CN102130959A - System and method for scheduling cloud storage resource - Google Patents
System and method for scheduling cloud storage resource Download PDFInfo
- Publication number
- CN102130959A CN102130959A CN2011100695717A CN201110069571A CN102130959A CN 102130959 A CN102130959 A CN 102130959A CN 2011100695717 A CN2011100695717 A CN 2011100695717A CN 201110069571 A CN201110069571 A CN 201110069571A CN 102130959 A CN102130959 A CN 102130959A
- Authority
- CN
- China
- Prior art keywords
- server
- information
- storage
- distributed cache
- resource information
- 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
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000013468 resource allocation Methods 0.000 claims description 8
- 238000005516 engineering process Methods 0.000 description 4
- 238000007726 management method Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000002452 interceptive effect Effects 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 230000003993 interaction Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 206010033799 Paralysis Diseases 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a system and a method for scheduling a cloud storage resource. The system at least comprises a plurality of storage servers and a distributed cache server; the plurality of storage servers are used for feeding storage resource information of the server back to the distributed cache server via a distributed object cache interface at set intervals; and the distributed cache server is used for storing the storage resource information fed back by each storage server. Through a protocol interface of a distributed cache memory object, the resource information of each storage node is organized into a data format according with the protocol interface to be fed back quickly and timely, so that a cloud storage system can reliably and rationally schedule the storage resource.
Description
Technical field
The present invention relates to cloud storage system, relate in particular to the system and method for cloud storage resources scheduling.
Background technology
Cloud storage system is meant by functions such as cluster application, grid or distributed file systems, a large amount of various dissimilar memory devices in the network are gathered collaborative work by application software, a system of storage and Operational Visit function externally is provided jointly.The storage of use cloud is not to use some memory devices, and a kind of data access service that is to use whole cloud storage system to bring.The core of cloud storage is that application software is combined with memory device, realizes the transformation of memory device to stores service by application software.
In a single day the storage resources of cloud storage system can be served as by exclusive storage server node, but has unnecessary storage resources to be added to high in the clouds in order to realize the notion of " cloud " better, to advocate, use just can be scheduled.So cloud storage system is not generally done exclusive restriction to storage resources, can be storage resources unnecessary in ordinary individual's computer, also can be storage resources unnecessary in the exclusive server.
The cloud storage system framework roughly can be divided into resource feedback layer, scheduling of resource layer and service broker's layer, and wherein, the resource feedback layer is mainly used in and obtains and write down the storage resources information of respectively storing child node; The scheduling of resource layer is mainly used in and utilizes the resource information of the storage server that the resource feedback layer obtains to carry out resource query, calculating and scheduling, and corresponding storage server resource allocation is used to requesting users; The request that service broker's layer then is mainly used in the user splits, and then needs to utilize the scheduling of resource layer to distribute corresponding storage server resource at new user, responds user's request by corresponding storage server then; At old user's reading and writing data request, will ask directly to send to the respective stored server according to the resource information of scheduling of resource layer record and handle.In the cloud storage system framework, the strategy that adopts data flow to separate with management is realized visit and the management to data, and this point is embodied in service broker's layer of the superiors.
Because the cloud storage system framework separates data flow by above-mentioned layered mode with management, so when thousands of memory nodes are arranged in the system, each memory node needs feedback resource information separately in time, so that system upper strata corresponding storage resources of reasonable distribution when carrying out scheduling of resource.Therefore, how in time or apace to obtain the information of each storage server resource node, just seem very important.
At present, some comparatively ripe cloud storage systems are (as Amazon Simple Storage System, Amazon simple storage system) is commercial system, its ins and outs are often externally underground, for example the layer architecture of above-mentioned cloud storage system is how specifically to implement to the feedback of the resource information of thousands of memory nodes and according to the resource information reasonable distribution of feedback and dispatch storage resources, and is all unexposed; Because the enforcement difficulty of high reliability cloud storage scheme is bigger, therefore for some users that the cloud storage demand is arranged, need the more cost of cost.
Summary of the invention
Technical problem to be solved by this invention provides a kind of system and method for realizing the scheduling of cloud storage resources, can in time get access to storage resources information funding source rational management.
In order to solve the problems of the technologies described above, the invention provides a kind of system that realizes the scheduling of cloud storage resources, comprise a plurality of storage servers and distributed cache server at least, wherein:
Storage server is used at set intervals the storage resources information of book server is fed back to distributed cache server by the distributed objects cache interface;
Distributed cache server is used to preserve the storage resources information that each storage server feeds back.
Further, the distributed objects cache interface is the Memcached protocol interface, promptly
Storage server regularly obtains the resource information of interior disk of book server and file system, and the line command that this resource information is organized into Memcached protocol interface form feeds back to distributed cache server.
Further, line command comprises following parameter at least:
Command type is used for the type of directive command, comprises that information obtains order, information and add order, information updating order and information and replace in the order one or more;
Key name is used for resources bank of unique identification, perhaps indicates the host address and the port of storage server;
Key assignments, the host address and the port of indication storage server when key name identifying resource storehouse provide the resource status value of corresponding storage server when key name indication host address and port;
The order end mark is used to indicate line command to finish.
Further, line command also comprises the time-out time parameter, is used to indicate the valid expiration date of resource information on distributed cache server; When this time-out time parameter is made as 0, effectively, when this time-out time parameter was made as greater than 0 integer k, this k represented the time-out time of resource information on distributed cache server to the expression resource information forever on distributed cache server.
Further, line command also comprises the data length parameter, is used to indicate the data of resource information or the length of blocks of data; Behind the order end mark of the order line that contains this data length parameter, also comprise the data or the blocks of data content of resource information.
Further,
Also initiatively send information to storage server obtains instruction to distributed cache server, the storage resources information of record storage server feedback;
Storage server also obtains the storage resources information of instruction by distributed objects cache interface feedback book server according to described information.
Further, this method also comprises:
Resource allocation server is used for visiting the resources bank of distributed cache server, the storage resources information of each storage server in the query resource storehouse, and carry out stores service according to the storage resources of the corresponding storage server of storage resources message scheduling of inquiry.
In order to solve the problems of the technologies described above, the invention provides a kind of storage server that is used for realizing as previously mentioned the system of cloud storage resources scheduling, comprise built-in resource feedback module, be used at set intervals the storage resources information of book server being fed back to by the distributed objects cache interface distributed cache server of this system.
Further, the distributed objects cache interface is the Memcached protocol interface, promptly
The resource feedback module regularly obtains the resource information of interior disk of book server and file system, and the line command that resource information is organized into Memcached protocol interface form feeds back to distributed cache server, and line command comprises following parameter at least:
Command type is used for the type of directive command, comprises that information obtains order, information and add order, information updating order and information and replace in the order one or more;
Key name is used for resources bank of unique identification, perhaps indicates the host address and the port of storage server;
Key assignments, the host address and the port of indication storage server when described key name identifying resource storehouse provide the resource status value of corresponding storage server when key name indication host address and port;
The order end mark is used to indicate line command to finish.
Further, line command also comprises time-out time and/or data length parameter, wherein:
Time-out time is used to indicate the valid expiration date of resource information on distributed cache server; When this time-out time parameter is made as 0, effectively, when this time-out time parameter was made as greater than 0 integer k, this k represented the time-out time of resource information on distributed cache server to the expression resource information forever on distributed cache server;
Data length is used to indicate the data of resource information or the length of blocks of data; Behind the order end mark of the order line that contains this data length parameter, also comprise the data or the blocks of data content of resource information.
In order to solve the problems of the technologies described above, the invention provides a kind of method that realizes the scheduling of cloud storage resources, relate to a plurality of storage servers and distributed cache server at least, this method comprises:
Storage server feeds back to distributed cache server with the storage resources information of book server by the distributed objects cache interface at set intervals;
Distributed cache server is preserved the storage resources information of each storage server feedback.
Further, the distributed objects cache interface is the Memcached protocol interface, and storage server feeds back to distributed cache server with the storage resources information of book server by the distributed objects cache interface at set intervals, specifically comprises:
Storage server regularly obtains the resource information of interior disk of book server and file system, and the line command that resource information is organized into Memcached protocol interface form feeds back to distributed cache server.
Further, line command comprises following parameter at least:
Command type is used for the type of directive command, comprises that information obtains order, information and add order, information updating order and information and replace in the order one or more;
Key name is used for resources bank of unique identification, perhaps indicates the host address and the port of storage server;
Key assignments, the host address and the port of indication storage server when key name identifying resource storehouse provide the resource status value of corresponding storage server when key name indication host address and port;
The order end mark is used to indicate line command to finish.
Further, line command also comprises time-out time and/or data length parameter, wherein:
Time-out time is used to indicate the valid expiration date of resource information on described distributed cache server; When this time-out time parameter is made as 0, represent described resource information on distributed cache server effectively forever, when this time-out time parameter was made as greater than 0 integer k, this k represented the time-out time of resource information on distributed cache server;
Data length is used to indicate the data of resource information or the length of blocks of data; Behind the order end mark of the order line that contains this data length parameter, also comprise the data or the blocks of data content of resource information.
Further, storage server feeds back to distributed cache server with the line command that resource information is organized into Memcached protocol interface form, specifically comprises:
Storage server is if judge that book server is the memory node that increases newly, and obtain order by information and know it is first feedback resources information all storage servers from distributed cache server, then add the order indication and add the resources bank of all node resource information of sign, and the resource information that will regularly obtain is updated to resources bank by information updating order indication by information; If obtaining order by information knows that book server is not first feedback resources information, then upgrade the storage resources information of resources bank by information updating order indication, then the resource information of regularly obtaining is added the order indication by information and add resources bank to;
Storage server is if judge that book server is not the memory node that increases newly, and then the resource information that will regularly obtain is updated to resources bank by information updating order indication.
Further, this method also comprises:
Storage server then feeds back to distributed cache server with resource information again if judge that the resource information feedback is unsuccessful, perhaps resource information is fed back to standby distributed cache server, feeds back successfully until resource information.
Further, this method also relates to resource allocation server, and this method also comprises:
Resource allocation server is visited the resources bank in the described distributed cache server, the storage resources information of each storage server in the query resource storehouse, and carry out stores service according to the storage resources of the corresponding storage server of storage resources message scheduling of inquiry.
The present invention adopts the protocol interface of distributed caching memory object, the resource information of each memory node is organized into the data format that meets this protocol interface and timely feedback apace, thereby makes cloud storage system can reliably, reasonably dispatch storage resources.
Description of drawings
Fig. 1 is the configuration diagram of the system embodiment of realization cloud storage resources scheduling of the present invention;
Fig. 2 is the flow chart of the method embodiment of realization cloud storage resources scheduling of the present invention.
Embodiment
Below in conjunction with accompanying drawing and preferred embodiment technical scheme of the present invention is at length set forth.The embodiment that below exemplifies only is used for description and interpretation the present invention, and does not constitute the restriction to technical solution of the present invention.
The framework of the system embodiment of realization cloud storage resources of the present invention scheduling as shown in Figure 1, comprise at least a plurality of storage servers (1～N) and distributed cache server, wherein:
A plurality of storage servers (1～N), be used at set intervals the storage resources information of book server is fed back to distributed cache server by the distributed objects cache interface;
Distributed cache server is used for the storage resources information record storage with each storage server feedback.
Perhaps, initiatively send information to each storage server obtains instruction to distributed cache server, writes down the storage resources information of each storage server feedback;
(1～N) obtains instruction according to the information of distributed cache server passes through the storage resources information that the distributed objects cache interface feeds back book server to a plurality of storage servers.
Distributed objects cache interface among the said system embodiment preferably adopts the Memcached protocol interface, be that each storage server regularly or according to the regular feedback information instruction of distributed cache server is made into the command format that meets the Memcached protocol interface with the storage resources information sets of book server, feed back to distributed cache server, can realize the high-speed feedback of storage resources information thus.
Memcached is a kind of caching system that adopts the distributed caching object, promptly can allow a plurality of different main frames to visit this caching system simultaneously, mainly be used in and carry out the speed that the website front end improves visit data in the Internet, at be distributed system (distributed data base and distributed computing system).And each memory node of cloud storage also is distributed, so be fit to quote the Memcached caching system.Because the Memcached caching system increases income,, can realize as long as meet the Memcache protocol interface that this caching system supports therefore when will or writing data by this caching system inquiry.
Certainly, the present invention is used for the information interaction between a plurality of memory nodes and the distributed cache server except adopting the Memcached protocol interface, also can adopt other distributed objects cache interface with the Memcached protocol interface compatibility to carry out information interaction, such as Redis interface TT (Tokyo Tyrant) interface etc.
Each storage server is as the client of Memcached, by built-in resource feedback module (being a kind of Memcached client-side program that the present invention develops), regularly get access to disk and the relevant resource information of file system on the book server, adopt transmission control protocol (TCP, Transfer Control Protocol) or User Datagram Protoco (UDP) (UDP, User Datagram Protocol) set up connection to distributed cache server feedback resources information;
Distributed cache server need be set up a plurality of (the hundreds of or thousands of) information interactive interface that can open simultaneously, in order to work efficiently, by these information interactive interface of buffer memory, can eliminate the expense of setting up them and bringing.When each client no longer needs the distributed caching service, but the closing information interactive interface.
Draw together two kinds by data packets for transmission in the memcache agreement: line of text and free data; Wherein line of text is used to pass the resource information feedback command of carrying storage server and the response of distributed cache server, and it finishes with " r n " (after carriage return character immediately following a newline) expression; Free data then are used for storage server from the distributed cache server access data, and it must determine end position by the block length of indicating in the line of text.
The built-in resource feedback module of storage server adopts the line command resource information of following form:
＜command?name＞＜key＞＜flags＞＜exptime＞＜bytes＞\r\n
Wherein, the implication of each parameter in this memory command form is expressed in the following table.
Table 1
The present invention except adopt the carriage return character in conjunction with newline as the line command end mark, also can adopt other symbol as the line command end mark, for example adopt branch "; " etc.
Parameter key, value all have two kinds of indicative functions in the table 1, are illustrated in the following table particularly.
Table 2
Hostname:port by parameter key indication, be that the built-in resource feedback module of storage server is from adopting file transfer protocol (FTP) (FTP, File Transfer Protocol) obtains in the configuration file, wherein hostname is the FTP address that the indication distributed cache server need be monitored, and port then is the FTP port that the indication distributed cache server will be monitored.
By the resource status of parameter value indication memory node (being storage server), the resource status value representation of included one or more is in following table in the table 2.
Table 3
| The resource status value | The state value implication |
| F_BSIZE | Be used to indicate the block size of file system |
| F_FRSIZE | Be used to indicate the chip size of file system |
| F_BLOCKS | Be used for indicating file system piece number altogether |
| F_BFREE | Be used for indicating the piece number of current free time of file system |
| F_BAVAIL | Be used to indicate the current maximum available block number of the non-root user of file system |
| F_FILES | Be used to the maximum number of files of indicating file system to support |
| F_FFREE | Be used to indicate the current maximum file available number of file system |
| F_FAVAIL | Be used to indicate the current maximum file available number of the non-root user of file system |
| F_FLAG | Be used to indicate the carry option of file system |
| F_NAMEMAX | Be used to indicate the maximum filename length of file system |
| BYTEMAX | Be used to indicate file system byte number altogether |
| BYTEAVAIL | Be used to indicate the byte number of file system current residual |
| TYPE | Be used to indicate file system type (some attribute of the file system of artificial regulation) |
The system embodiment of the realization cloud storage resources scheduling of the invention described above also comprises:
Resource allocation server is used to visit the resources bank that is kept at distributed cache server, the resource information of each storage server in the query resource storehouse, and carry out stores service according to the resource that the resource information of inquiry is dispatched corresponding storage server.
Below will specify the usage of each parameter in the mentioned order line of text at flow process shown in Figure 2.
Fig. 2 represents is the flow process of the method embodiment of realization cloud storage resources scheduling of the present invention, be storage server set up with distributed cache server between be connected after carry out, comprise the steps:
110: judge whether this storage server is newly-increased node, is then to carry out the following step, otherwise execution in step 140;
120: judge whether this storage server is first feedback server, is then to carry out the following step, otherwise execution in step 170;
Storage server carries out above-mentioned judgement by sending as give an order to distributed cache server:
get?SS\r\n
Wherein SS promptly represents the unique identification of resources bank as the key name value of parameter key; If storage server obtains the SS of distributed cache server response for empty, is the storage server of first feedback resources information in the resources bank just know book server; Otherwise, be not empty if the response that storage server obtains is SS, not the storage server of first feedback resources information in the resources bank just know book server.
130: the indication distributed cache server adds the resources bank of all node resource information of sign;
Storage server carries out above-mentioned indication by sending instruction to distributed cache server:
add?SS?0?data_len\r\n
block_data\r\n
Wherein, add order add indication distributed cache server and add the resources bank that a name is called SS, the exptime parameter of this resources bank is made as 0, and the resources bank that expression is added is effective forever; Data_len is the byte value of consult volume, the length (is unit with byte) of expression back information data; The block_data of next line represents information data, is expressed as the Hostname and the port hostname:port thereof of storage server at this, such as 192.168.0.1:2020.Owing to may have a plurality of storage resources (for example 2020 ports and 3030 ports are all opened up the FTP stores service) on same the storage server, therefore need indicate a plurality of hostname:port at a plurality of storage resources, for example, 192.168.0.1:2020,192.168.0.1:3030.
140: collect the resource information of this storage server, form memcache protocol interface set order and feed back to distributed cache server;
Storage server feeds back by sending instruction to distributed cache server:
add?hostname:port?060data?len\r\n
block_data\r\n
Wherein, the Hostname and the port thereof that add a memory node of order add indication distributed cache server interpolation are the resource information of hostname:port, the exptime parameter of this resources bank is made as 60, be illustrated in time-out time 60 seconds then, if distributed cache server is not received this storage server and upgraded feedback resources information, its corresponding resource information is deleted just think this storage server to break down; Data_len is the byte value of consult volume, the length (is unit with byte) of expression back information data; The block_data of next line is the resource information data of the one or more feedbacks shown in the table 3.
150: judging whether feedback information is successful, is then to close connection, ending message feedback; Otherwise execution next step;
160: feed back again or feed back to standby distributed cache server, return step 150 and carry out;
Storage server is not for example when exceeding the predetermined anti-number of times of weight, again send corresponding feedback command to original distributed cache server, after exceeding the anti-number of times of predetermined weight if do not feed back successfully yet, then send corresponding feedback command, until feeding back successfully to standby distributed cache server.
170: the indication distributed cache server upgrades the storage resources information of resources bank;
Storage server carries out above-mentioned indication by sending instruction to distributed cache server:
set?SS?0?data_len\r\n
block_data\r\n
Wherein, update command set indication distributed cache server more newname be the resource information of a storage server in the resources bank of SS; Data_len is the byte value of consult volume, the length (is unit with byte) of expression back information data; The block_data of next line represents information data, is expressed as the Hostname and the port hostname:port thereof of this storage server at this, such as 192.168.0.1:2020.
180: collect the resource information of book server, form memcache protocol interface add order and feed back to distributed cache server, execution in step 150.
Storage server carries out above-mentioned feedback by sending instruction to distributed cache server:
add?hostname:port?0?60?data?len\r\n
block_data\r\n
The implication of each parameter is described out in step 140 in the order, and this repeats no more.
Said method embodiment also comprises step:
Dispatch server visit is kept at the resources bank of distributed cache server, the resource information of each storage server in the query resource storehouse, and carry out stores service according to the resource that the resource information of inquiry is dispatched corresponding storage server.
For example, dispatch server sends to distributed cache server and carries out above-mentioned visit with down order:
get?SS\r\n
Can obtain the storage server information in the resources bank that distributed cache server returns:
192.168.0.1:2020
192.168.0.2:2020
That is to say has this two storage servers in the present resources bank, if will continue to inquire about the resource information of seeing storage server wherein, then continue to inquire about with down order to the distributed cache server transmission:
get?192.168.0.1:2020
Can obtain the resource information of the corresponding stores service implement body that distributed cache server returns, such as the one or more resource status values in the table 3.
The present invention adopts the resource information of each storage server in the distributed caching system cache cloud storage system, can accomplish high-speed cache by the Memcached interface on speed.Because the storage resources feedback information itself is second a level, even if therefore the distributed cache server paralysis also can not have influence on the performance of the dispatching patcher on cloud storage system upper strata.
More than explanation only is the preferable embodiment of the present invention; but protection scope of the present invention is not limited thereto; anyly be familiar with those skilled in the art in the technical scope that the present invention discloses, the variation that can expect easily or replacement all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with the protection range of claims.
Claims (17)
1. a system that realizes the scheduling of cloud storage resources comprises a plurality of storage servers and distributed cache server at least, wherein:
Storage server is used at set intervals the storage resources information of book server is fed back to distributed cache server by the distributed objects cache interface;
Distributed cache server is used to preserve the storage resources information that each storage server feeds back.
2. according to the described system of claim 1, it is characterized in that described distributed objects cache interface is the Memcached protocol interface, promptly
Described storage server regularly obtains the resource information of interior disk of book server and file system, and the line command that described resource information is organized into described Memcached protocol interface form feeds back to described distributed cache server.
3. according to the described system of claim 2, it is characterized in that described line command comprises following parameter at least:
Command type is used for the type of directive command, comprises that information obtains order, information and add order, information updating order and information and replace in the order one or more;
Key name is used for resources bank of unique identification, perhaps indicates the host address and the port of described storage server;
Described key assignments is indicated the host address and the port of described storage server when described key name identifies described resources bank, provide the resource status value of corresponding storage server when described key name is indicated described host address and port;
The order end mark is used to indicate described line command to finish.
4. according to the described system of claim 3, it is characterized in that described line command also comprises the time-out time parameter, be used to indicate the valid expiration date of described resource information on described distributed cache server; When this time-out time parameter is made as 0, represent that described resource information is effective forever on described distributed cache server, when this time-out time parameter was made as greater than 0 integer k, this k represented the time-out time of described resource information on described distributed cache server.
5. according to the described system of claim 4, it is characterized in that described line command also comprises the data length parameter, be used to indicate the data of described resource information or the length of blocks of data; Behind the order end mark of the order line that contains this data length parameter, also comprise the data or the blocks of data content of described resource information.
6. according to each described system of claim 1 to 5, it is characterized in that,
Also initiatively send information to described storage server obtains instruction to described distributed cache server, writes down the storage resources information of described storage server feedback;
Described storage server also obtains the storage resources information of instruction by described distributed objects cache interface feedback book server according to described information.
7. according to each described system of claim 1 to 5, it is characterized in that, also comprise:
Resource allocation server, be used for visiting the resources bank of described distributed cache server, inquire about the storage resources information of each storage server in the described resources bank, and carry out stores service according to the storage resources of the corresponding storage server of inquiring about of described storage resources message scheduling.
8. storage server that is used for the system as claimed in claim 1, it is characterized in that, comprise built-in resource feedback module, be used at set intervals the storage resources information of book server being fed back to by the distributed objects cache interface distributed cache server of described system.
9. according to the described storage server of claim 8, it is characterized in that described distributed objects cache interface is the Memcached protocol interface, promptly
Described resource feedback module regularly obtains the resource information of interior disk of book server and file system, the line command that described resource information is organized into described Memcached protocol interface form feeds back to described distributed cache server, and described line command comprises following parameter at least:
Command type is used for the type of directive command, comprises that information obtains order, information and add order, information updating order and information and replace in the order one or more;
Key name is used for resources bank of unique identification, perhaps indicates the host address and the port of described storage server;
Described key assignments is indicated the host address and the port of described storage server when described key name identifies described resources bank, provide the resource status value of corresponding storage server when described key name is indicated described host address and port;
The order end mark is used to indicate described line command to finish.
10. according to according to the described storage server of claim 9, it is characterized in that described line command also comprises time-out time and/or data length parameter, wherein:
Time-out time is used to indicate the valid expiration date of described resource information on described distributed cache server; When this time-out time parameter is made as 0, represent that described resource information is effective forever on described distributed cache server, when this time-out time parameter was made as greater than 0 integer k, this k represented the time-out time of described resource information on described distributed cache server;
Data length is used to indicate the data of described resource information or the length of blocks of data; Behind the order end mark of the order line that contains this data length parameter, also comprise the data or the blocks of data content of described resource information.
11. a method that realizes the scheduling of cloud storage resources relates to a plurality of storage servers and distributed cache server at least, this method comprises:
Storage server feeds back to distributed cache server with the storage resources information of book server by the distributed objects cache interface at set intervals;
Distributed cache server is preserved the storage resources information of each storage server feedback.
12. in accordance with the method for claim 11, it is characterized in that, described distributed objects cache interface is the Memcached protocol interface, described storage server feeds back to distributed cache server with the storage resources information of book server by the distributed objects cache interface at set intervals, specifically comprises:
Described storage server regularly obtains the resource information of interior disk of book server and file system, and the line command that described resource information is organized into described Memcached protocol interface form feeds back to described distributed cache server.
13. in accordance with the method for claim 12, it is characterized in that described line command comprises following parameter at least:
Command type is used for the type of directive command, comprises that information obtains order, information and add order, information updating order and information and replace in the order one or more;
Key name is used for resources bank of unique identification, perhaps indicates the host address and the port of described storage server;
Described key assignments is indicated the host address and the port of described storage server when described key name identifies described resources bank, provide the resource status value of corresponding storage server when described key name is indicated described host address and port;
The order end mark is used to indicate described line command to finish.
14. in accordance with the method for claim 13, it is characterized in that described line command also comprises time-out time and/or data length parameter, wherein:
Time-out time is used to indicate the valid expiration date of described resource information on described distributed cache server; When this time-out time parameter is made as 0, represent that described resource information is effective forever on described distributed cache server, when this time-out time parameter was made as greater than 0 integer k, this k represented the time-out time of described resource information on described distributed cache server;
Data length is used to indicate the data of described resource information or the length of blocks of data; Behind the order end mark of the order line that contains this data length parameter, also comprise the data or the blocks of data content of described resource information.
15. in accordance with the method for claim 14, it is characterized in that described storage server feeds back to described distributed cache server with the line command that described resource information is organized into described Memcached protocol interface form, specifically comprises:
Described storage server is if judge that book server is the memory node that increases newly, and obtain order by described information and know it is first feedback resources information all storage servers from described distributed cache server, then add the order indication and add the resources bank of all node resource information of sign, and the described resource information that will regularly obtain is updated to described resources bank by described information updating order indication by described information; Know that book server is not first feedback resources information if obtain order by described information, then upgrade the storage resources information of described resources bank, then the described resource information of regularly obtaining is added the order indication by described information and add described resources bank to by described information updating order indication;
Described storage server is if judge that book server is not the memory node that increases newly, and the described resource information that then will regularly obtain is updated to described resources bank by described information updating order indication.
16. according to each described method of claim 11 to 15, it is characterized in that, also comprise:
Described storage server then feeds back to described distributed cache server with resource information again if judge that described resource information feedback is unsuccessful, perhaps described resource information is fed back to standby distributed cache server, feeds back successfully until described resource information.
17., it is characterized in that also relate to resource allocation server, this method also comprises according to each described method of claim 11 to 15:
Resource allocation server is visited the resources bank in the described distributed cache server, inquire about the storage resources information of each storage server in the described resources bank, and carry out stores service according to the storage resources of the corresponding storage server of inquiring about of described storage resources message scheduling.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011100695717A CN102130959A (en) | 2011-03-22 | 2011-03-22 | System and method for scheduling cloud storage resource |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011100695717A CN102130959A (en) | 2011-03-22 | 2011-03-22 | System and method for scheduling cloud storage resource |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102130959A true CN102130959A (en) | 2011-07-20 |
Family
ID=44268848
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011100695717A Pending CN102130959A (en) | 2011-03-22 | 2011-03-22 | System and method for scheduling cloud storage resource |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102130959A (en) |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102377827A (en) * | 2011-12-13 | 2012-03-14 | 方正国际软件有限公司 | Multilevel cloud storage system and storage method thereof |
| CN102857546A (en) * | 2012-03-20 | 2013-01-02 | 深圳市朗科科技股份有限公司 | Cloud storage platform and cloud storage terminal |
| CN103491197A (en) * | 2013-10-12 | 2014-01-01 | 北京海联捷讯信息科技发展有限公司 | Distributed automatic inspection system and resource collecting method thereof |
| CN103841129A (en) * | 2012-11-21 | 2014-06-04 | 华为技术有限公司 | Cloud computing resource information acquisition server, cloud computing resource information acquisition client and information processing method |
| CN104283913A (en) * | 2013-07-04 | 2015-01-14 | 高宏亮 | Network storage system and method |
| WO2015196369A1 (en) * | 2014-06-24 | 2015-12-30 | 华为技术有限公司 | Distributed cache control method and apparatus |
| CN105721543A (en) * | 2015-06-15 | 2016-06-29 | 清华大学深圳研究生院 | Cross-protocol distribution-type cloud storage system based on operation control unit and data management method |
| CN105893031A (en) * | 2016-03-28 | 2016-08-24 | 广州华多网络科技有限公司 | Cache operation implementation method, calling method of business layer method and device |
| CN105978744A (en) * | 2016-07-26 | 2016-09-28 | 浪潮电子信息产业股份有限公司 | Resource allocation method, device and system |
| WO2017092414A1 (en) * | 2015-12-03 | 2017-06-08 | 杭州海康威视数字技术股份有限公司 | Data cloud storage system, client terminal, storage server and application method |
| CN106959820A (en) * | 2016-01-11 | 2017-07-18 | 杭州海康威视数字技术股份有限公司 | A kind of data extraction method and system |
| CN107463598A (en) * | 2017-06-09 | 2017-12-12 | 中国邮政储蓄银行股份有限公司 | Distributed cache system |
| CN108769244A (en) * | 2018-06-14 | 2018-11-06 | 郑州云海信息技术有限公司 | A kind of store tasks information acquisition method and relevant apparatus |
| CN108810043A (en) * | 2017-04-28 | 2018-11-13 | 北京云中融信网络科技有限公司 | A kind of method and apparatus of storage comment information |
| CN108880876A (en) * | 2018-05-31 | 2018-11-23 | 郑州云海信息技术有限公司 | Storage cluster dispositions method, device, equipment and computer readable storage medium |
| CN109639752A (en) * | 2018-10-23 | 2019-04-16 | 深圳壹账通智能科技有限公司 | Method, apparatus, computer equipment and the storage medium that agency is scheduled |
| CN110113176A (en) * | 2018-02-01 | 2019-08-09 | 北京京东尚科信息技术有限公司 | Information synchronization method and device for configuration server |
| CN111131440A (en) * | 2019-12-20 | 2020-05-08 | 腾讯科技（深圳）有限公司 | Resource scheduling method and device and server |
| CN114390069A (en) * | 2022-01-30 | 2022-04-22 | 青岛海尔科技有限公司 | Data access method, system, equipment and storage medium based on distributed cache |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101193079A (en) * | 2006-11-29 | 2008-06-04 | 索尼株式会社 | Data management server, data management system, data management method and data management program |
| US20100325199A1 (en) * | 2009-06-22 | 2010-12-23 | Samsung Electronics Co., Ltd. | Client, brokerage server and method for providing cloud storage |
-
2011
- 2011-03-22 CN CN2011100695717A patent/CN102130959A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101193079A (en) * | 2006-11-29 | 2008-06-04 | 索尼株式会社 | Data management server, data management system, data management method and data management program |
| US20100325199A1 (en) * | 2009-06-22 | 2010-12-23 | Samsung Electronics Co., Ltd. | Client, brokerage server and method for providing cloud storage |
Non-Patent Citations (1)
| Title |
|---|
| 曾海军，詹舒波: "Memcached在呼叫中心系统中的应用", 《软件 COMPUTER ENGINEERING & SOFTWARE》, vol. 31, no. 10, 31 December 2010 (2010-12-31), pages 62 - 64 * |
Cited By (33)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102377827A (en) * | 2011-12-13 | 2012-03-14 | 方正国际软件有限公司 | Multilevel cloud storage system and storage method thereof |
| CN102857546B (en) * | 2012-03-20 | 2016-09-28 | 深圳市朗科科技股份有限公司 | Cloud storage platform and cloud storage terminal |
| CN102857546A (en) * | 2012-03-20 | 2013-01-02 | 深圳市朗科科技股份有限公司 | Cloud storage platform and cloud storage terminal |
| WO2013139153A1 (en) * | 2012-03-20 | 2013-09-26 | 深圳市朗科科技股份有限公司 | Cloud storage platform and cloud storage terminal |
| CN103841129A (en) * | 2012-11-21 | 2014-06-04 | 华为技术有限公司 | Cloud computing resource information acquisition server, cloud computing resource information acquisition client and information processing method |
| CN103841129B (en) * | 2012-11-21 | 2017-04-26 | 华为技术有限公司 | Cloud computing resource information acquisition server, cloud computing resource information acquisition client and information processing method |
| CN104283913A (en) * | 2013-07-04 | 2015-01-14 | 高宏亮 | Network storage system and method |
| CN103491197A (en) * | 2013-10-12 | 2014-01-01 | 北京海联捷讯信息科技发展有限公司 | Distributed automatic inspection system and resource collecting method thereof |
| CN103491197B (en) * | 2013-10-12 | 2016-08-10 | 北京海联捷讯信息科技发展有限公司 | Distributed automatic tour inspection system and resource collection method thereof |
| WO2015196369A1 (en) * | 2014-06-24 | 2015-12-30 | 华为技术有限公司 | Distributed cache control method and apparatus |
| CN105393518A (en) * | 2014-06-24 | 2016-03-09 | 华为技术有限公司 | Distributed cache control method and apparatus |
| CN105393518B (en) * | 2014-06-24 | 2019-03-19 | 华为技术有限公司 | Distributed cache control method and device |
| CN105721543A (en) * | 2015-06-15 | 2016-06-29 | 清华大学深圳研究生院 | Cross-protocol distribution-type cloud storage system based on operation control unit and data management method |
| CN105721543B (en) * | 2015-06-15 | 2017-08-08 | 清华大学深圳研究生院 | Based on operation control unit across protocol distributed cloud storage system and data managing method |
| CN106850710B (en) * | 2015-12-03 | 2020-02-28 | 杭州海康威视数字技术股份有限公司 | Data cloud storage system, client terminal, storage server and application method |
| CN106850710A (en) * | 2015-12-03 | 2017-06-13 | 杭州海康威视数字技术股份有限公司 | A kind of safe data cloud storage system, client terminal, storage server and application process |
| WO2017092414A1 (en) * | 2015-12-03 | 2017-06-08 | 杭州海康威视数字技术股份有限公司 | Data cloud storage system, client terminal, storage server and application method |
| US10623516B2 (en) | 2015-12-03 | 2020-04-14 | Hangzhou Hikvision Digital Technology Co., Ltd. | Data cloud storage system, client terminal, storage server and application method |
| CN106959820A (en) * | 2016-01-11 | 2017-07-18 | 杭州海康威视数字技术股份有限公司 | A kind of data extraction method and system |
| CN105893031A (en) * | 2016-03-28 | 2016-08-24 | 广州华多网络科技有限公司 | Cache operation implementation method, calling method of business layer method and device |
| CN105978744A (en) * | 2016-07-26 | 2016-09-28 | 浪潮电子信息产业股份有限公司 | Resource allocation method, device and system |
| CN105978744B (en) * | 2016-07-26 | 2018-10-26 | 浪潮电子信息产业股份有限公司 | A kind of resource allocation methods, apparatus and system |
| CN108810043B (en) * | 2017-04-28 | 2021-11-02 | 北京云中融信网络科技有限公司 | Method and device for storing comment information |
| CN108810043A (en) * | 2017-04-28 | 2018-11-13 | 北京云中融信网络科技有限公司 | A kind of method and apparatus of storage comment information |
| CN107463598A (en) * | 2017-06-09 | 2017-12-12 | 中国邮政储蓄银行股份有限公司 | Distributed cache system |
| CN110113176A (en) * | 2018-02-01 | 2019-08-09 | 北京京东尚科信息技术有限公司 | Information synchronization method and device for configuration server |
| CN108880876A (en) * | 2018-05-31 | 2018-11-23 | 郑州云海信息技术有限公司 | Storage cluster dispositions method, device, equipment and computer readable storage medium |
| CN108769244A (en) * | 2018-06-14 | 2018-11-06 | 郑州云海信息技术有限公司 | A kind of store tasks information acquisition method and relevant apparatus |
| CN109639752A (en) * | 2018-10-23 | 2019-04-16 | 深圳壹账通智能科技有限公司 | Method, apparatus, computer equipment and the storage medium that agency is scheduled |
| CN109639752B (en) * | 2018-10-23 | 2022-07-19 | 深圳壹账通智能科技有限公司 | Method and device for scheduling agent, computer equipment and storage medium |
| CN111131440A (en) * | 2019-12-20 | 2020-05-08 | 腾讯科技（深圳）有限公司 | Resource scheduling method and device and server |
| CN114390069A (en) * | 2022-01-30 | 2022-04-22 | 青岛海尔科技有限公司 | Data access method, system, equipment and storage medium based on distributed cache |
| CN114390069B (en) * | 2022-01-30 | 2024-03-22 | 青岛海尔科技有限公司 | Data access method, system, equipment and storage medium based on distributed cache |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102130959A (en) | System and method for scheduling cloud storage resource | |
| CN107590001B (en) | Load balancing method and device, storage medium and electronic equipment | |
| CN105897946B (en) | A kind of acquisition methods and system of access address | |
| CN101493826B (en) | Database system based on WEB application and data management method thereof | |
| CN107818112B (en) | Big data analysis operating system and task submitting method | |
| JP4740897B2 (en) | Virtual network configuration method and network system | |
| CN103209223A (en) | Distributed application conversation information sharing method and system and application server | |
| CN102694861B (en) | A kind of terminal application software sorting technique, Apparatus and system based on cloud | |
| CN101902473B (en) | Method for synchronously updating data based on grid GIS (Geographic Information System) | |
| CN102609488B (en) | Client, data query method of client, server and data query system | |
| EP1758042B1 (en) | Document distribution system and method | |
| CN104954468A (en) | Resource allocation method and resource allocation device | |
| CN104731516A (en) | Method and device for accessing files and distributed storage system | |
| KR20100137323A (en) | Client, brokerage sever and method for providing cloud storage | |
| CN102722405B (en) | Counting method and system in high concurrent and multithreaded application | |
| CN102638584A (en) | Data distributing and caching method and data distributing and caching system | |
| CN104601736A (en) | Method and device for realizing short uniform resource locator (URL) service | |
| CN103607428A (en) | Method of accessing shared memory and apparatus thereof | |
| CN104536965B (en) | A kind of data query display systems under the conditions of big data and method | |
| CN104050248A (en) | File storage system and storage method | |
| CN103853714A (en) | Data processing method and device | |
| CN102333108A (en) | Distributed cache synchronization system and method | |
| CN115640110B (en) | Distributed cloud computing system scheduling method and device | |
| CN105024839A (en) | Method and device for software version upgrading | |
| CN101521606B (en) | Log information processing method, log server and communication 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 | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20110720 |