CN106095534B - A kind of calculating task processing method and system - Google Patents
A kind of calculating task processing method and system Download PDFInfo
- Publication number
- CN106095534B CN106095534B CN201610399492.5A CN201610399492A CN106095534B CN 106095534 B CN106095534 B CN 106095534B CN 201610399492 A CN201610399492 A CN 201610399492A CN 106095534 B CN106095534 B CN 106095534B
- Authority
- CN
- China
- Prior art keywords
- task
- module
- operational module
- operational
- calculating
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/466—Transaction processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
Abstract
The invention discloses a kind of calculating task processing method and system, which comprises operational module obtains calculating task to be processed, and operational module is deployed in the storage equipment in cloud storage system；Operational module passes through the computing resource being multiplexed on a storage device, completes the processing to calculating task.Using scheme of the present invention, the consumption etc. to resource and cost can reduce.
Description
[technical field]
The present invention relates to Computer Applied Technology, in particular to a kind of calculating task processing method and system.
[background technique]
The files such as a large amount of user video, picture are stored on cloud storage system, with the continuous development evolution of product, are needed
A large amount of computing capability is provided to handle the file of user, as video code conversion, file decompression, pictorial information extract
Deng.
In the prior art, it generallys use and builds the mode of physical computing cluster to provide above-mentioned computing capability, turned with video
For code, need to develop the dedicated frame of a set of video code conversion, it is anticipated that video code conversion scale, Quantity arrives several hundred
Thousands of physical equipment clusters is used for operation, and this needs to expend huge resource and cost.
[summary of the invention]
The present invention provides a kind of calculating task processing method and system, the consumption etc. to resource and cost can reduce.
Specific technical solution is as follows:
A kind of calculating task processing method, comprising:
Operational module obtains calculating task to be processed, and the operational module is deployed in the storage equipment in cloud storage system
On；
The operational module passes through the computing resource being multiplexed on a storage device, completes the place to the calculating task
Reason.
According to one preferred embodiment of the present invention, the calculating task includes: offline task, online task and appoints in real time
Business.
According to one preferred embodiment of the present invention, the operational module obtains calculating task to be processed and includes:
The operational module pulls offline task；
The operational module receives the online task or real-time task sent.
According to one preferred embodiment of the present invention, the operational module pulls offline task and includes:
The operational module sends scheduling request before pulling offline task every time, to scheduler module, and described in reception
The dispatch command that scheduler module returns is corresponded to from the offline business mark carried in the dispatch command is pulled in message-oriented middleware
Offline task, the offline task in the message-oriented middleware is to be sent among the message after the front-end module receives
In part.
According to one preferred embodiment of the present invention, the operational module receives the online task sent or real-time task packet
It includes:
The operational module receives the front-end module and is directed to the online task received or real-time task to scheduler module
After issuing scheduling request, sent according to the operational module mark carried in the dispatch command of scheduler module return online
Task or real-time task.
According to one preferred embodiment of the present invention, it is described completion include: to the processing of the calculating task
When the calculating task is offline task or online task, it is corresponding that the operational module tune plays the calculating task
Business process, complete processing to the calculating task；
When the calculating task is real-time task, the operational module uses own thread, completes to appoint described calculate
The processing of business.
According to one preferred embodiment of the present invention, this method further comprises:
Before handling the calculating task, the calculating task is sent to proxy module by the operational module, so as to
The proxy module saves the calculating task, and if receiving the operational module hair before being more than scheduled duration
The task completion notice sent, then delete the calculating task, otherwise, turns to carry out the calculating task by other operational modules
Processing.
A kind of calculating task processing system, comprising: operational module；
The operational module is deployed in the storage equipment in cloud storage system,
The operational module for obtaining calculating task to be processed, and passes through the calculating being multiplexed on a storage device
Resource completes the processing to the calculating task.
According to one preferred embodiment of the present invention, the calculating task includes: offline task, online task and appoints in real time
Business.
According to one preferred embodiment of the present invention, the operational module pulls offline task；
The operational module receives the online task or real-time task sent.
According to one preferred embodiment of the present invention, the system further comprises: scheduler module and front-end module；
For the operational module before pulling offline task every time, Xiang Suoshu scheduler module sends scheduling request, receives institute
State scheduler module return dispatch command, from pulled in message-oriented middleware carried in the dispatch command offline business mark pair
The offline task answered；
The scheduler module, for after receiving from the scheduling request of the operational module, Xiang Suoshu operational module
Return to the dispatch command for carrying the offline business mark in need pulled；
The front-end module, for sending the offline task received in the message-oriented middleware.
According to one preferred embodiment of the present invention, the front-end module is further used for, when receive online task or in real time
When task, Xiang Suoshu scheduler module sends scheduling request, and the online task or real-time task are sent to the scheduler module
The operational module carried in the dispatch command of return identifies corresponding operational module；
The scheduler module is further used for, after receiving from the scheduling request of the front-end module, to before described
End module returns to the dispatch command for carrying the operational module mark of responsible processing task.
According to one preferred embodiment of the present invention, when the calculating task is offline task or online task, the work
Module tune plays the corresponding business process of the calculating task, completes the processing to the calculating task；
When the calculating task is real-time task, the operational module uses own thread, completes to appoint described calculate
The processing of business.
According to one preferred embodiment of the present invention, the system further comprises: proxy module；
The operational module is further used for, and before handling the calculating task, the calculating task is sent to institute
Proxy module is stated, and notifies the proxy module after the completion of task processing；
The proxy module, for being saved to the calculating task, and if being received before being more than scheduled duration
The task completion notice that the operational module is sent, then delete the calculating task, otherwise, turns by other operational modules to institute
Calculating task is stated to be handled.
Scheme of the present invention is used by described above can be seen that, map out the work module on a storage device, work
Module can complete the processing to calculating task, thus with intimate zero cost by the computing resource left unused in multiplexing storage equipment
Mode got a large amount of computing capability, compared with the prior art, no longer need to specially build physical computing cluster, Jin Erji
The earth reduces the consumption to resource and cost, and improves the resource utilization of storage equipment.
[Detailed description of the invention]
Fig. 1 is the flow chart of calculating task processing method embodiment of the present invention.
Fig. 2 is the composed structure schematic diagram of calculating task processing system embodiment of the present invention.
[specific embodiment]
Cloud storage system possesses the memory device set group of substantial amounts, due to being mainly used for depositing to store user file
Store up file, therefore common features of these storage equipment relatively high, but computing resource such as central processing unit that is that disk occupies
There are a large amount of free time in (CPU, Central Processing Unit) etc., to cause the huge waste of computing resource.
It for this purpose, proposing a kind of calculating task processing system in the present invention, or is mammoth system, wherein can include: work
Make module (mammoth worker), scheduler module (mammoth observer) and front-end module (mommoth front) etc.,
By the mutual cooperation between each module, the various processing such as task access, task schedule and task computation are completed.
Wherein, front-end module is mainly responsible for task access, and scheduler module is mainly responsible for task schedule, and operational module is mainly born
Responsibility business handles (task computation).
Operational module is deployed in the storage equipment in cloud storage system, using the computing resource left unused in storage equipment
Calculating task is handled, without specially building physical computing cluster again, and then significantly reduce to resource and
The consumption of cost, and improve the resource utilization of storage equipment.
In order to be clearer and more clear technical solution of the present invention, hereinafter, referring to the drawings and the embodiments, to institute of the present invention
The scheme of stating is described in further detail.
Embodiment one
Fig. 1 is the flow chart of calculating task processing method embodiment of the present invention, as shown in Figure 1, including realizing in detail below
Mode.
In 11, operational module obtains calculating task to be processed, and operational module is deployed in the storage in cloud storage system
In equipment.
In order to which using the computing resource left unused in storage equipment, can map out the work module on a storage device, each storage
An operational module need to be only disposed in equipment.
The calculating task can include: offline task, online task and real-time task.
According to the difference of task type, the mode that operational module obtains calculating task also can be different, as follows:
1) operational module pulls offline task
Operational module sends scheduling request before pulling offline task every time, to scheduler module, and receives scheduler module
The dispatch command of return identifies corresponding offline task from the offline business carried in dispatch command is pulled in message-oriented middleware,
Offline task in message-oriented middleware is to be sent in message-oriented middleware after front-end module receives；
2) operational module receives the online task or real-time task sent
Operational module receiving front-end module is directed to the online task received or real-time task and issues scheduling to scheduler module
After request, the online task sent is identified according to the operational module carried in the dispatch command of scheduler module return or is appointed in real time
Business.
Front-end module is responsible for task access, by interacting with user, obtains all types of tasks of user's publication.Scheduling
Module is the scheduling component in mammoth system, is responsible for the task schedule of whole system.
Front-end module is receiving one after line task or real-time task, can send scheduling request to scheduler module, ask
Which operational module ask that scheduler module is responsible for handling the task is, scheduler module receives the scheduling request from front-end module
Afterwards, one can be selected from numerous operational modules to be responsible for handling the task, and the operational module selected mark is carried and is being adjusted
Front-end module is returned in degree instruction, and then the task is sent to selected operational module by front-end module and is handled.
Front-end module can send the offline task received in message-oriented middleware.Each operational module is pulling every time
Before offline task, can all send scheduling request to scheduler module, inquiry scheduler module this which offline business needed to pull
Task, after scheduler module receives the scheduling request from operational module, can be taken to operational module return according to scheduling rule
Dispatch command with offline business mark notifies the operational module to go to pull appointing for which offline business in this way
Business.
Scheduler module needs to meet simultaneously the various characteristics of offline scheduling and on-line scheduling as scheduling component.It is online to adjust
The characteristic of degree can include: available operational module is provided, guarantees the load balancing etc. between each operational module.It dispatches offline
Characteristic can include: control task quantity and speed that each operational module pulls each offline business, protect the safety of operational module
And execution efficiency, prevent excess pressure etc..
In addition, each operational module can handle the task of all offline business, it, can be right according to the scheduling of scheduler module
Multiple offline business carry out the distribution of computing resources and time, i.e., in the limited situation of computing resource each offline business it
Between deploy computing resource, and need to guarantee calculate service will not to operational module on a storage device original storage service make
At influence.
How scheduler module is specifically scheduled with no restriction, can be determined according to actual needs.
In 12, operational module passes through the computing resource being multiplexed on a storage device, completes the processing to calculating task.
In the present embodiment, calculating task can be transmitted in form of a message, and messaging flow is transparent to business, this can
To greatly simplify the complexity and learning cost of business development, quick business development and access may be implemented whereby.
For different types of calculating task, different transmitting and processing mode can be used, as follows:
1) offline task
Message is transmitted by message-oriented middleware；
The corresponding business process of offline task has been adjusted to complete task processing；
2) online task
Message is transmitted by remote procedure call protocol (RPC, Remote Procedure Call Protocol)；
The corresponding business process of online task has been adjusted to complete task processing；
3) real-time task
Message is transmitted by RPC agreement；
Operational module is completed task using own thread and is handled.
Usually, offline task is lower to requirement of real-time, but more demanding to throughput, and online task is to real-time
More demanding but lower to throughput requirement, real-time task is higher to requirement of real-time, and synchronous return of more requirements handles knot
Fruit.
User can choose suitable calculating mode, i.e. off-line mode, on-line mode according to the calculation features etc. of own service
Or real-time mode, corresponding task are then respectively offline task, online task or real-time task.
It in practical applications, can also be further in mammoth system in addition to operational module, scheduler module and front-end module
Including proxy module (mammoth agent).
The calculating task can be sent to proxy module, so as to generation before one calculating task of every processing by operational module
Reason module saves the calculating task, and if complete more than the task that operational module is sent is received before scheduled duration
At notice, then the calculating task is deleted, otherwise, turns to handle the calculating task by other operational modules.
Proxy module can start timing after receiving the calculating task, if reach scheduled duration such as 5 minutes it
Before, the task completion notice that operational module is sent is received, then can delete the calculating task, if since operational module occurs
Failure and other reasons do not receive task completion notice before reaching scheduled duration, then can turn by other operational modules to the meter
Calculation task is handled, and such as can be pulled the calculating from proxy module under the scheduling of scheduler module by other operational modules and be appointed
It is engaged in and completes to handle.
By the above-mentioned means, task may make not lose, ensure as far as possible that each task is correctly processed.It is described
The specific value of scheduled duration can be determined according to actual needs.
It is the introduction about embodiment of the present invention method above, is further situated between to present system embodiment below
It continues.
Embodiment two
Fig. 2 is the composed structure schematic diagram of calculating task processing system embodiment of the present invention, as shown in Figure 2, comprising: work
In addition module 21 can also further comprise: scheduler module 22, front-end module 23 and proxy module 24 etc., by between each module
Mutual cooperation, complete the various processing such as task access, task schedule and task computation.
Usually, the number of operational module 21, scheduler module 22, front-end module 23 and proxy module 24 is multiple,
To simplify attached drawing, one is only indicated respectively in Fig. 2.
In addition, the storage that operational module 21, scheduler module 22 and proxy module 24 can be deployed in cloud storage system is set
It is standby upper.That is, can dispose operational module 21, scheduler module 22 above for each storage equipment or act on behalf of mould
One in block 24.
Usually far fewer than the number of operational module 21, each scheduler module 22 can respectively correspond more the number of scheduler module 22
A operational module 21, i.e., for providing dispatch service for corresponding multiple operational modules 21.
Operational module 21 obtains calculating task to be processed, and passes through the computing resource being multiplexed on a storage device, complete
The processing of pairs of calculating task.
The calculating task can include: offline task, online task and real-time task.
According to the difference of task type, the mode that operational module 21 obtains calculating task also can be different, as follows:
1) operational module 21 pulls offline task
Operational module 21 sends scheduling request before pulling offline task every time, to scheduler module 22, receives scheduling mould
The dispatch command that block 22 returns identifies corresponding offline from the offline business carried in dispatch command is pulled in message-oriented middleware
It is engaged in, the offline task in message-oriented middleware is to be sent in message-oriented middleware after front-end module 23 receives；
2) operational module 21 receives the online task or real-time task sent
21 receiving front-end module 23 of operational module is directed to the online task received or real-time task and sends out to scheduler module 22
Out after scheduling request, the online task sent is identified according to the operational module carried in the dispatch command of the return of scheduler module 22
Or real-time task.
Front-end module 23 is responsible for task access, by interacting with user, obtains all types of tasks of user's publication.It adjusts
Degree module 22 is the scheduling component in system, is responsible for the task schedule of whole system.
Front-end module 23 is receiving one after line task or real-time task, can send scheduling to scheduler module 22 and ask
It asks, the responsible processing task of inquiry scheduler module 22 is which operational module 21, and scheduler module 22 is received from front end mould
After the scheduling request of block 23, one can be selected from numerous operational modules 21 and is responsible for handling the task, and the work that will be selected
Module id carrying returns to front-end module 23 in dispatch command, and then by front-end module 23 is sent to the task selected
Operational module 21 handled.
Front-end module 23 can send the offline task received in message-oriented middleware.Each operational module 21 is each
Before pulling offline task, all can to scheduler module 22 send scheduling request, inquiry scheduler module 22 this need which is pulled
The task of offline business can be according to scheduling rule, Xiang Gong after scheduler module 22 receives the scheduling request from operational module 21
Make module 21 and return to the dispatch command for carrying offline business mark, notifies that operational module 21 goes which is pulled in this way
The task of a offline business.
Scheduler module 22 needs to meet simultaneously the various characteristics of offline scheduling and on-line scheduling as scheduling component.Online
The characteristic of scheduling can include: available operational module 21 is provided, guarantees the load balancing etc. between each operational module 21.Offline
The characteristic of scheduling can include: control task quantity and speed that each operational module 21 pulls each offline business, protect Working mould
The safety and execution efficiency of block 21, prevent excess pressure etc..
In addition, the task that each operational module 21 can handle all offline business can according to the scheduling of scheduler module 22
Multiple offline business are carried out with the distribution of computing resource and time, i.e., in each offline industry in the limited situation of computing resource
Computing resource is deployed between business, and needing to guarantee to calculate service will not be to 21 original storage on a storage device of operational module
Service impacts.
How scheduler module 22 is specifically scheduled with no restriction, can be determined according to actual needs.
In the present embodiment, calculating task can be transmitted in form of a message, moreover, appointing for different types of calculate
Business, can be used different transmitting and processing mode, as follows:
1) offline task
Message is transmitted by message-oriented middleware；
Operational module 21 has adjusted the corresponding business process of offline task to complete task processing；
2) online task
Message is transmitted by RPC agreement；
Operational module 21 has adjusted the corresponding business process of online task to complete task processing；
3) real-time task
Message is transmitted by RPC agreement；
Operational module 21 is completed task using own thread and is handled.
Usually, offline task is lower to requirement of real-time, but more demanding to throughput, and online task is to real-time
More demanding but lower to throughput requirement, real-time task is higher to requirement of real-time, and synchronous return of more requirements handles knot
Fruit.
User can choose suitable calculating mode, i.e. off-line mode, on-line mode according to the calculation features etc. of own service
Or real-time mode, corresponding task are then respectively offline task, online task or real-time task.
In addition, the calculating task can also be sent to before one calculating task of every processing and act on behalf of mould by operational module 21
Block 24, and the notification agent module 24 after the completion of task processing.
Proxy module 24 saves the calculating task, and if receiving operational module 21 before being more than scheduled duration
The task completion notice sent, then delete the calculating task, otherwise, turns to carry out the calculating task by other operational modules 21
Processing.
By the above-mentioned means, task may make not lose, ensure as far as possible that each task is correctly processed.It is described
The specific value of scheduled duration can be determined according to actual needs.
In short, map out the work module on a storage device using scheme of the present invention, operational module can be deposited by multiplexing
The computing resource left unused in equipment is stored up, the processing to calculating task is completed, to got in a manner of intimate zero cost big
The computing capability of amount no longer needs to specially build physical computing cluster, and then significantly reduce to resource compared with the prior art
With the consumption of cost, and improve storage equipment resource utilization.
In several embodiments provided by the present invention, it should be understood that disclosed system and method can pass through it
Its mode is realized.For example, system embodiment described above is only schematical, for example, the division of the unit, only
Only a kind of logical function partition, there may be another division manner in actual implementation.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list
Member both can take the form of hardware realization, can also realize in the form of hardware adds SFU software functional unit.
The above-mentioned integrated unit being realized in the form of SFU software functional unit can store and computer-readable deposit at one
In storage media.Above-mentioned SFU software functional unit is stored in a storage medium, including some instructions are used so that a computer
It is each that equipment (can be personal computer, server or the network equipment etc.) or processor (processor) execute the present invention
The part steps of embodiment the method.And storage medium above-mentioned include: USB flash disk, mobile hard disk, read-only memory (ROM,
Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic or disk etc. it is each
Kind can store the medium of program code.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Within mind and principle, any modification, equivalent substitution, improvement and etc. done be should be included within the scope of the present invention.
Claims (10)
1. a kind of calculating task processing method characterized by comprising
Operational module obtains calculating task to be processed, and the operational module is deployed in the storage equipment in cloud storage system
's；
The operational module passes through the computing resource being multiplexed on a storage device, completes the processing to the calculating task, packet
Include: when the calculating task is offline task or online task, the operational module tune plays the corresponding industry of the calculating task
Business process completes the processing to the calculating task；When the calculating task is real-time task, the operational module is used certainly
Body thread completes the processing to the calculating task.
2. the method according to claim 1, wherein
The operational module obtains calculating task to be processed
The operational module pulls offline task；
The operational module receives the online task or real-time task sent.
3. according to the method described in claim 2, it is characterized in that,
The operational module pulls offline task
The operational module sends scheduling request before pulling offline task every time, to scheduler module, and receives the scheduling
Module return dispatch command, from pulled in message-oriented middleware carried in the dispatch command offline business mark it is corresponding from
Line task, the offline task in the message-oriented middleware are to be sent in the message-oriented middleware after front-end module receives.
4. according to the method described in claim 2, it is characterized in that,
The operational module receives the online task sent or real-time task includes:
The operational module receiving front-end module is directed to the online task received or real-time task and issues scheduling to scheduler module
After request, the online task or reality sent is identified according to the operational module carried in the dispatch command of scheduler module return
When task.
5. the method according to claim 1, wherein
This method further comprises:
Before handling the calculating task, the calculating task is sent to proxy module by the operational module, so as to described
Proxy module saves the calculating task, and if receiving the operational module before being more than scheduled duration and sending
Task completion notice, then delete the calculating task, otherwise, turn the calculating task is carried out by other operational modules
Reason.
6. a kind of calculating task processing system characterized by comprising operational module；
The operational module is deployed in the storage equipment in cloud storage system,
The operational module for obtaining calculating task to be processed, and passes through the computing resource being multiplexed on a storage device,
Complete the processing to the calculating task；
Wherein, when the calculating task is offline task or online task, the operational module tune plays the calculating task pair
The business process answered completes the processing to the calculating task；
When the calculating task is real-time task, the operational module uses own thread, completes to the calculating task
Processing.
7. system according to claim 6, which is characterized in that
The operational module pulls offline task；
The operational module receives the online task or real-time task sent.
8. system according to claim 7, which is characterized in that
The system further comprises: scheduler module and front-end module；
For the operational module before pulling offline task every time, Xiang Suoshu scheduler module sends scheduling request, receives the tune
Spend module return dispatch command, from pulled in message-oriented middleware carried in the dispatch command offline business mark it is corresponding
Offline task；
The scheduler module, for after receiving from the scheduling request of the operational module, Xiang Suoshu operational module to be returned
Carry the dispatch command of the offline business mark in need pulled；
The front-end module, for sending the offline task received in the message-oriented middleware.
9. system according to claim 8, which is characterized in that
The front-end module is further used for, and when receiving online task or real-time task, Xiang Suoshu scheduler module, which is sent, to be adjusted
The online task or real-time task, are sent to the Working mould carried in the dispatch command that the scheduler module returns by degree request
The corresponding operational module of block identification；
The scheduler module is further used for, after receiving from the scheduling request of the front-end module, the front end Xiang Suoshu mould
Block returns to the dispatch command for carrying the operational module mark of responsible processing task.
10. system according to claim 6, which is characterized in that
The system further comprises: proxy module；
The operational module is further used for, and before handling the calculating task, the calculating task is sent to the generation
Module is managed, and notifies the proxy module after the completion of task processing；
The proxy module, for being saved to the calculating task, and if being received before being more than scheduled duration described
The task completion notice that operational module is sent, then delete the calculating task, otherwise, turns by other operational modules to the meter
Calculation task is handled.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610399492.5A CN106095534B (en) | 2016-06-07 | 2016-06-07 | A kind of calculating task processing method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610399492.5A CN106095534B (en) | 2016-06-07 | 2016-06-07 | A kind of calculating task processing method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106095534A CN106095534A (en) | 2016-11-09 |
| CN106095534B true CN106095534B (en) | 2019-07-23 |
Family
ID=57228122
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610399492.5A Active CN106095534B (en) | 2016-06-07 | 2016-06-07 | A kind of calculating task processing method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106095534B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111435943B (en) * | 2019-01-14 | 2022-07-19 | 阿里巴巴集团控股有限公司 | Data processing method, device, system and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101441580A (en) * | 2008-12-09 | 2009-05-27 | 华北电网有限公司 | Distributed paralleling calculation platform system and calculation task allocating method thereof |
| CN104102543A (en) * | 2014-06-27 | 2014-10-15 | 北京奇艺世纪科技有限公司 | Load regulation method and load regulation device in cloud computing environment |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103795804A (en) * | 2014-02-24 | 2014-05-14 | 华为技术有限公司 | Storage resource scheduling method and storage calculation system |
| CN104750549A (en) * | 2015-04-13 | 2015-07-01 | 飞狐信息技术（天津）有限公司 | Computational task processing device, method and system |
-
2016
- 2016-06-07 CN CN201610399492.5A patent/CN106095534B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101441580A (en) * | 2008-12-09 | 2009-05-27 | 华北电网有限公司 | Distributed paralleling calculation platform system and calculation task allocating method thereof |
| CN104102543A (en) * | 2014-06-27 | 2014-10-15 | 北京奇艺世纪科技有限公司 | Load regulation method and load regulation device in cloud computing environment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106095534A (en) | 2016-11-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10467725B2 (en) | Managing access to a resource pool of graphics processing units under fine grain control | |
| JP7112919B2 (en) | Smart device task processing method and device | |
| CN101652750B (en) | Data processing device, distributed processing system and data processing method | |
| CN110888721A (en) | Task scheduling method and related device | |
| CN107025139A (en) | A kind of high-performance calculation Scheduling Framework based on cloud computing | |
| CN112632566B (en) | Vulnerability scanning method and device, storage medium and electronic equipment | |
| CN112148455B (en) | Task processing method, device and medium | |
| CN102299959A (en) | Load balance realizing method of database cluster system and device | |
| CN102904961A (en) | Method and system for scheduling cloud computing resources | |
| CN113691611B (en) | Block chain distributed high-concurrency transaction processing method, system, equipment and storage medium | |
| CN111200606A (en) | Deep learning model task processing method, system, server and storage medium | |
| CN113703997A (en) | Bidirectional asynchronous communication middleware system integrating multiple message agents and implementation method | |
| CN108170417B (en) | Method and device for integrating high-performance job scheduling framework in MESOS cluster | |
| CN103067486A (en) | Big-data processing method based on platform-as-a-service (PaaS) platform | |
| CN104052677A (en) | Soft load balancing method and apparatus of single data source | |
| CN106095534B (en) | A kind of calculating task processing method and system | |
| CN108628677A (en) | Distributed task scheduling processing system, method and apparatus | |
| CN112261125B (en) | Centralized unit cloud deployment method, device and system | |
| CN112099930A (en) | Quantum computer cluster distributed queue scheduling method | |
| CN115361382B (en) | Data processing method, device, equipment and storage medium based on data group | |
| CN111190731A (en) | Cluster task scheduling system based on weight | |
| CN116010065A (en) | Distributed task scheduling method, device and equipment | |
| CN105549911B (en) | The data transmission method and device of NVRAM | |
| CN103777593A (en) | Automatic product control and production system and realizing method thereof | |
| CN108694544A (en) | A kind of order dispatch processing method and processing device |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230928 Address after: Room 219, floor 2, building 4, courtyard 1, Shangdi East Road, Haidian District, Beijing 100085 Patentee after: Beijing Duyou Information Technology Co.,Ltd. Address before: 100085 Baidu Building, 10 Shangdi Tenth Street, Haidian District, Beijing Patentee before: BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY Co.,Ltd. |