CN101354720B - Distributed memory database data system and sharing method thereof - Google Patents
Distributed memory database data system and sharing method thereof Download PDFInfo
- Publication number
- CN101354720B CN101354720B CN2008101418546A CN200810141854A CN101354720B CN 101354720 B CN101354720 B CN 101354720B CN 2008101418546 A CN2008101418546 A CN 2008101418546A CN 200810141854 A CN200810141854 A CN 200810141854A CN 101354720 B CN101354720 B CN 101354720B
- Authority
- CN
- China
- Prior art keywords
- data
- submodule
- module
- service
- edge treated
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a distributed memory database data system and a sharing method thereof. The method comprises the following steps: the data receiving sub module receives data sent by an edge processer, the index and the address data of the edge processor are added and stored to a sub module for a usable data pool; a servable data management sub module sends a request to a data request process sub module; the data request process sub module distributes the data from the sub module for the usable data pool according to the request and sends the data and the index of the data to the usable service data management sub module as well as adds the date and the index of the data into the servable data; and when the application process has access to the database, the data is shared. The sharing method adopts a center processor to manage all the data indexes of the edge processer and the method for distributing the data to the edge processer, thereby guaranteeing the rapid searching, acquiring and data modification of the application process without acquiring data from the network, avoiding the expenditure of the additional network, effectively preventing the access conflict and realizing the data sharing at high speed.
Description
Technical field
The present invention relates to a kind of data of database system and shared method thereof, relate in particular to a kind of distributed data system and shared method thereof.
Background technology
In the prior art, the high capacity telecommunication apparatus adopts embedded multi-computer system usually, wherein, on a plurality of processors, use memory database, data management and service are provided, and the efficient of its memory database requires high, and most data can only realize visit on this processor, and can not use the access to netwoks mechanism of common distributed data base, remote procedure call (RPC) for example.
An in store class is referred to as the data of " resource " in the memory database, and application process needs when realizing telecommunication service to database application resource, and obtaining just can provide telecommunications service after the resource, and service is given back resource to database after finishing again.Each processor is managed a part of resource, non-overlapping copies, resource itself also exists " unavailable ", " available ", " in the use " three kinds of states, have only " available " resource just can be employed the process application, after the available resources of this locality all have been applied, processor can't directly provide telecommunication service, and application process need redirect to service on other processors that have resource usually.This process not only causes the decline of efficient, has also increased the complexity of handling application process.
Therefore, prior art awaits improving and development.
Summary of the invention
The object of the present invention is to provide a kind of distributed memory database data system to share method, manage the index that all are distributed in the data on the edge treated machine by adopting central processing unir, central processing unir is according to the method for edge treated machine module to the data information request distribution portion data of its application, when guaranteeing on the application process visit edge treated machine database, can search, obtain, revise data fast, can effectively prevent access conflict, realize data sharing efficiently.
Technical scheme of the present invention is as follows:
A kind of distributed memory database data system is characterized in that, this system comprises: at least one edge treated machine module; One central processing unir module; Described at least one edge treated machine module is connected with described central processing unir module respectively;
Described central processing unir module comprises: a data available pond module, a Data Receiving submodule, a data request processing submodule;
Described data available pond module is connected with Data Receiving submodule, data request processing submodule respectively, be used to store the index of the data on the described edge treated machine module and the edge treated machine address date that data are belonged to, and be changed to the data of preserving available;
Described Data Receiving submodule, be used to receive the data on the described edge treated machine module that edge treated machine module sends, and obtain the data available index and give the edge treated machine of being asked module as response feedback from the module of described data available pond, and the data that described data available pond module is acquired are changed in the use;
Described data request processing submodule is used to receive the data application information of each edge treated machine module and feedback request information.
Described system, wherein, described edge treated machine module comprises: memory database submodule, application process submodule.
Described system, wherein, described memory database submodule comprises: local data management submodule; But service data management submodule; But described local data management submodule, described service data management submodule communicate with the central processing unir module respectively and are connected;
Described local data management submodule is used for the data on the edge treated machine module are sent in the central processing unir module;
But described service data management submodule is used for sending request of data information to the central processing unir module, and receives the data of described central processing unir module assignment.
Described system, wherein, described central processing unir module also comprises data retrieval management submodule, described data retrieval management submodule, but be used for recycling service data information.
Described system, wherein, described memory database submodule also comprises: but data queue in application process data processing submodule, the formation of data source data formation service data, the service.
A kind of distributed memory database data sharing method that is applied to the described data system of claim 1, the edge treated machine module of this system comprises: memory database submodule, application process submodule, described memory database submodule comprises again: but local data management submodule service data is managed submodule, application process data processing submodule, but described local data management submodule, described service data management submodule are connected with the central processing unir module respectively, and it comprises the steps:
A, described Data Receiving submodule receive the data on its edge treated machine module that described edge treated machine module sends, and the edge treated machine address date that this data directory and data are belonged to is added, and are saved in described data available pond module;
But B, described service data management submodule send request of data information to described data request processing submodule, and the outstanding message feedback;
C, described data request processing submodule from described data available pond module assignment partial data, and with these data and this data directory, but send to described service data management submodule according to described request information.
But D, described service data management submodule, but the data of described distribution are added service data, when on the application process submodule visit edge treated machine during database, carry out data sharing and handle.
Described method wherein, also comprises before the described steps A:
A01, described local data management submodule send to described Data Receiving submodule with the data on the described edge treated machine module.
Described method, wherein, described step D is further comprising the steps of:
D1, described application process submodule are to described application process data processing submodule request for data, but described memory database obtains data from the service data formation and offers described application process submodule, but and these data are transferred to data queue the service from the service data formation;
After D2, described application process submodule are finished and are used for the process service, discharge data to described application process data processing submodule, but described memory database is transferred to described data data queue from service in the service data formation, for follow-up application process application.
Described method, wherein, when the central processing unir module of described data system also comprises data retrieval management submodule, further comprising the steps of after the described step D2:
But but D3, described data retrieval management submodule data volume in the data queue in described service data management submodule inquiry service data formation and service, but and send the data collection order according to the predetermined policy decision to described service data management submodule;
But after D4, described service data management submodule are received the data collection order, send to described central processing unir module, but and with of the service data formation deletion of these data from this locality but obtain data directory from the service data formation.
Described method, wherein, after the described step D4, further comprising the steps of:
After D5, described application process data processing submodule provide the data application, remaining data amount in the data queue in the inspection service; When data are about to use up, but trigger described service data management submodule, initiate request of data once more.
The data system of a kind of distributed memory database provided by the present invention and shared method thereof, owing to adopted central processing unir to manage all edge treated machine data index, central processing unir is according to the data information request of edge treated machine module to its application, the method of distribution portion data, guaranteed that all application processes look in local express quick checking, obtain, revise data, and need not obtain data from network, avoided the expense of complementary network, prevented simultaneously the generation of access conflict effectively, and can realize data sharing efficiently, and the logical topology structure of software module and the management of distributed data base have been simplified.
Description of drawings
Fig. 1 is a data system structural representation of the present invention;
Fig. 2 is a central processing unir inside modules illustrative view of functional configuration of the present invention;
Fig. 3 is an edge treated machine inside modules illustrative view of functional configuration of the present invention;
Fig. 4 is that data system of the present invention is shared method flow diagram.
Embodiment
Below in conjunction with accompanying drawing, each preferred embodiment of the present invention is made a more detailed description.
The data system of distributed memory database of the present invention comprises: central processing unir and many edge of table processor, described many edge of table processor is connected with described central processing unir respectively, as shown in Figure 1, many edge of table processor is represented with B1, B2...Bn respectively, wherein edge treated machine B1 and central processing unir are on same physical treatment machine, its internal logic structure is consistent with edge treated machine B2...Bn, and edge treated machine B1 provides edge treated machine function on central processing unir.This data system can comprise or not comprise edge treated machine B1, when described data system does not comprise this edge treated machine, central processing unir is not then managed this edge treated machine resource data, does not promptly manage the local resource data, does not provide the request for data service to application process service submodule simultaneously.
Central processing unir A of the present invention includes: data available pond modules A 101; Data Receiving submodule A1; Data request processing submodule A2; Data retrieval management submodule A3, as shown in Figure 2.
Described data available pond modules A 101, it does not preserve complete data available, only preserves the edge treated machine ID address date that data available index and data are belonged to;
Described Data Receiving submodule A1, be used to receive the ready data message that the edge treated machine reports, and the ready data directory that each edge treated machine is reported and the ID address date of ownership edge treated machine add described data available pond modules A 101, and these data are changed to " available ";
Described data request processing submodule A2, be used for receiving the data application of each edge treated machine, and from described data available pond modules A 101, obtain a collection of data available index according to certain strategy, give the edge treated of being asked machine as response feedback, and this partial data in the modules A 101 of described data available pond is changed to " in the use ", then these data can only be used by the edge treated of being applied for machine, and can not reallocate to other edge treated machines;
Described data retrieval management submodule A3, be used for regularly inquiring about each edge treated machine and " can serve " data queue's ratio data, but when the service data large percentage, then but the edge treated machine initiates to reclaim command operation to this service data, redundant data is reclaimed, and be " available " state its " in the use " state-transition in data available pond modules A 101.
Edge treated machine B1 of the present invention includes again: application process submodule b4, internal storage data library module b100, wherein, described internal storage data library module b100 comprises again: but local data management submodule b1 service data is managed submodule b2, application process data processing submodule b3, data source formation b101, " can serve " b102 of data queue, " in the service " b103 of data queue, as shown in Figure 3.
Described local data management submodule b1, be in charge of the data source formation b101 that belongs to this edge treated machine, after local data among the described data source formation b101 was ready, described local data management submodule b1 was responsible for the information ready for data to described Data Receiving submodule A1;
But described service data management submodule b2, be responsible for providing to application process the data of service to described data request processing submodule A2 request, the data that described data request processing submodule A2 returns may be that the ready data directory among the described local data management submodule b1 also may be the data available index on other edge treated machines, if the former, but then described service data management submodule b2, this partial data is added number " can serve " among the b102 of data queue, but if the described service data management of latter submodule b2, but need be to the service data management submodule b2 of the edge treated machine of attribution data, obtain complete data, and with among this data adding data source formation b101, and then add among " can serve " b102 of data queue;
But described service data management submodule b2, also be responsible for receiving the query requests of described data retrieval management submodule A3, return the data bulk that " can serve " described in this edge treated machine among the b102 of data queue and described " in the service " b103 of data queue;
But described service data management submodule b2, also be responsible for receiving the data recovery request of described data retrieval management submodule A3, and the data of some are taken out from " can serve " b102 of data queue return to described data retrieval management submodule A3, preferentially the data of non-processor are discharged to described data retrieval management submodule A3 in the processing procedure;
Described application process data processing submodule b3, be responsible for receiving the application of application process data, services, described application process data processing submodule b3, at first from described " can serve " b102 of data queue, obtain data available, preferentially choose the data that belong to this processor when obtaining, after obtaining successfully data are deleted from described " can serve " b102 of data queue, add described " in the service " b103 of data queue, and in " in the service " b103 of data queue, write down the data that offered application process submodule b4;
Described application process data processing submodule b3 also is responsible for the data of reception application process and gives back, and the data of giving back are deleted from described " the service " b103 of data queue, joins among described " can serve " b102 of data queue;
Described application process data processing submodule b3, also be responsible for checking remaining data amount among described " can serve " b102 of data queue, according to certain strategy when the remaining data quantity not sufficient, but notify described service data management submodule b2, but, initiate operation once more to described data request processing submodule A2 application supplemental data by described service data management submodule b2;
Described application process submodule b4 is used for finishing the application process service to described application process data processing submodule b3 application service data.
The shared method of no colliding data system between a plurality of processor distributed memory database of the present invention, as shown in Figure 4, its data system includes: central processing unir module, a plurality of edge treated machine module, described a plurality of edge treated machine modules are connected with described central processing unir module respectively.Described central processing unir module comprises again: data available pond module, Data Receiving submodule, data request processing submodule; Described data available pond module, described Data Receiving submodule, described data request processing submodule are connected with the edge treated machine respectively again.Edge treated machine module comprises: memory database submodule, application process submodule, and described memory database submodule comprises again: but local data management submodule service data is managed submodule, application process data processing submodule, " data source " data queue, " can serve " data queue, " in the service " data queue.But described local data management submodule, described service data management submodule communicate with the central processing unir submodule respectively and are connected.Described memory database submodule includes again: its method step is as follows:
A, local data management submodule b1 check in data source formation b101 whether local data is ready, if ready these data are sent to Data Receiving submodule A1, described Data Receiving submodule A1 joins data available pond A101 with this data directory and processor ID address
But B service data management submodule b2, A2 sends requesting data information to the data request processing submodule, after described data request processing submodule A2 receives this solicited message, according to certain strategy, from described data available pond A101, obtain a part of data directory, send to the edge treated machine, and in described data available pond A101, this partial data index is designated " in the use ", avoid duplicate allocation.
But C, described service data management submodule b2, after receiving the data available that described data request processing submodule A2 returns, if local, then directly its adding " can be served " b102 of data queue, otherwise the ownership edge treated machine address and the data directory of the data available of sending according to described data request processing submodule A2, use request of data to the edge treated machine transmission that this partial data belonged to, after this ownership edge treated machine is received, then complete data are sent to the requesting party.The requesting party adds data source formation b101 with data, and with its adding described " can serve " b102 of data queue.
D, application process submodule b4, b3 sends the data application to application process data processing submodule, described application process data processing submodule b3 is to obtain data from local " can serve " b102 of data queue, offer described application process submodule b4, and with these data from described " can serve " b102 of data queue, transfer among " in the service " b103 of data queue.
After E, described application process submodule b4 finish the application process service, discharge data to the b3 of application process data processing submodule described in the described memory database in this locality, described memory database moves to these data among described " can serve " b102 of data queue from described " the service " b103 of data queue, for subsequent applications process service request.
But the data bulk among the regularly described service data management of F, data retrieval management submodule A3 submodule b2 inquiry " can serve " b102 of data queue and " in the service " b103 of data queue, but and according to certain strategy decision to described service data management submodule b2, send the data collection order.
But G, described service data management submodule b2, after receiving the data collection order, according to certain strategy from described " can serve " b102 of data queue, obtain data directory and send to central processing unir, and these data are deleted from the b102 of data queue of this locality described " can serve ", if these data are not local datas, but the data of borrowing on other processor are then deleted it from described data source formation b101.
H, described application process data processing submodule b3, after the data application is provided, check " in the service " b103 of data queue remaining data amount, when data are about to use up, but trigger described service data management submodule b2, initiate request of data once more to central processing unir by step B.
The data system of a kind of distributed memory database provided by the present invention and shared method thereof, owing to adopted central processing unir to manage all edge treated machine data index, central processing unir is according to the data information request of edge treated machine module to its application, the method of distribution portion data, guaranteed that all application processes look for, obtain, revise the data that not necessarily belong to this processor in local express quick checking, need not obtain data from network, avoid the expense of complementary network.Prevented simultaneously the generation of access conflict effectively, guaranteed that promptly same data can not visited by a plurality of processors, and can realize data sharing efficiently, and simplified the logical topology structure of software module and the management of distributed data base.
Should be understood that the description of above-mentioned specific embodiment is comparatively detailed, can not therefore be interpreted as the restriction to scope of patent protection of the present invention, scope of patent protection of the present invention should be as the criterion with claims.
Claims (10)
1. a distributed memory database data system is characterized in that, this system comprises: at least one edge treated machine module; One central processing unir module; Described at least one edge treated machine module is connected with described central processing unir module respectively;
Described central processing unir module comprises: a data available pond module, a Data Receiving submodule, a data request processing submodule;
Described data available pond module is connected with Data Receiving submodule, data request processing submodule respectively, is used to store the index of the data on the described edge treated machine module and the edge treated machine address date that data are belonged to;
Described Data Receiving submodule, be used to receive the data on the described edge treated machine module that edge treated machine module sends, and the edge treated machine address date that this data directory and data are belonged to added and be saved in described data available pond module, and be changed to the data of preserving available;
Described data request processing submodule, be used to receive the data application information of each edge treated machine module, and obtain the data available index and give the edge treated machine of being asked module as response feedback from the module of described data available pond, and the data that described data available pond module is acquired are changed in the use.
2. system according to claim 1 is characterized in that, described edge treated machine module comprises: memory database submodule, application process submodule.
3. system according to claim 2 is characterized in that, described memory database submodule comprises: local data management submodule; But service data management submodule; But described local data management submodule, described service data management submodule communicate with the central processing unir module respectively and are connected;
Described local data management submodule is used for the data on the edge treated machine module are sent in the central processing unir module;
But described service data management submodule is used for sending request of data information to the central processing unir module, and receives the data of described central processing unir module assignment.
4. system according to claim 1 is characterized in that, described central processing unir module also comprises data retrieval management submodule, described data retrieval management submodule, but be used for recycling service data information.
5. system according to claim 2 is characterized in that, described memory database submodule also comprises: but data queue in application process data processing submodule, the formation of data source data formation service data, the service.
6. data sharing method that is applied to the described data system of claim 1, it is characterized in that, the edge treated machine module of this system comprises: memory database submodule, application process submodule, described memory database submodule comprises again: but local data management submodule service data is managed submodule, application process data processing submodule, but described local data management submodule, described service data management submodule are connected with the central processing unir module respectively, and described method comprises the steps:
A, described Data Receiving submodule receive the data on its edge treated machine module that described edge treated machine module sends, and the edge treated machine address date that this data directory and data are belonged to is added, and are saved in described data available pond module;
But B, described service data management submodule send request of data information to described data request processing submodule, and the outstanding message feedback;
C, described data request processing submodule from described data available pond module assignment partial data, and with these data and this data directory, but send to described service data management submodule according to described request information;
But D, described service data management submodule, but the data of described distribution are added service data, when on the application process submodule visit edge treated machine during database, carry out data sharing and handle.
7. method according to claim 6 is characterized in that, also comprises before the described steps A:
A01, described local data management submodule send to described Data Receiving submodule with the data on the described edge treated machine module.
8. method according to claim 6 is characterized in that, described step D is further comprising the steps of:
D1, described application process submodule are to described application process data processing submodule request for data, but described memory database obtains data from the service data formation and offers described application process submodule, but and these data are transferred to data queue the service from the service data formation;
After D2, described application process submodule are finished and are used for the process service, discharge data to described application process data processing submodule, but described memory database is transferred to described data data queue from service in the service data formation, for follow-up application process application.
9. method according to claim 8 is characterized in that, and is when the central processing unir module of described data system also comprises data retrieval management submodule, further comprising the steps of after the described step D2:
But but D3, described data retrieval management submodule data volume in the data queue in described service data management submodule inquiry service data formation and service, but and send the data collection order according to the predetermined policy decision to described service data management submodule;
But after D4, described service data management submodule are received the data collection order, send to described central processing unir module, but and with of the service data formation deletion of these data from this locality but obtain data directory from the service data formation.
10. method according to claim 9 is characterized in that, and is after the described step D4, further comprising the steps of:
After D5, described application process data processing submodule provide the data application, remaining data amount in the data queue in the inspection service; When data are about to use up, but trigger described service data management submodule, initiate request of data once more.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008101418546A CN101354720B (en) | 2008-09-04 | 2008-09-04 | Distributed memory database data system and sharing method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008101418546A CN101354720B (en) | 2008-09-04 | 2008-09-04 | Distributed memory database data system and sharing method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101354720A CN101354720A (en) | 2009-01-28 |
| CN101354720B true CN101354720B (en) | 2011-08-10 |
Family
ID=40307528
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2008101418546A Expired - Fee Related CN101354720B (en) | 2008-09-04 | 2008-09-04 | Distributed memory database data system and sharing method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101354720B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102088389B (en) * | 2009-12-02 | 2015-01-28 | 中兴通讯股份有限公司 | Distributed content access scheduling device and content reading method |
| CN101872356B (en) * | 2010-05-31 | 2013-08-07 | 中兴通讯股份有限公司 | Method and system for improving processing performance of memory database |
| CN102385677B (en) * | 2010-09-01 | 2015-04-29 | 北京启明星辰信息技术股份有限公司 | Unified threat management system and data processing method thereof |
| CN103577457B (en) * | 2012-07-31 | 2017-09-08 | 国际商业机器公司 | For the method and system manipulated to multi-tenant database |
| CN110716690B (en) * | 2018-07-12 | 2023-02-28 | 阿里巴巴集团控股有限公司 | Data recovery method and system |
| US20210149918A1 (en) * | 2019-11-15 | 2021-05-20 | International Business Machines Corporation | Intelligent data pool |
| CN112417229A (en) * | 2020-11-12 | 2021-02-26 | 成都渊数科技有限责任公司 | Method for acquiring corresponding equipment model based on mac address |
| CN113360534B (en) * | 2021-08-11 | 2022-05-06 | 北京快立方科技有限公司 | Memory recovery method for multi-copy data of memory database |
-
2008
- 2008-09-04 CN CN2008101418546A patent/CN101354720B/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN101354720A (en) | 2009-01-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101354720B (en) | Distributed memory database data system and sharing method thereof | |
| US8583756B2 (en) | Dynamic configuration and self-tuning of inter-nodal communication resources in a database management system | |
| CN106294190B (en) | Storage space management method and device | |
| US20150026213A1 (en) | System and method for providing a shareable global cache for use with a database environment | |
| CN108900626B (en) | Data storage method, device and system in cloud environment | |
| JP2000503785A (en) | Method and apparatus for updating or changing a network directory | |
| CN103744719A (en) | Lock management method, lock management system, lock management system configuration method and lock management system configuration device | |
| WO2007041899A1 (en) | A system and method of managing the dynamic adaptive distributed resource | |
| CN107864211A (en) | Cluster resource dispatching method and system | |
| CN102508872A (en) | Data processing method and system of online processing system based on memory | |
| CN116150160B (en) | Adjustment method and device for database cluster processing nodes and storage medium | |
| US20090100436A1 (en) | Partitioning system including a generic partitioning manager for partitioning resources | |
| CN110162395B (en) | Memory allocation method and device | |
| CN102243655A (en) | Method and device for managing database connections | |
| CN101963978A (en) | Distributed database management method, device and system | |
| CN101065732A (en) | Computer system and information processing method | |
| CN110543357B (en) | Method, related device and system for managing application program object | |
| CN101754156A (en) | Office data management method, device and charging system | |
| CN112346848A (en) | Method, device and terminal for managing memory pool | |
| CN112001800B (en) | Method and device for processing business in block chain system | |
| CN1455558A (en) | Communication system, recording medium of connecting management servor device and recording programme | |
| CN109347991B (en) | File distribution method, device, equipment and medium | |
| US20100199281A1 (en) | Managing the Processing of Processing Requests in a Data Processing System Comprising a Plurality of Processing Environments | |
| CN110019113B (en) | Database service processing method and database server | |
| CN112616143A (en) | Method and device for distributing communication number, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20110810 Termination date: 20150904 |
|
| EXPY | Termination of patent right or utility model |