CN103491155B - Cloud computing method and system for achieving mobile computing and obtaining mobile data - Google Patents
Cloud computing method and system for achieving mobile computing and obtaining mobile data Download PDFInfo
- Publication number
- CN103491155B CN103491155B CN201310425325.XA CN201310425325A CN103491155B CN 103491155 B CN103491155 B CN 103491155B CN 201310425325 A CN201310425325 A CN 201310425325A CN 103491155 B CN103491155 B CN 103491155B
- Authority
- CN
- China
- Prior art keywords
- distributed
- file system
- storage
- scheduling
- server
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Computer And Data Communications (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Multi Processors (AREA)
Abstract
The invention relates to a cloud computing method and system for achieving mobile computing and obtaining mobile data. The cloud computing system comprises an application program server, a distributed type dispatching management unit, a processing pool, a query pool, a storage pool and a cloud management platform, the application program server generates service requests and sends the service requests to the processing pool or the query pool, the distributed type dispatching management unit is used for generating dispatching strategies and load balancing strategies, the processing pool is used for carrying out distributed type processing according to the dispatching strategies and storing processing results in the storage pool in a distributed mode, the query pool is used for seeking related data in the storage pool according to the load balancing strategies in a load balancing mode, the storage pool is used for achieving the virtualization of file storage, file sharing and transparent storage, and the cloud management platform is used for monitoring and managing the processing pool, the distributed type dispatching management unit, the storage pool and the query pool. Based on a distributed type dispatching system and a distributed type file system, service application can be clouded, data moving and mobile computing are achieved in a configurable mode, and therefore high-efficiency parallel computing is achieved.
Description
Technical field
The present invention relates to computer realm, more particularly to a kind of cloud computing side for realizing mobile computing and obtaining mobile data
Method and system.
Background technology
In many key areas such as internet, telecommunications, with the continuous development of business event, the I/T environment of enterprise also exists
Constantly developed, oneself substantial amounts of server system has been gradually formed on IT machine rooms or wide area network, run this enterprise various
The Distributed Application software of business.The features such as there is wide area distribution, isomery, dynamic due to each server so that Distributed Application
Each calculating task Effec-tive Function becomes an extremely complex and challenging problem in system.
For example to the detailed menu manager field of telecom charging, there is problems with existing processing system：
1. system architecture close coupling：At present detailed menu manager and core business close coupling, affect core business, and IO and resource are competing
Fierceness is striven, is stored and is not separated with calculating, cut machine difficult.
2. hardware cost is high：Be entirely at present high-end resource being stored and be calculated, without classification, high cost.
3. dynamic expansion is unable to：Need stop system to migrate detailed monofile during Host Extensions, and change zone configuration.Can not stop
Machine, according to great Qu host task is distributed, and the disposal ability of main frame is unbalanced.
4. system upgrade extension is difficult：The resource of system great Qu system distribution at present, data are not separated with application, system extension,
Upgrading is dumb；Without unified management and control interface.
The content of the invention
The technical problem to be solved is for the deficiencies in the prior art, there is provided one kind is realized mobile computing and obtained
Take the cloud computing method and system of mobile data.
The technical scheme that the present invention solves above-mentioned technical problem is as follows：It is a kind of to realize mobile computing and obtain mobile data
Cloud computing method, comprises the steps：
Step 1：Apps server produces service request；
Step 2：Judge the type of service request, ask then to be sent to processing pond if processing, execution step 3, if
Inquiry pond, execution step 5 are then sent to for inquiry request；
Step 3：The processing pond searches corresponding scheduling according to the service request for receiving in distributed scheduling administrative unit
Strategy, and distributed treatment is carried out to service request according to scheduling strategy；
Step 4：Result is sent to into storage pool, result is carried out distributed storage by storage pool, and will storage
As a result apps server is returned to by processing pond, is terminated；
Step 5：Searched in distributed scheduling administrative unit according to the service request for receiving corresponding negative in the inquiry pond
Balance policy is carried, and related data is searched in storage pool by load balancing mode according to load balancing；
Step 6：Query Result is returned to into apps server, is terminated.
On the basis of above-mentioned technical proposal, the present invention can also do following improvement.
Further, processing pond described in step 3 carries out the concrete reality of distributed treatment according to scheduling strategy to service request
It is now as follows：
Step 3.1：The distributed scheduling client receives the process request that apps server sends；
Step 3.2：Corresponding scheduling strategy, and then basis are searched in distributed scheduling administrative unit according to request is processed
Scheduling strategy is sent to corresponding distributed scheduling proxy server by request is processed；
Step 3.3：The distributed scheduling proxy server is processed the process request for receiving, and by result
It is sent to corresponding distributed type file system client side.
Further, result is carried out being implemented as follows for distributed storage by storage pool in step 4：
Step 4.1：The distributed type file system client side takes according to result in metadata of distributed type file system
The store path of the result is searched on business device；
Step 4.2：Result is stored in corresponding distributed file system storage server according to store path.
Further, step 5 is implemented as follows：
Step 5.1：The load balancing client receives the inquiry request that apps server sends；
Step 5.2：Corresponding load balancing is searched in distributed scheduling administrative unit according to inquiry request, and then
Inquiry request is sent to by corresponding responsible equilibrium proxy server according to load balancing；
Step 5.3：Inquiry request is sent to corresponding distributed file system visitor by the load balancing proxy server
Family end；
Step 5.4：The distributed type file system client side is first in distributed file system according to the content of inquiry request
Corresponding data storage path is searched on data server；
Step 5.5：Distributed type file system client side is deposited according to data storage path in corresponding distributed file system
Desired data is inquired about on storage server；
Step 5.6：The data of lookup are returned to into corresponding apps server by inquiring about pond.
Further, above-mentioned technical proposal also includes that cloud management platform is single to processing pond, inquiry pond, distributed scheduling management
Unit, storage pool are monitored management.
Further, result storage is arrived distributed file system by step 4.2 distributed type file system client side
While storage server, connect in local cache unit storage certain period of time that also can be in distributed type file system client side
The result of receipts.
Another technical scheme that the present invention solves above-mentioned technical problem is as follows：One kind realizes mobile computing and obtains mobile number
According to cloud computing system, including several apps servers, distributed scheduling administrative unit, processing pond, inquiry pond, storage
Pond and cloud management platform；
The apps server, it is used to produce service request, is sent out service request according to the type of service request
Give processing pond or inquiry pond；
The distributed scheduling administrative unit, it is used to formulate scheduling strategy and load balancing for processing pond and inquiry
Transfer in pond；
The processing pond, it is used to that corresponding tune to be searched in distributed scheduling administrative unit according to the service request for receiving
Degree strategy, and carries out distributed treatment according to scheduling strategy, is stored in storage pool by result is distributed；
The inquiry pond, it is used to that corresponding bearing to be searched in distributed scheduling administrative unit according to the service request for receiving
Balance policy is carried, and related data is searched in storage pool by load balancing mode according to load balancing；
The storage pool, it is used to realize virtualization, file-sharing and the transparent storage that file is stored；
The cloud management platform, it is used to supervise distributed scheduling administrative unit, processing pond, inquiry pond and storage pool
Keyholed back plate is managed.
On the basis of above-mentioned technical proposal, the present invention can also do following improvement.
Further, the processing pond includes distributed scheduling client and several distributed scheduling proxy servers；
The distributed scheduling client, it is used to receive the service request of apps server transmission, and according to industry
Corresponding scheduling strategy is searched in business request in distributed scheduling administrative unit, and then is sent service request according to scheduling strategy
To corresponding distributed scheduling proxy server；
The distributed scheduling proxy server, it is used for the service request to receiving and processes, and by result
It is sent to corresponding distributed type file system client side.
Further, the inquiry pond includes load balancing client and several load balancing proxy servers；
The load balancing client, it is used to receive the inquiry request of apps server transmission, and distributed
Corresponding load balancing is searched in management and running unit, and then is sent to inquiry request accordingly according to load balancing
Load balancing proxy server；
The load balancing proxy server, it is used to for inquiry request to be sent to corresponding distributed file system client
End.
Further, the storage pool includes several distributed type file system client sides, several distributed file systems
Storage server, distributed file system metadata service device and distributed file system management tool；
The distributed type file system client side, it is used for the result sent according to distributed scheduling proxy server
The store path of the result is searched on distributed file system metadata service device, and then will be processed according to store path
As a result store in corresponding distributed file system storage server,
Or, according to load balancing proxy server send inquiry request in distributed file system metadata service device
The middle store path for searching desired data, and search according in store path to corresponding distributed file system storage server
Desired data；
The distributed file system metadata service device, it is used to storing metadata, and to storage pool in it is distributed
File system client and distributed file system storage server are monitored management；
The distributed file system storage server, it is used for the number of distributed storage file system client transmission
According to；
The distributed file system management tool, it is used to monitor the function situation of storage pool.
The invention has the beneficial effects as follows：
1. system architecture loose coupling, using Dispatching System and distributed file system processing pond, inquiry pond are realized
And storage pool, each computing pool independently of one another can be with dynamic expansion；
2. hardware cost is low, and each host deployments of the system reduce hardware cost in X86, therefore support not
Under conditions of increasing hardware input, system concurrency disposal ability is improved；
3. dynamic scalability is strong：The task in each computing resource pond of Dispatching System Coordination Treatment so that each to calculate section
The equilibrium of point disposal ability and dynamic retractility, the storage pool is realized by distributed file system, it is ensured that the dynamic of storage pool is stretched
Contracting；
4. online upgrading easily, carries out unifying task scheduling by Dispatching System, is realized by cloud management platform
Online upgrading.
Based on Dispatching System and distributed file system, service application cloud is capable of achieving, by configurable mode,
Mobile data and mobile computing are realized, so as to complete efficient parallel computation.
Description of the drawings
Fig. 1 is a kind of cloud computing method flow chart realized mobile computing and obtain mobile data of the present invention；
Fig. 2 is step 3 flow chart of the present invention；
Fig. 3 is step 4 flow chart of the present invention；
Fig. 4 is step 5 flow chart of the present invention；
Fig. 5 is a kind of cloud computing system structure chart realized mobile computing and obtain mobile data of the present invention；
Fig. 6 is processing pond structure chart of the present invention；
Fig. 7 is inquiry pool structure figure of the present invention；
Fig. 8 is storage pool structure chart of the present invention.
In accompanying drawing, the list of parts representated by each label is as follows：
1st, apps server, 2, distributed scheduling administrative unit, 3, processing pond, 4, inquiry pond, 5, storage pool, 6, cloud
Management platform, 301, distributed scheduling client, 302, distributed scheduling proxy server, 401, load balancing client,
402nd, load balancing proxy server, 501, distributed type file system client side, 502, distributed file system metadata service
Device, 503, distributed file system storage server, 504, distributed file system management tool.
Specific embodiment
The principle and feature of the present invention are described below in conjunction with accompanying drawing, example is served only for explaining the present invention, and
It is non-for limiting the scope of the present invention.
As shown in figure 1, a kind of cloud computing method realized mobile computing and obtain mobile data, comprises the steps：
Step 1：Apps server produces service request；
Step 2：Judge the type of service request, ask then to be sent to processing pond if processing, execution step 3, if
Inquiry pond, execution step 5 are then sent to for inquiry request；
Step 3：The processing pond searches corresponding scheduling according to the service request for receiving in distributed scheduling administrative unit
Strategy, and distributed treatment is carried out to service request according to scheduling strategy；
Step 4：Result is sent to into storage pool, result is carried out distributed storage by storage pool, and will storage
As a result apps server is returned to by processing pond, is terminated；
Step 5：Searched in distributed scheduling administrative unit according to the service request for receiving corresponding negative in the inquiry pond
Balance policy is carried, and related data is searched in storage pool by load balancing mode according to load balancing；
Step 6：Query Result is returned to into apps server, is terminated.
Wherein, above-mentioned technical proposal also include cloud management platform to processing pond, inquiry pond, distributed scheduling administrative unit,
Storage pool is monitored management.
As shown in Fig. 2 processing pond described in step 3 carries out the concrete of distributed treatment according to scheduling strategy to service request
Realize as follows：
Step 3.1：The distributed scheduling client receives the process request that apps server sends；
Step 3.2：Corresponding scheduling strategy, and then basis are searched in distributed scheduling administrative unit according to request is processed
Scheduling strategy is sent to corresponding distributed scheduling proxy server by request is processed；
Step 3.3：The distributed scheduling proxy server is processed the process request for receiving, and by result
It is sent to corresponding distributed type file system client side.
As shown in figure 3, result is carried out being implemented as follows for distributed storage by storage pool in step 4：
Step 4.1：The distributed type file system client side takes according to result in metadata of distributed type file system
The store path of the result is searched on business device；
Step 4.2：Result is stored in corresponding distributed file system storage server according to store path.
Wherein, step 4.2 distributed type file system client side deposits result storage to distributed file system
While storage server, receive in local cache unit storage certain period of time that also can be in distributed type file system client side
Result.
As shown in figure 4, step 5 is implemented as follows：
Step 5.1：The load balancing client receives the inquiry request that apps server sends；
Step 5.2：Corresponding load balancing is searched in distributed scheduling administrative unit according to inquiry request, and then
Inquiry request is sent to by corresponding responsible equilibrium proxy server according to load balancing；
Step 5.3：Inquiry request is sent to corresponding distributed file system visitor by the load balancing proxy server
Family end；
Step 5.4：The distributed type file system client side is first in distributed file system according to the content of inquiry request
Corresponding data storage path is searched on data server；
Step 5.5：Distributed type file system client side is deposited according to data storage path in corresponding distributed file system
Desired data is inquired about on storage server；
Step 5.6：The data of lookup are returned to into corresponding apps server by inquiring about pond.
As shown in figure 5, a kind of cloud computing system realized mobile computing and obtain mobile data, including several apply journey
Sequence server 1, distributed scheduling administrative unit 2, processing pond 3, inquiry pond 4, storage pool 5 and cloud management platform 6；
The apps server 1, it is used to produce service request, is sent out service request according to the type of service request
Give processing pond 2 or inquiry pond 3；
The distributed scheduling administrative unit 2, it is used to formulate scheduling strategy and load balancing for processing pond and looks into
Ask pond to transfer；
The processing pond 3, it is used for according to the service request for receiving, and 2 lookups are corresponding in distributed scheduling administrative unit
Scheduling strategy, and carry out distributed treatment according to scheduling strategy, is stored in storage pool by result is distributed；
The inquiry pond 4, it is used to be searched in distributed scheduling administrative unit accordingly according to the service request for receiving
Load balancing, and related data is searched in storage pool by load balancing mode according to load balancing；
The storage pool 5, it is used to realize virtualization, file-sharing and the transparent storage that file is stored；
The cloud management platform 6, it is used to carry out distributed scheduling administrative unit, processing pond, inquiry pond and storage pool
Monitoring management.
Wherein, cloud management platform, there is provided visualization interface, monitors and manages to whole mobile computing mobile data.Can
To monitor each server resource, processing pond, inquiry pond and each module of storage pool whether normal operation can be monitored, if it find that certain
Module running is abnormal, can automatically process partial fault with alarm prompt attendant, recovers the normal operation of each module
State.
Processing pond, distributed scheduling client is appointed according to the process that scheduling strategy dispatches each distributed scheduling proxy server
Business, realizes the coordination and parallel running of each distributed scheduling proxy server, meets parallel computation, the dynamic expansion of data management
Require.
Inquiry pond, by Load Balancing Task Scheduling, realizes inquiring about looking into parallel for each load balancing proxy server in pond
Inquiry is processed, and realizes the parallel computation of query processing.
Storage pool, based on distributed file system storage pool is built, and realizes file storage virtualization, file-sharing and transparent
Storage.
As shown in fig. 6, the processing pond 3 includes distributed scheduling client 301 and several distributed schedulings agency's clothes
Business device 302；
The distributed scheduling client 301, it is used to receive the service request of the transmission of apps server 1, and root
Corresponding scheduling strategy is searched in distributed scheduling administrative unit 2 according to service request, and then please by business according to scheduling strategy
Ask and be sent to corresponding distributed scheduling proxy server 302；
The distributed scheduling proxy server 302, it is used for the service request to receiving and processes, and will process knot
Fruit is sent to corresponding distributed type file system client side 401.
As shown in fig. 7, the inquiry pond 4 includes load balancing client 401 and several load balancing proxy servers
402；
The load balancing client 401, it is used to receive the inquiry request of the transmission of apps server 1, and is dividing
Corresponding load balancing is searched in cloth management and running unit 2, and then is sent inquiry request according to load balancing
To corresponding load balancing proxy server 402；
The load balancing proxy server 402, it is used to for inquiry request to be sent to corresponding distributed file system
Client 401.
As shown in figure 8, the storage pool 5 includes several distributed type file system client sides 501, several distributed texts
Part system storage server 503, distributed file system metadata service device 502 and distributed file system management tool 504；
The distributed type file system client side 501, it is used for the place sent according to distributed scheduling proxy server 302
Reason result searches the store path of the result on distributed file system metadata service device 502, and then according to storage
Path stores result in corresponding distributed file system storage server 503,
Or, taken in metadata of distributed type file system according to the inquiry request that load balancing proxy server 402 sends
The store path of desired data is searched in business device 502, and according to store path to corresponding distributed file system storage service
Desired data is searched in device 503；
The distributed file system metadata service device 502, it is used to store metadata, and to the distribution in storage pool
Formula file system client 501 and distributed file system storage server 503 are monitored management；
The distributed file system storage server 503, it is used for distributed storage file system client 501 and sends
Data；
The distributed file system management tool 504, it is used to monitor the function situation of storage pool 5.
(the generally same day connects in local cache unit storage certain period of time in distributed type file system client side 501
The result of receipts) receive result.
Wherein the internal memory of distributed type file system client side 501 contains the configuration strategy of mobile data and mobile computing, is counting
It is investigated that when asking, mobile data and mobile computing strategy are configured with configurable mode, for the inquiry of same day data, using local slow
Deposit mode, i.e. storage pool data according to configuration information by same day data buffer storage inquiry pond be located server in, inquiry pond
Inquire about in the buffer cell of the server at place, parallel computation is carried out with mobile computing pattern, for the inquiry of data in the past, match somebody with somebody
Put mobile data strategy, in the way of multiple index, data movement carried out by distributed file system, carry out data query and
Parallel computation.
Wherein, mobile computing:Be people can at any time and any place, can incessantly from network obtain institute's phase
The technology general name of the service of prestige.Mobile data:Be people can at any time and any place, can obtain from network incessantly
Take desired data.For example：Telephone expenses are inquired about using mobile phone terminal, novel is read, this are all to obtain service to listen to online music,
Belong to mobile computing.And during this, the content of telephone expenses, the lteral data of novel, the audio file of music is all to move number
According to.
The distributed scheduling client of processing pond is distributed to task according to the scheduling strategy obtained from management and running unit
Each distributed scheduling agent side, each distributed scheduling agent side calculating task creates, processes and exit, and realizes calculating task
Movement, complete mobile computing, the calculating task larger for processing data amount, calculate before processing, using file system number
Data buffer storage by distributed scheduling, is moved into calculating local according to caching technology.
The distributed scheduling client acquisition tasks of processing pond, for the task of mobile computing class, data storage generates rope
Draw, generate data file and journal file.For the task of mobile data class, data storage generates data file, storage pool profit
With caching mechanism, data cached information.
The load balancing proxy server in inquiry pond, for the task of mobile computing class, reads index, reads data text
Part.For the task of mobile data class, reading storage pool is data cached, if data cached without hit, from storage pool
Data are read in distributed file system, while updating storage pool cache information.
Dispatching System is based on distributed scheduling model, and each calculating task of management control is created, runs, suspends, stopped
Only, processing pond, inquiry pond, data query and store tasks are exited, while carrying out fail safe, Data Migration.
The Dispatching System is mainly by distributed scheduling administrative unit, distributed scheduling client, distributed tune
Degree proxy server, load balancing client and load balancing proxy server composition, distributed scheduling administrative unit is responsible for tune
The formulation of degree strategy, data division, task scheduling, load balancing, fault-tolerant processing etc.；Distributed scheduling client (load balancing
Client) with upstream interface applications, upstream is applied by distributed scheduling client (load balancing client) with distributed tune
Degree system interaction；Distributed scheduling proxy server (load balancing proxy server) is responsible for (negative from distributed scheduling client
Carry balanced client) task is received, and according to the scheduling strategy (load balancing) searched in distributed scheduling administrative unit
Local task scheduling is completed, data processing and calculating is carried out, and is responsible for data transfer communications.
By the distributed file system of storage pool, distributed file system storage server data are operated, it is complete
Calculate into mobile data.For the less calculating task of processing data amount, using data retrieval technology, data are accessed, by dividing
Cloth file system carries out data movement, completes data mobile computing.
The distributed file system of storage pool is a for mass data storage, for the cluster that cloud storage service is designed
Storage system, there is provided the memory capacity of PB levels, realizes file system sas storage virtualization, and making to apply can pellucidly access actual point
The file physical data of cloth, it is ensured that the data access performance of mass memory, with high fault tolerance, high availability, extensibility with
And manageability and on-demand service, not there are problems that Single Point of Faliure.By distributed type file system client side, distributed file system
Meta data server, distributed file system data storage server and distributed file system management tool composition.In order to carry
High-performance improves local file caching mechanism with configurable mode.
The client of distributed file system, in each the distributed scheduling proxy server in processing pond and inquiry pond
It is required to that the client of distributed file system is installed on each load balancing proxy server, by distributed file system visitor
Family end carries out efficient distributed file system and accesses storage.
Distributed file system metadata service device, manages the metadata of distributed file system (including file directory tree
Tissue, attribute maintenance, file operation log recording, mandate access etc.), the NameSpace of whole storage system is managed, externally carry
For single system image, and it is responsible for the management and monitoring of whole storage cluster.Distributed file system metadata service device is coordinated
Activity between commander's application server and storage server, there is provided to name space, access control information, file and memory block
The management of the information such as map information, block position
Distributed file system storage server, stores the real data of user, is whole distributed file system storage
Resource provider.When application server carries out data access, storage server cluster provides actual data I/O service.Data
What pressure can be in admirable proportion is distributed between storage server cluster.
The foregoing is only presently preferred embodiments of the present invention, not to limit the present invention, all spirit in the present invention and
Within principle, any modification, equivalent substitution and improvements made etc. should be included within the scope of the present invention.
Claims (5)
1. it is a kind of realize mobile computing and obtain mobile data cloud computing method, it is characterised in that comprise the steps：
Step 1：Apps server produces service request；
Step 2：Judge the type of service request, if request is processed processing pond is then sent to, execution step 3, if looking into
Ask request and be then sent to inquiry pond, execution step 5；
Step 3：The processing pond searches corresponding scheduling strategy according to the service request for receiving in distributed scheduling administrative unit,
And distributed treatment is carried out to service request according to scheduling strategy；
Step 3 is implemented as follows：
Step 3.1：The distributed scheduling client receives the process request that apps server sends；
Step 3.2：Corresponding scheduling strategy is searched in distributed scheduling administrative unit according to request is processed, and then according to scheduling
Strategy is sent to corresponding distributed scheduling proxy server by request is processed；
Step 3.3：The distributed scheduling proxy server is processed the process request for receiving, and result is sent
To corresponding distributed type file system client side；
Step 4：Result is sent to into storage pool, result is carried out distributed storage by storage pool, and will store result
Apps server is returned to by processing pond, is terminated；
Step 5：It is equal that corresponding load is searched in the inquiry pond according to the service request for receiving in distributed scheduling administrative unit
Weighing apparatus strategy, and related data is searched in storage pool by load balancing mode according to load balancing；
Step 5 is implemented as follows：
Step 5.1：The load balancing client receives the inquiry request that apps server sends；
Step 5.2：Corresponding load balancing, and then basis are searched in distributed scheduling administrative unit according to inquiry request
Inquiry request is sent to corresponding responsible equilibrium proxy server by load balancing；
Step 5.3：Inquiry request is sent to corresponding distributed type file system client side by the load balancing proxy server；
Step 5.4：The distributed type file system client side is according to the content of inquiry request in metadata of distributed type file system
Corresponding data storage path is searched on server；
Step 5.5：Distributed type file system client side takes according to data storage path in the storage of corresponding distributed file system
Desired data is inquired about on business device；
Step 5.6：The data of lookup are returned to into corresponding apps server by inquiring about pond；
Step 6：Query Result is returned to into apps server, is terminated.
2. according to claim 1 it is a kind of realize mobile computing and obtain mobile data cloud computing method, it is characterised in that
Result is carried out being implemented as follows for distributed storage by storage pool in step 4：
Step 4.1：The distributed type file system client side is according to result in distributed file system metadata service device
The upper store path for searching the result；
Step 4.2：Result is stored in corresponding distributed file system storage server according to store path.
3. according to claim 1 it is a kind of realize mobile computing and obtain mobile data cloud computing method, it is characterised in that
Also include that cloud management platform is monitored management to processing pond, inquiry pond, distributed scheduling administrative unit, storage pool.
4. according to claim 2 it is a kind of realize mobile computing and obtain mobile data cloud computing method, it is characterised in that
Step 4.2 distributed type file system client side is by result storage to the same of distributed file system storage server
When, the result received in local cache unit storage certain period of time that also can be in distributed type file system client side.
5. it is a kind of realize mobile computing and obtain mobile data cloud computing system, it is characterised in that apply journey including several
Sequence server, distributed scheduling administrative unit, processing pond, inquiry pond, storage pool and cloud management platform；
The apps server, it is used to produce service request, is sent to service request according to the type of service request
Processing pond or inquiry pond；
The distributed scheduling administrative unit, it is used to formulate scheduling strategy and load balancing for processing pond and inquiry pond tune
Take；
The processing pond, it is used to that corresponding scheduling plan to be searched in distributed scheduling administrative unit according to the service request for receiving
Slightly, and distributed treatment is carried out according to scheduling strategy, is stored in storage pool by result is distributed；
The processing pond includes distributed scheduling client and several distributed scheduling proxy servers；
The distributed scheduling client, it is used to receive the service request of apps server transmission, and please according to business
Ask and corresponding scheduling strategy is searched in distributed scheduling administrative unit, and then service request is sent to by phase according to scheduling strategy
The distributed scheduling proxy server answered；
The distributed scheduling proxy server, it is used for the service request to receiving and processes, and result is sent
To corresponding distributed type file system client side；
The inquiry pond, it is used to that corresponding load to be searched in distributed scheduling administrative unit according to the service request for receiving
Weighing apparatus strategy, and related data is searched in storage pool by load balancing mode according to load balancing；The inquiry Chi Bao
Include load balancing client and several load balancing proxy servers；
The load balancing client, it is used to receive the inquiry request of apps server transmission, and in distributed scheduling
Corresponding load balancing is searched in administrative unit, and then is sent to inquiry request according to load balancing corresponding negative
Carry balanced proxy server；
The load balancing proxy server, it is used to for inquiry request to be sent to corresponding distributed type file system client side；
The storage pool, it is used to realize virtualization, file-sharing and the transparent storage that file is stored；If the storage pool includes
Dry distributed type file system client side, several distributed file system storage server, metadata of distributed type file system
Server and distributed file system management tool；
The distributed type file system client side, its result for being used to be sent according to distributed scheduling proxy server is being divided
Search the store path of the result on cloth file system metadata server, and then according to store path by result
Store in corresponding distributed file system storage server,
Or, looked in distributed file system metadata service device according to the inquiry request that load balancing proxy server sends
The store path of desired data is looked for, and according to needed for searching in store path to corresponding distributed file system storage server
Data；
The distributed file system metadata service device, it is used to store metadata, and to the distributed document in storage pool
System client and distributed file system storage server are monitored management；
The distributed file system storage server, it is used for the data of distributed storage file system client transmission；
The distributed file system management tool, it is used to monitor the function situation of storage pool；
The cloud management platform, it is used to be monitored pipe to distributed scheduling administrative unit, processing pond, inquiry pond and storage pool
Reason.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310425325.XA CN103491155B (en) | 2013-09-17 | 2013-09-17 | Cloud computing method and system for achieving mobile computing and obtaining mobile data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310425325.XA CN103491155B (en) | 2013-09-17 | 2013-09-17 | Cloud computing method and system for achieving mobile computing and obtaining mobile data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103491155A CN103491155A (en) | 2014-01-01 |
| CN103491155B true CN103491155B (en) | 2017-04-12 |
Family
ID=49831108
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310425325.XA Active CN103491155B (en) | 2013-09-17 | 2013-09-17 | Cloud computing method and system for achieving mobile computing and obtaining mobile data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103491155B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105635196B (en) * | 2014-10-27 | 2019-08-09 | 中国电信股份有限公司 | A kind of method, system and application server obtaining file data |
| CN107423301B (en) * | 2016-05-24 | 2021-02-23 | 华为技术有限公司 | Data processing method, related equipment and storage system |
| CN106060594B (en) * | 2016-05-25 | 2019-04-23 | 武汉斗鱼网络科技有限公司 | A kind of white list generation system and method based on player heartbeat |
| CN106612218A (en) * | 2017-01-01 | 2017-05-03 | 国云科技股份有限公司 | Regional feature extraction method of data packet of virtual access entry |
| CN107453951A (en) * | 2017-08-15 | 2017-12-08 | 郑州云海信息技术有限公司 | A kind of storage pool monitoring method and device |
| CN107491551B (en) * | 2017-08-30 | 2021-04-06 | 郑州云海信息技术有限公司 | Method and device for realizing client access of distributed file system |
| CN107958033A (en) * | 2017-11-20 | 2018-04-24 | 郑州云海信息技术有限公司 | Lookup method, device, distributed file system and the storage medium of metadata |
| CN109901552B (en) * | 2019-03-12 | 2021-09-03 | 杭州哲达科技股份有限公司 | Vibration multidimensional perception and cloud computing method for machine-pump health management and control |
| CN111897792A (en) * | 2020-08-11 | 2020-11-06 | 北京无线电测量研究所 | Distributed file access method, system, medium and device |
| CN112541038A (en) * | 2020-12-01 | 2021-03-23 | 杭州海康威视数字技术股份有限公司 | Time series data management method, system, computing device and storage medium |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100403315C (en) * | 2006-09-25 | 2008-07-16 | 华为技术有限公司 | System and method for database access for implementing load sharing |
| US20080137830A1 (en) * | 2006-12-12 | 2008-06-12 | Bhogal Kulvir S | Dispatching A Message Request To A Service Provider In A Messaging Environment |
| CN101969391B (en) * | 2010-10-27 | 2012-08-01 | 北京邮电大学 | Cloud platform supporting fusion network service and operating method thereof |
| CN103106249B (en) * | 2013-01-08 | 2016-04-20 | 华中科技大学 | A kind of parallel data processing system based on Cassandra |
-
2013
- 2013-09-17 CN CN201310425325.XA patent/CN103491155B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN103491155A (en) | 2014-01-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103491155B (en) | Cloud computing method and system for achieving mobile computing and obtaining mobile data | |
| CN108885582B (en) | Multi-tenant memory services for memory pool architecture | |
| JP7221975B2 (en) | Transaction processing in a multi-master distributed data management system | |
| US10936659B2 (en) | Parallel graph events processing | |
| CN105144121B (en) | Cache content addressable data block is for Storage Virtualization | |
| US20100332629A1 (en) | Secure custom application cloud computing architecture | |
| US11327970B1 (en) | Context dependent execution time prediction for redirecting queries | |
| Sowmya et al. | Layers of cloud–IaaS, PaaS and SaaS: a survey | |
| CN110737643A (en) | big data analysis, processing and management center station based on catering information management system | |
| US20120166492A1 (en) | Database transfers using constraint free data | |
| US11076020B2 (en) | Dynamically transitioning the file system role of compute nodes for provisioning a storlet | |
| CN103595799A (en) | Method for achieving distributed shared data bank | |
| US11003493B2 (en) | Application and storage based scheduling | |
| Chandra et al. | A study on cloud database | |
| US20220050728A1 (en) | Dynamic data driven orchestration of workloads | |
| CN115811546A (en) | System and method for realizing network cooperative distributed processing for scientific and technological service | |
| US11714573B1 (en) | Storage optimization in a distributed object store | |
| US11625273B1 (en) | Changing throughput capacity to sustain throughput for accessing individual items in a database | |
| CN106570151A (en) | Data collection processing method and system for mass files | |
| US11477273B2 (en) | Feedback control based load balancing for containerized workloads | |
| Chen et al. | The research about video surveillance platform based on cloud computing | |
| US9229659B2 (en) | Identifying and accessing reference data in an in-memory data grid | |
| US11656957B1 (en) | Managing nodes of a DBMS | |
| Saifeng | VFS_CS: a light-weight and extensible virtual file system middleware for cloud storage system | |
| Xie | Improving Performance of Hadoop Clusters |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |