CN101271458A - Distributed computation file management method, system and device thereof - Google Patents
Distributed computation file management method, system and device thereof Download PDFInfo
- Publication number
- CN101271458A CN101271458A CNA2007100871464A CN200710087146A CN101271458A CN 101271458 A CN101271458 A CN 101271458A CN A2007100871464 A CNA2007100871464 A CN A2007100871464A CN 200710087146 A CN200710087146 A CN 200710087146A CN 101271458 A CN101271458 A CN 101271458A
- Authority
- CN
- China
- Prior art keywords
- global profile
- file
- master server
- distributed
- log
- 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
- Computer And Data Communications (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The present invention discloses a distributed computer file management method, including the following steps: registering to the overall file of a major server; scheduling the metadata of the overall file from the major server when in need of processing the overall file; taking back the files included in the overall file from various distributed nodes according to the metadata of the overall file. By adopting the management method, the user only needs to know the overall file instead of which files are included in the overall file and which distributed nodes the files are stored in, thus decreasing the complexity of distributed calculation and reducing the workload of users.
Description
Technical field
The present invention relates to networking technology area, particularly relate to a kind of distributed computation file management method, system and its apparatus.
Background technology
Along with the develop rapidly of Internet technology, just become a difficult problem for the processing that has magnanimity information on the internet.If on unit, handle the very huge information of these data volumes, just need very strong processing power and port ability, yet this high-performance unit is expensive on the one hand, and the tupe of this dependence unit also is limited.Therefore people have proposed the notion of Distributed Calculation, utilize the idling-resource that a lot of interconnecting computer are arranged on the internet to handle magnanimity information on the internet.
And in Distributed Calculation, be exactly distributed storage aspect important, because the continuous increase of data volume, unit can not satisfy the needs of storage, needs special equipment store data; The use of data simultaneously not only has been confined to a place, a kind of application of system, but to a plurality of users, a plurality of systems are shared.Therefore distributed storage can be stored large-capacity data information, can satisfy the Real-Time Sharing to multisystem, multi-user again.
Prior art be with file storage on special high speed storing equipment, system provides the interface of access file.The user visits file by the interface that these systems provide, and reaches fast by memory device high speed, reliable the reading and writing of file are operated.If but file is very big, and file read-write is very frequent, and efficient is just very low.This is for general and sacrificed performance.
Prior art also proposes a kind of distributed document disposal route at the specific occasion, the mode of file according to piece is kept on the distributed node machine, preserve file metadata information (how many pieces file comprises, and which node each piece is distributed in respectively on) by a master control machine and carry out the storage of Distributed Calculation file.When the user reads file, read the block message of file earlier from the master control machine, from distributed node, read file data according to block message then; When the user writes file, read fileinfo from the master control machine earlier,, select the node of a document storage then according to load balancing; End user is written to file data in the distributed node.But this method can only allow the read-write of user order, and the user also writes corresponding distributed code according to the distribution situation of each piece of file, is very loaded down with trivial details to the user, has therefore increased user's workload.
Summary of the invention
The problem to be solved in the present invention provides a kind of distributed computation file management method, system and its apparatus, reduces the complexity of Distributed Calculation, alleviates user's workload.
For achieving the above object, the technical scheme of embodiments of the invention proposes a kind of distributed computation file management method, may further comprise the steps: to the global profile registration of master server; When needs are handled described global profile, transfer the metadata of described global profile from described master server; Metadata according to described global profile is fetched the file that described global profile comprises from each distributed node.
Wherein, after the described file that global profile is comprised is fetched from each node, further comprising the steps of: that the described file of fetching from each distributed node is merged.
Wherein, the metadata of described global profile specifically comprises: the corresponding relation of the file that global profile comprises and described file and distributed node.
Wherein, the described metadata of transferring global profile from master server specifically may further comprise the steps: send the request of reading described global profile to described master server; Described server issues after according to described request the metadata of described global profile being taken out.
Wherein, described global profile registration to master server, specifically may further comprise the steps: each distributed node is registered the local file on the described distributed node to described master server; Described master server is revised described global profile metadata according to the local file log-on message on the described distributed node.
The technical scheme of embodiments of the invention also proposes a kind of distributed computation file management system, comprise client, master server and distributed node, described master server, be used for the metadata of global profile is managed, and when described client need be handled described global profile, the metadata information of described global profile is handed down to described client; Described distributed node is used for the local file on the described distributed node is registered to the global profile of described master server, and after receiving described client-requested described local file is handed down to described client.
Wherein, described master server comprises that global profile administration module and global profile issue module, and described global profile administration module is used for the metadata of global profile is managed; Described global profile issues module and is used for when described client need be handled described global profile the global profile metadata information in the described global profile administration module being handed down to described client
Wherein, described master server also comprises the registration receiver module, is used to receive the log-on message that described distributed node sends to described master server, and described log-on message is passed to described global profile administration module.
Wherein, described distributed node comprises that registration reporting module and file data issue module, and described registration reporting module is used for the local file on the described distributed node is registered to the global profile of described master server; Described file data issues module and is used for after receiving described client-requested described local file being handed down to described client.
Wherein, before described global profile registration to master server, further comprising the steps of: as to set the local file on each distributed node and the corresponding relation of global profile.
The technical scheme of embodiments of the invention also proposes a kind of master server, comprises that global profile administration module and global profile issue module, and described global profile administration module is used for the metadata of global profile is managed; Described global profile issues module and is used for when described client need be handled described global profile the global profile metadata information in the described global profile administration module being handed down to described client
Wherein, master server also comprises the registration receiver module, is used to receive the log-on message that described distributed node sends to described master server, and described log-on message is passed to described global profile administration module.
The technical scheme of the embodiment of the invention has realized that the user only operates global profile, and does not need to write distributed correlative code, the distributed distributed computation file management method that improves the performance that file reads that can utilize many machines again.By this management method, the user only need know a global profile, and does not need to know which file is this global profile have constitute, which be kept at again on the distributed node, for the user, then reduced the complexity of Distributed Calculation, alleviated user's workload.
Description of drawings
Fig. 1 is an embodiment of the invention distributed computation file management system construction drawing;
Fig. 2 is the structural drawing of the another kind of distributed computation file management of embodiment of the invention system;
Fig. 3 is the process flow diagram of embodiment of the invention distributed computation file management method;
Fig. 4 is the process flow diagram of the distributed computation file management method of example with the web log file disposal system for the embodiment of the invention;
Fig. 5 is the process flow diagram of the embodiment of the invention to the Distributed Calculation file modification;
Fig. 6 be the embodiment of the invention a kind of on distributed node to the process flow diagram of Distributed Calculation file modification.
Embodiment
Below in conjunction with drawings and Examples, the specific embodiment of the present invention is described in further detail:
Core concept of the present invention is that this global profile is made of the local file that much is distributed on each distributed node by defining a kind of global profile.In the metadata of the global profile of the information storage of this global profile on master server, the information of global profile comprises: which file the title of global profile, global profile comprise, and these files that comprise and the corresponding relation of each distributed node.And the file that this global profile comprises is formed from trend master server registration by distributed node, and the user only knows that this global profile is used for What for or handle what data, and need not know detailed problems such as comprising which file in this global profile.Therefore can hide the file processing details of distributed treatment by this global profile, and when reading file, utilize the advantage of a plurality of distributed node parallel processings of distributed system to improve access speed.
As shown in Figure 1, be embodiment of the invention distributed computation file management system construction drawing,
This distributed computation file management system comprises master server 1, distributed node 2 and client 3, master server 1 is used for the metadata of global profile is managed, and when client 3 need be handled global profile, the metadata information of global profile is handed down to client 3; Distributed node 2 is used for the local file on the distributed node 2 is registered to global profile on the master server 1, and after receiving client 3 requests the local file on the distributed node 2 is handed down to client 3.
As follows at above-mentioned distributed computation file management system distributed file management process, at first distributed node 2 calls registration interface, local file is registered on the global profile by ICP/IP protocol according to predefined corresponding relation, this predefined corresponding relation can be set as required by the user, and the local file on the distributed node 2 can corresponding a plurality of global profiles; The log-on message of local file on each distributed node 2 of master server 1 management, i.e. the metadata of global profile, this metadata comprises which file is global profile comprise, the corresponding relation of the distributed node 2 of these files and its existence; When client 3 need be transferred global profile and handles, then call file interface sends the request from certain global profile metadata of inquiry to master server 1 of reading; Master server 1 is according to the global profile title of client 3 request inquiry, by ICP/IP protocol the metadata of this global profile is handed down to client 3; The file that the metadata information of the global profile that client 3 bases are passed back comprises this global profile from each distributed node 2 is all fetched and is merged, and is combined into a global profile.
Wherein, master server 1 comprises that global profile issues module 11, global profile administration module 12 and registration receiver module 13, and global profile administration module 12 is used for the metadata of global profile is managed; Global profile issues module 11 and is used for when client 3 need be handled global profile the global profile metadata information in the global profile administration module 12 being handed down to client 3; Registration receiver module 13 is used to receive the log-on message that distributed node 2 sends to master server 1, and this log-on message is passed to global profile administration module 12.
Wherein, distributed node 2 comprises that registration reporting module 21 and file data issue module 22, and registration reporting module 21 is used for the local file on the distributed node 2 is registered to global profile on the master server 1; File data issues module 22 and is used for after receiving client 3 requests local file being handed down to client 3.
As shown in Figure 2, be the structural drawing of the another kind of distributed computation file management of embodiment of the invention system, this system also is considered as distributed node 2 with client in the foregoing description 3.
This distributed computation file management system comprises master server 1 and distributed node 2, master server 1 is used for the metadata of global profile is managed, and when needs are handled global profile, the metadata information of global profile is handed down to distributed node 2; Distributed node 2 is used for the local file on the distributed node 2 is registered to global profile on the master server 1; File data issues module 22 and is used for after receiving other distributed node 2 requests local file being issued.
As follows at above-mentioned distributed computation file management system distributed file management process, at first distributed node 2 calls registration interface, local file is registered on the global profile by ICP/IP protocol according to predefined corresponding relation, this predefined corresponding relation can be set as required by the user, and the local file on the distributed node 2 can corresponding a plurality of global profiles; The log-on message of local file on each distributed node 2 of master server 1 management, i.e. the metadata of global profile, this metadata comprises which file is global profile comprise, the corresponding relation of the distributed node 2 of these files and its existence; When needs are transferred global profile and handled, then call file interface sends the request from certain global profile metadata of inquiry to master server 1 of reading; Master server 1 is according to the global profile title of request inquiry, by ICP/IP protocol the metadata of this global profile is handed down to distributed node 2; The file that the metadata information of the global profile that distributed node 2 bases are passed back comprises this global profile from each distributed node 2 is all fetched and is merged, and is combined into a global profile.
Wherein, master server 1 comprises that global profile issues module 11, global profile administration module 12 and registration receiver module 13, and global profile administration module 12 is used for the metadata of global profile is managed; Global profile issues module 11 and is used for when needs are handled global profile the global profile metadata information in the global profile administration module 12 being issued; Registration receiver module 13 is used to receive the log-on message that distributed node 2 sends to master server 1, and this log-on message is passed to global profile administration module 12.
Wherein, distributed node 2 comprises that registration reporting module 21 and file data issue module 22, and registration reporting module 21 is used for the local file on the distributed node 2 is registered to global profile on the master server 1; File data issues module 22 and is used for after receiving other distributed node 2 requests local file being issued.
As shown in Figure 3, the process flow diagram for embodiment of the invention distributed computation file management method may further comprise the steps:
Step S301 is to the global profile registration of master server; Distributed node calls registration interface, by ICP/IP protocol local file is registered to master server according to predefined corresponding relation, and master server is revised the metadata of global profile according to log-on message; This predefined corresponding relation can be set as required by the user, and the local file on the distributed node can corresponding a plurality of global profiles; For example: suppose that whole distributed system has 2 distributed nodes, a master server.Local file a1 is arranged on the distributed node m, and a2 is totally 2 files, and distributed node 2 has local file b1, and b2 is totally 2 files; The user is made as the local file a1 on the distributed node 1 corresponding with global profile file1 and global profile file2 as required; Be made as the local file a2 on the distributed node m corresponding with global profile file1; Also be made as the local file b1 on the distributed node n corresponding with global profile file1, and be made as the local file b2 on the distributed node n corresponding with global profile file2, according to above-mentioned default corresponding relation, distributed node m calls registration interface and passes through ICP/IP protocol, send register instruction to master server, with the a1 file registration is global profile file1, and is global profile file2 with the a1 file registration, is global profile file1 with the a2 file registration; Simultaneously, distributed node n calls registration interface by ICP/IP protocol, sends register instruction to master server, is global profile file1 with the b1 file registration, and is global profile file2 with the b2 file registration.By above-mentioned registration process, exist 2 global profiles to be respectively file1 and file2 at present on the master server, wherein global profile file1 comprises a1 on the distributed node m and the b1 on a2 and the distributed node 2, and global profile file2 comprises a1 on the distributed node m and the b2 on the distributed node n.Master server is managed the log-on message of local file on each distributed node, i.e. the metadata of global profile, and this metadata comprises which file is global profile comprise, the corresponding relation of the distributed node of these files and its existence; For the foregoing description, the metadata of global profile file1 comprises a1, a2 and b1, and the corresponding relation of a1, a2 and b1 and distributed node.
Step S302 when needs are handled global profile, transfers the metadata of global profile from master server.When client need be transferred global profile and handles, then call and read file interface, send the request of certain global profile metadata of inquiry to master server; Master server issues by the metadata of ICP/IP protocol with this global profile according to the global profile title of request inquiry.Wherein, the client of registration global profile can not be same client with the client that reads global profile, and client is when reading global profile and do not know what global profile comprises, only after master server issued the metadata of global profile, client just knew according to the metadata of global profile which file is global profile comprise.
Step S303 fetches the file that global profile comprises according to the metadata of global profile from each distributed node; For the foregoing description, then a1, a2 are fetched from distributed node m, b1 is fetched from distributed node n.
As shown in Figure 4, for the embodiment of the invention is the process flow diagram of the distributed computation file management method of example with the web log file disposal system, may further comprise the steps:
Step S401, set the local file on the distributed node and the corresponding relation of global profile as required, for example then need local file news.log (news daily record) on the distributed node one for the web log file disposal system, local file club.log on the node two (club's daily record), local file bbs.log on the node three (forum's daily record), local file search.log on the node four (search daily record), and the local file blog.log on the node five (blog daily record), the local file on these each nodes has constituted this web log file jointly and has handled needed global profile cookie.log.
Step S402, register to the global profile of master server according to each distributed node of corresponding relation that above-mentioned steps is set, for example distributed node one finds that local file has news.log to exist, then call registration interface and by ICP/IP protocol local file news.log is registered to master server, the corresponding relation that master server is set according to above-mentioned steps is registered to global profile cookie.log with news.log; Then respectively club.log, bbs.log, s earch.log and blog.log are registered to master server by that analogy, the corresponding relation that master server is set according to above-mentioned steps is registered above-mentioned local file to global profile cookie.log.As long as knowing global profile cookie.log so for the user is that to handle the file of web log file just passable, and does not need to know that global profile cookie.log comprises the corresponding relation of what file and these files and distributed node actually.
Step S403 if the user needs above-mentioned network log is carried out statistical treatment on the distributed system client, then calls and reads file interface, sends the request of inquiry global profile cookie.log metadata to master server; Master server is according to the global profile cookie.log of request inquiry, by ICP/IP protocol the metadata of this global profile is handed down to client, be about to file news.log, club.log, bbs.log, search.log and blog.log that global profile cookie.log comprises, and the corresponding relation of these files and distributed node is handed down to client.Wherein, the client of registration global profile can not be same client with the client that reads global profile, and client is when reading global profile and do not know what global profile comprises, only after master server issued the metadata of global profile, client just knew according to the metadata of global profile which file is global profile comprise.
Step S404, client is according to the metadata information of the above-mentioned global profile that issues, and the file news.lo g, club.log, bbs.log, search.log and the blog.log that global profile cookie.log are comprised from other node fetch the back and merge.For example fetch news.log (news daily record) from distributed node one, fetch club.log (club's daily record) from node two by ICP/IP protocol, and by that analogy.The All Files of the processing network log of user's needs has all focused on the employed client of user and has used for the user like this.
As shown in Figure 5, being the process flow diagram of the embodiment of the invention to the Distributed Calculation file modification, is that example is made amendment to global profile cookie.log with Fig. 4 embodiment, may further comprise the steps:
Step S501 sets the local file on the distributed node and the corresponding relation of global profile as required.
Step S502, according to the corresponding relation that above-mentioned steps is set, each distributed node is to the global profile registration of master server.Distributed node one finds that local file has news.log to exist, then call registration interface and by ICP/IP protocol local file news.log is registered to master server, the corresponding relation that master server is set according to above-mentioned steps is registered to global profile cookie.log with news.log; Then respectively club.log, bbs.log, search.log and blog.log are registered to master server by that analogy, the corresponding relation that master server is set according to above-mentioned steps is registered above-mentioned local file to global profile cookie.log.As long as knowing global profile cookie.log so for the user is that to handle the file of web log file just passable, and does not need to know that global profile cookie.log comprises the corresponding relation of what file and these files and distributed node actually.
Step S503 if the user needs global profile cookie.log is made amendment on distributed node, then calls and reads file interface, sends the request of inquiry global profile cookie.log metadata to master server; Master server is according to the global profile cookie.log of request inquiry, by ICP/IP protocol the metadata of this global profile is handed down to client, be about to file news.log, club.log, bbs.log, search.log and blog.log that global profile cookie.log comprises, and the corresponding relation of these files and distributed node is handed down to client.
Step S504, the metadata information of the global profile that distributed node issues according to step S503, file news.log, the club.log, bbs.log, search.log and the blog.log that global profile cookie.log are comprised from other node fetch the back and merge.For example fetch news.log (news daily record) from distributed node one, fetch club.log (club's daily record) from node two by ICP/IP protocol, and by that analogy.
Step S505 makes amendment to global profile cookie.log on distributed node.
Step S506 according to predefined corresponding relation, registers global profile cookie.log with cookie.log as the global profile corresponding of the local file on this distributed node on master server.
As shown in Figure 6, for the embodiment of the invention a kind of on distributed node to the process flow diagram of Distributed Calculation file modification, with Fig. 4 embodiment be example to exist among the global profile cookie.log with distributed node one on local file news.log (news daily record) on distributed node one, make amendment, may further comprise the steps:
Step S601 sets the local file on the distributed node and the corresponding relation of global profile as required.
Step S602 registers to the global profile of master server according to the corresponding relation that above-mentioned steps is set.Distributed node one finds that local file has news.log to exist, then call registration interface and by ICP/IP protocol local file news.log is registered to master server, the corresponding relation that master server is set according to above-mentioned steps is registered to global profile cookie.log with news.log; Then respectively club.log, bbs.log, search.log and blog.log are registered to master server by that analogy, the corresponding relation that master server is set according to above-mentioned steps is registered above-mentioned local file to global profile cookie.log.As long as knowing global profile cookie.log so for the user is that to handle the file of web log file just passable, and does not need to know that global profile cookie.log comprises the corresponding relation of what file and these files and distributed node actually.
Step S603 makes amendment to the news.log on the distributed node one.
Step S604, distributed node one register to the global profile of master server again.
The embodiment of the invention has realized that by the global profile of setting for the user user only operates global profile, and do not need to write distributed correlative code, the distributed distributed computation file management method that improves the performance that file reads that can utilize many machines again.By this management method, the user only need know a global profile, and does not need to know which file is this global profile have constitute, and is kept at again on which distributed node.For the user, then reduced the complexity of Distributed Calculation, alleviated user's workload.
The above only is a preferred implementation of the present invention; should be pointed out that for those skilled in the art, under the prerequisite that does not break away from the principle of the invention; can also make some improvements and modifications, these improvements and modifications also should be considered as protection scope of the present invention.
Claims (12)
1. a distributed computation file management method is characterized in that, may further comprise the steps:
Global profile registration to master server;
When needs are handled described global profile, transfer the metadata of described global profile from described master server;
Metadata according to described global profile is fetched the file that described global profile comprises from each distributed node.
2. distributed computation file management method according to claim 1 is characterized in that, and is after the described file that global profile is comprised is fetched from each node, further comprising the steps of:
The described file of fetching from each distributed node is merged.
3. distributed computation file management method according to claim 1 is characterized in that the metadata of described global profile specifically comprises: the corresponding relation of the file that global profile comprises and described file and distributed node.
4. distributed computation file management method according to claim 1 is characterized in that, the described metadata of transferring global profile from master server specifically may further comprise the steps:
Send the request of reading described global profile to described master server;
Described server issues after according to described request the metadata of described global profile being taken out.
5. distributed computation file management method according to claim 1 is characterized in that, described global profile registration to master server specifically may further comprise the steps:
Each distributed node is registered the local file on the described distributed node to described master server;
Described master server is revised described global profile metadata according to the log-on message of the local file on the described distributed node.
6. distributed computation file management method according to claim 1 is characterized in that, and is before the global profile registration of master server, further comprising the steps of described:
Set the local file on each distributed node and the corresponding relation of global profile.
7. a distributed computation file management system comprises client, it is characterized in that, also comprises master server and distributed node,
Described master server is used for the metadata of global profile is managed, and when described client need be handled described global profile, and the metadata information of described global profile is handed down to described client;
Described distributed node is used for the local file on the described distributed node is registered to the global profile of described master server, and after receiving described client-requested described local file is handed down to described client.
8. as distributed computation file management system as described in the claim 7, it is characterized in that described master server comprises that global profile administration module and global profile issue module,
Described global profile administration module is used for the metadata of global profile is managed;
Described global profile issues module and is used for when described client need be handled described global profile the global profile metadata information in the described global profile administration module being handed down to described client
9. as distributed computation file management system as described in the claim 8, it is characterized in that, described master server also comprises the registration receiver module, is used to receive the log-on message that described distributed node sends to described master server, and described log-on message is passed to described global profile administration module.
10. as distributed computation file management system as described in the claim 7, it is characterized in that described distributed node comprises that registration reporting module and file data issue module,
Described registration reporting module is used for the local file on the described distributed node is registered to the global profile of described master server;
Described file data issues module and is used for after receiving described client-requested described local file being handed down to described client.
11. a master server is characterized in that, comprises that global profile administration module and global profile issue module,
Described global profile administration module is used for the metadata of global profile is managed;
Described global profile issues module and is used for when described client need be handled described global profile the global profile metadata information in the described global profile administration module being handed down to described client
12., it is characterized in that as master server as described in the claim 11, also comprise the registration receiver module, be used to receive the log-on message that described distributed node sends to described master server, and described log-on message is passed to described global profile administration module.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNA2007100871464A CN101271458A (en) | 2007-03-22 | 2007-03-22 | Distributed computation file management method, system and device thereof |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNA2007100871464A CN101271458A (en) | 2007-03-22 | 2007-03-22 | Distributed computation file management method, system and device thereof |
Publications (1)
Publication Number | Publication Date |
---|---|
CN101271458A true CN101271458A (en) | 2008-09-24 |
Family
ID=40005435
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CNA2007100871464A Pending CN101271458A (en) | 2007-03-22 | 2007-03-22 | Distributed computation file management method, system and device thereof |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101271458A (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102045389A (en) * | 2010-11-25 | 2011-05-04 | 深圳市科陆电子科技股份有限公司 | Distributed data storage method based on OSGI |
CN102147809A (en) * | 2011-03-22 | 2011-08-10 | 曙光信息产业股份有限公司 | Parallel file system and management method thereof |
CN102385507A (en) * | 2010-08-27 | 2012-03-21 | 卡斯柯信号有限公司 | Multi-machine information processing method and device for automatic generation of Boolean algebra |
CN102436501A (en) * | 2011-12-02 | 2012-05-02 | 曙光信息产业(北京)有限公司 | Parallel file managing system based on web |
CN103838761A (en) * | 2012-11-23 | 2014-06-04 | 阿里巴巴集团控股有限公司 | Metadata updating method and device |
CN105740348A (en) * | 2016-01-25 | 2016-07-06 | 武汉精伦电气有限公司 | Key log information extraction method and system based on distributed terminal system |
CN107194271A (en) * | 2017-04-18 | 2017-09-22 | 华南农业大学 | A kind of shared private cloud storage system of weak center |
CN108021578A (en) * | 2016-11-03 | 2018-05-11 | 北京国双科技有限公司 | The relation query method and device of data file |
CN111212138A (en) * | 2019-12-31 | 2020-05-29 | 曙光信息产业(北京)有限公司 | Cross-site storage system and data information access method |
-
2007
- 2007-03-22 CN CNA2007100871464A patent/CN101271458A/en active Pending
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102385507A (en) * | 2010-08-27 | 2012-03-21 | 卡斯柯信号有限公司 | Multi-machine information processing method and device for automatic generation of Boolean algebra |
CN102385507B (en) * | 2010-08-27 | 2014-12-24 | 卡斯柯信号有限公司 | Multi-machine information processing method and device for automatic generation of Boolean algebra |
CN102045389A (en) * | 2010-11-25 | 2011-05-04 | 深圳市科陆电子科技股份有限公司 | Distributed data storage method based on OSGI |
CN102147809A (en) * | 2011-03-22 | 2011-08-10 | 曙光信息产业股份有限公司 | Parallel file system and management method thereof |
CN102436501A (en) * | 2011-12-02 | 2012-05-02 | 曙光信息产业(北京)有限公司 | Parallel file managing system based on web |
CN103838761A (en) * | 2012-11-23 | 2014-06-04 | 阿里巴巴集团控股有限公司 | Metadata updating method and device |
CN103838761B (en) * | 2012-11-23 | 2017-03-01 | 阿里巴巴集团控股有限公司 | A kind of update method of metadata and device |
CN105740348A (en) * | 2016-01-25 | 2016-07-06 | 武汉精伦电气有限公司 | Key log information extraction method and system based on distributed terminal system |
CN108021578A (en) * | 2016-11-03 | 2018-05-11 | 北京国双科技有限公司 | The relation query method and device of data file |
CN107194271A (en) * | 2017-04-18 | 2017-09-22 | 华南农业大学 | A kind of shared private cloud storage system of weak center |
CN111212138A (en) * | 2019-12-31 | 2020-05-29 | 曙光信息产业(北京)有限公司 | Cross-site storage system and data information access method |
CN111212138B (en) * | 2019-12-31 | 2022-11-22 | 曙光信息产业(北京)有限公司 | Cross-site storage system and data information access method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101271458A (en) | Distributed computation file management method, system and device thereof | |
CN101667144B (en) | Virtual machine communication method based on shared memory | |
CN108027805B (en) | Load distribution in a data network | |
US9386117B2 (en) | Server side data cache system | |
US10831612B2 (en) | Primary node-standby node data transmission method, control node, and database system | |
US20150215405A1 (en) | Methods of managing and storing distributed files based on information-centric network | |
CN102307206B (en) | Caching system and caching method for rapidly accessing virtual machine images based on cloud storage | |
CN101222490B (en) | Network server system and method | |
CN102209087A (en) | Method and system for MapReduce data transmission in data center having SAN | |
WO2011134875A1 (en) | Data center operation | |
CN102541990A (en) | Database redistribution method and system utilizing virtual partitions | |
US10708379B1 (en) | Dynamic proxy for databases | |
JP7012079B2 (en) | Communication method, system, program between on-premises and off-premises | |
CN115470156A (en) | RDMA-based memory use method, system, electronic device and storage medium | |
CN103442034A (en) | Streaming media service method and system based on cloud computing technology | |
US9787755B2 (en) | Method and device for browsing network data, and storage medium | |
CN107908713A (en) | A kind of distributed dynamic cuckoo filtration system and its filter method based on Redis clusters | |
Chen et al. | Big data generation and acquisition | |
KR100718271B1 (en) | Embedded system employing a software platform for sharing hardware device and content resources | |
CN107357532A (en) | A kind of new cache pre-reading implementation method of new cluster-based storage | |
US7698424B1 (en) | Techniques for presenting multiple data storage arrays to iSCSI clients as a single aggregated network array | |
CN111030856B (en) | Cloud-based data access method, electronic device and computer readable medium | |
CN111625701B (en) | Searching method, searching device, server and storage medium | |
Qian | Design and evaluation of efficient collective communications on modern interconnects and multi-core clusters | |
CN101572689A (en) | Method and device for transmitting data between network interface card and accelerators in multi-processor system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 1120886 Country of ref document: HK |
|
C12 | Rejection of a patent application after its publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20080924 |
|
REG | Reference to a national code |
Ref country code: HK Ref legal event code: WD Ref document number: 1120886 Country of ref document: HK |