2009100858651A patent/CN101576912A/en active Pending
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102097122B (en) * | 2009-12-10 | 2013-03-27 | 上海华虹集成电路有限责任公司 | NAND flash controller circuit of multi-channel shared data cache region |
| WO2011147073A1 (en) * | 2010-05-24 | 2011-12-01 | 中兴通讯股份有限公司 | Data processing method and device in distributed file system |
| CN101937465A (en) * | 2010-09-10 | 2011-01-05 | 中兴通讯股份有限公司 | Access method of distributed file system and upper file system thereof |
| CN101937465B (en) * | 2010-09-10 | 2013-09-11 | 中兴通讯股份有限公司 | Access method of distributed file system and upper file system thereof |
| CN103324491A (en) * | 2012-03-20 | 2013-09-25 | 腾讯科技（深圳）有限公司 | Method and device for loading data |
| WO2017028719A1 (en) * | 2015-08-19 | 2017-02-23 | 阿里巴巴集团控股有限公司 | Metadata output method, client side, and metadata server |
| CN105843693A (en) * | 2016-03-22 | 2016-08-10 | 同济大学 | High-speed maglev transportation simulation oriented memory sharing method |
| CN105843693B (en) * | 2016-03-22 | 2019-05-14 | 同济大学 | A kind of shared drive method towards high speed Maglev emulation |
| CN108121594A (en) * | 2016-11-29 | 2018-06-05 | 阿里巴巴集团控股有限公司 | A kind of process management method and device |
| WO2018099292A1 (en) * | 2016-11-29 | 2018-06-07 | 阿里巴巴集团控股有限公司 | Process management method and apparatus |
| CN108153637A (en) * | 2016-12-05 | 2018-06-12 | 武汉比亚迪汽车有限公司 | Detect the method and device of EEPROM read-write capabilities |
| CN108153637B (en) * | 2016-12-05 | 2021-07-09 | 武汉比亚迪汽车有限公司 | Method and device for detecting read-write function of EEPROM |
| CN108737397A (en) * | 2018-05-09 | 2018-11-02 | 烽火通信科技股份有限公司 | A kind of method for realizing data interaction between business and protocol stack in router |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101576912A (en) | System and reading and writing method for realizing asynchronous input and output interface of distributed file system | |
| CN110663019B (en) | File system for Shingled Magnetic Recording (SMR) | |
| RU2463648C2 (en) | Selective use of set of different cells of solid-state storage | |
| JP5159884B2 (en) | Network adapter resource allocation between logical partitions | |
| CN100458757C (en) | Inter core communication method and apparatus for multi-core processor in embedded real-time operating system | |
| US7783853B1 (en) | Memory usage techniques in middleware of a real-time data distribution system | |
| CN103019949A (en) | Allocation method and device of write combine attribute memory space | |
| CN101382953A (en) | Interface system for accessing file system in user space and file reading and writing method | |
| US9454481B2 (en) | Affinity group access to global data | |
| RU2641244C2 (en) | Unified access to jointly used and controlled memory | |
| US10013264B2 (en) | Affinity of virtual processor dispatching | |
| US11687276B2 (en) | Data streaming for computational storage | |
| EP1094392B1 (en) | Method and apparatus for interfacing with a secondary storage system | |
| CN113535395A (en) | Descriptor queue and memory optimization method, system and application of network storage service | |
| CN112817887B (en) | Far memory access optimization method and system under separated combined architecture | |
| WO2023201987A1 (en) | Request processing method and apparatus, and device and medium | |
| CN105408875A (en) | Distributed procedure execution and file systems on a memory interface | |
| CN112948293A (en) | DDR arbiter and DDR controller chip of multi-user interface | |
| WO2011012003A1 (en) | Method and system for controlling service based on service componentized application software | |
| CN117492661A (en) | Data writing method, medium, device and computing equipment | |
| CN116302105B (en) | Access instruction scheduling method, system, hard disk, controller, storage medium and program product | |
| EP1408508A1 (en) | Storage device | |
| US20230236730A1 (en) | Storage device and storage system | |
| CN104025026B (en) | Configuration and status register of the access for configuration space | |
| US8892807B2 (en) | Emulating a skip read command |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20091111 |
|
| C20 | Patent right or utility model deemed to be abandoned or is abandoned |