CN109214132A - A kind of big Throughput Asynchronous task processing system of non-coupled streaming towards LVC emulation - Google Patents
A kind of big Throughput Asynchronous task processing system of non-coupled streaming towards LVC emulation Download PDFInfo
- Publication number
- CN109214132A CN109214132A CN201811280764.5A CN201811280764A CN109214132A CN 109214132 A CN109214132 A CN 109214132A CN 201811280764 A CN201811280764 A CN 201811280764A CN 109214132 A CN109214132 A CN 109214132A
- Authority
- CN
- China
- Prior art keywords
- task
- node
- processing
- queue node
- task queue
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Debugging And Monitoring (AREA)
- Multi Processors (AREA)
Abstract
A kind of big Throughput Asynchronous task processing system of non-coupled streaming towards LVC emulation, including external interface, task queue node, task processing node and data center.External interface is received from extraneous female task and resource data, and extraneous female task and resource data are fed back to task queue node, the task processing result of task queue node feeding back is received and issues outward.Task queue node be responsible for the fractionation of task with merge, and task processing node is assigned the task to, by resource data store to data center；External interface is fed back to after handling the processing result from task processing node.Task handles node and obtains task and resource data respectively from task queue node and data center, handles task, the new task generated in processing result and treatment process is fed back to task queue node.Data center's storage resource data.Present invention reduces the delays of analogue system, improve communication efficiency, realize the emulation of LVC integrated high-efficiency.
Description
Technical field
The present invention relates to a kind of big Throughput Asynchronous task processing systems of non-coupled streaming towards LVC emulation, using distribution
Formula and Stream Processing mechanism by task processing and data loose coupling, and fully consider the load balancing between node, belong to emulation neck
Domain.
Background technique
LVC integrative simulation refer in analogue system while having actual load, simulator, virtual troops' three types it is imitative
Very.Actual load refers to that true people carries out practice using practical equipment.Simulator refers to that true people manipulates analogue system, past
It is simulated training system of the people in circuit toward performance.Virtual troops refer to mathematics simulation system, it is a kind of deduction analysis tool.
Currently, some researchs are expanded to LVC integrative simulation technology both at home and abroad, and in recent years, external military product test
The compbined test verifying direction that Validation Mode is being combined from the mode based on actual loading test to actual situation is developed, in military product
The architecture of compbined test, intelligence, networking, standardization etc., achieve a series of theory and practice achievement, and shape
It include High Level Architecture (HLA), basic object model (BOM), the enabled system of Test And Training at series of standards specification
Structure (TENA), Model Driven Architecture (MDA) etc..LVC integrative simulation is related to a large amount of actual loads, simulator and mathematical modulo
Type, each simulation step length, interaction between model is frequent, computationally intensive and data volume is big, this will lead to the delay of analogue system
Increase, communication efficiency reduces, and the present invention provides a kind of big Throughput Asynchronous task processing method of non-coupled streaming towards LVC emulation
Effectively promote the communication performance of LVC integrated simulation system.
Summary of the invention
Technical problem solved by the present invention is overcoming the deficiencies of the prior art and provide a kind of towards the non-coupled of LVC emulation
The big Throughput Asynchronous task processing system of streaming, reduces the delay of analogue system, improves communication efficiency, realizes LVC one
Change efficient emulation.
The technical solution of the invention is as follows:
A kind of big Throughput Asynchronous task processing system of non-coupled streaming towards LVC emulation, including external interface, task team
Column node, task processing node and data center；
External interface: receiving from extraneous female task and resource data, will be extraneous according to the request of task queue node
Female task and resource data feed back to task queue node, receive the task processing result of task queue node feeding back and to outgoing
Cloth；
Task queue node: store tasks and resource data, be responsible for task fractionation with merge, and assign the task to appoint
Business processing node, by resource data store to data center；It is anti-after handling the processing result from task processing node
It feeds external interface；The task includes extraneous female task and the new task that task processing node generates；
Task handle node: obtain task and resource data respectively from task queue node and data center, to task into
Row processing, feeds back to task queue node for the new task generated in processing result and treatment process；
Data center: storage resource data.
The task queue node include it is multiple, each task queue node judges whether according to current load situation to right
External tapping sends request.
The task and resource data that task queue node will acquire are numbered, and obtain mission number and resource index number,
Then by mission number, resource index number, whether have subtask, whether be subtask, subtask registration storage to task queue
In list, the mission number and resource index number are corresponded.
The subtask is registered as an individual list；
Task queue node when receiving task, according to resource judgment consumed by task complexity, calculating whether
It needs to carry out task fractionation；
When task does not need to split, the subtask is registered as sky table；
When task needs to split, appointing for all subtasks that the task is included is stored in the subtask register list
Whether business number resource index number and is completed.
Task queue node is as follows handled the processing result of task processing node:
When predecessor's business is split, after all subtasks all handle completion, it is responsible for own by task queue node
Subtask processing result merges, and obtains the processing result of predecessor's business；
When predecessor is engaged in not being split, the processing result of corresponding task processing node is the processing result formerly held the post of and be engaged in.
Task queue node by mission number, resource index number and completes the money wanted of required by task when receiving task
Source data stores in data center together.
When task queue node assigns the task to task processing node, while the task is sent to task processing node
Mission number and resource index number.
Task processing node has multiple, and each task, which handles node and submits to task queue node, applies, from task queue
Node obtains task, and obtains the resource data that processing task needs from data center according to resource index number, carries out task
Calculation processing；
When the new task not generated in task computation completion and treatment process, processing result is fed back into task queue section
Point；
When the new task generated in task computation completion and treatment process, by what is generated in processing result, treatment process
The resource data that new task and new task need feeds back to task queue node；
When task computation does not complete, do not feed back.
The task processing node after task queue node acquisition task, is needing to safeguard a task processing column
Table, it includes: mission number which, which is handled in list, resource index number, whether generates new task, wherein mission number and resource
Call number is directly acquired from task queue node.
The task, which is handled while node is submitted and applied to task queue node, to be needed the software and hardware information of its own
It is supplied to task queue node, the software and hardware information includes computer display card, CPU usage, memory usage and handling capacity.
The mode that the task queue node assigns the task to task processing node is as follows: task queue node is periodical
To application calculating task task processing node according to software and hardware information carry out priority ranking, at the high task of priority
It manages node and distributes task, realize the load balancing of task processing node.
The invention has the following advantages over the prior art:
(1) task queue node of the present invention splits task, and is assigned on multiple tasks processing node, realizes
Distributed treatment greatly improves calculating speed, reduces the delay of analogue system, improves communication efficiency, is capable of handling
The calculating task of big flux.
(2) task queue node of the present invention, which does not need task processing node feeding back result, to be that task is handled according to request
Node distributes task, realizes the Stream Processing to task, improves communication efficiency, realizes the emulation of LVC integrated high-efficiency.
(3) when task queue node of the present invention receives task, to task and resource data reference numeral, and individually will money
Source data is stored in data center, realizes the loose coupling of task and resource data, the resource consumption of node is reduced, is improved
The safety of calculating speed and system.
Detailed description of the invention
Fig. 1 is composite structural diagram of the invention；
Fig. 2 is task queue list structure figure of the invention；
Fig. 3 is that task of the invention handles list structure figure.
Specific embodiment
Before describing the specific embodiment of the invention, the technical term used to the present invention is illustrated: female task is
Refer to the task from the received most original of exterior.Subtask refers to complete a task, female task is split into multiple
Small task.
The present invention will be further described in the following with reference to the drawings and specific embodiments.
As shown in Figure 1, a kind of big Throughput Asynchronous task processing system of non-coupled streaming towards LVC emulation of the present invention
System, comprising: external interface, task queue node, task processing node and four part of data center, external interface are received from outer
Extraneous female task and resource data are fed back to task team according to the request of task queue node by female task and resource data on boundary
Column node receives the task processing result of task queue node feeding back and issues outward.Task queue node store tasks (including
The new task that extraneous mother's task and task processing node generate) and resource data, be responsible for the fractionation of task with merge, and by task
Task processing node is distributed to, by resource data store to data center；Processing result from task processing node is carried out
External interface is fed back to after processing.Task handles node and obtains task and number of resources respectively from task queue node and data center
According to handling task, the new task generated in processing result and treatment process fed back to task queue node.Data
Center is for storing data.
External interface is present system and extraneous interface, is responsible for processing and extraneous interaction, receives from exterior
Resource data required for female task and operation feeds back extraneous female task and resource data according to the request of task queue node
Task queue node is given, the task processing result of task queue node feeding back is received and is issued outward.
Task queue node, store tasks and resource data, be responsible for task fractionation with merge, and assign the task to appoint
Business processing node, by resource data store to data center；It is anti-after handling the processing result from task processing node
It feeds external interface；Task includes extraneous female task and the new task that task processing node generates.
Task queue node include it is multiple, each task queue node judges whether according to current load situation to external
Mouth sends request.
Task queue node periodically carries out priority ranking to the task processing node of application calculating task, to preferential
The high node of grade distributes task, realizes node load balancing.
The extraneous next information of task queue node storage and internal temporary information, will acquire for task and resource data carry out
Number, obtain mission number and resource index number, then by mission number, resource index number, whether have subtask, whether be son
Into task queue lists, the mission number and resource index number are corresponded for task, subtask registration storage.Such as Fig. 2 institute
To show, wherein mission number is numbered by task reception sequence, in order to distinguish the number of female task and subtask, female task
Number is started by M, and the number of subtask is started by Z.Resource index number is in order to appoint task processing node in processing
The string number that corresponding resource data generates at random can be found when business, task queue node is when receiving task, by task
The resource data that number, resource index number and completion required by task are wanted stores in data center together.Whether there is subtask to be
In order to mark whether the task has been split into multiple subtasks.Whether be subtask be in order to mark the task whether be son appoint
Business.Subtask registration is an individual list, correctly can efficiently merge task to handle in subtask, deposit in list
It contains the mission numbers of all subtasks that mother's task is included, resource index number and whether completes.Task addition is taken office
Be engaged in queue lists in after, task handle node can to task queue node application task, meanwhile, task queue node need
Mission number, resource index number, resource data are supplied to data center.
Internal temporary information include: task queue list 1. to be allocated (the task queue list include three classes: first is that,
Female task of the most original obtained from external interface；Second is that complete the calculating of female task, multiple sons that female task is split into
Task；Third is that task handles node newly generated task when completing a certain task)；2. task processing result (task processing knot
Fruit includes: first is that, it is obtained from external interface, does not need to split merging, the mother of new task will not be generated during processing
Task processing result, this kind of result, which only needs task handling the result that node calculates, feeds back to external interface；Second is that complete
The calculating of female task, the processing result for multiple subtasks that female task is split into, this kind of result are needed by task queue node
It carries out just feeding back to external interface after result merges；Third is that task processing node is newly generated when completing a certain task
The processing result of task, this kind of result need to be carried out by task queue node just feed back to external interface after result treatment).
The calculating of one female task may be complex, in order to make full use of the performance of calculate node, improves at task
Efficiency is managed, a female task can be split into multiple subtasks after external interface acquisition task by task queue node,
For example, needing to split into female task into two subtasks to complete the processing of some female task, at this point, arranging in task queue
Need to be respectively created information of the column for storing them in table for two subtasks.Task processing node is responsible at subtask
Processing result is fed back to task queue node after task processing is completed by reason, and task queue node is receiving subtask
After processing result, by the corresponding subtask stored in task queue node whether complement mark position is to complete, when all
After subtask all handles completion, it is responsible for all subtask processing results being merged into female required by task by task queue node
It is wanting as a result, final feed back to external interface.
Task queue node handles what node provided according to task when the task of the task of response processing node handles application
Node soft hardware performance is analyzed (computer display card, CPU usage, memory usage, handling capacity etc.), periodically to Shen
Please calculating task task processing node carry out priority ranking, rationally carry out task distribution, give priority it is high node distribution
Task realizes the load balancing of node.
Task handles node, is responsible for obtaining task from task queue node, according to mission number and resource index number from number
The resource data that processing task needs is obtained according to center, the calculation processing of all female tasks and subtask is completed, by processing result
And the new task generated in treatment process feeds back to task queue node.Task processing node can be set to " hurry " and " sky
It is not busy " two states, when not carrying out task processing, the state that task handles node is " free time ", can be carried out at task at this time
Reason application；When getting a certain item calculating task from task queue node, state was both set as " hurrying ", was no longer appointed at this time
Business processing application.Task handles node after task queue node acquisition task, it is also desirable to safeguard a task processing column
Table, as shown in figure 3, including: mission number in task processing list, resource index number, whether generating new task, wherein task
Number and resource index number are obtained from task queue list, if are generated new task and referred to complete a certain item task and producing
Raw new task.Task handles node and needs while submitting and applying to task queue node by the software and hardware information of its own
(computer display card, CPU usage, memory usage, handling capacity etc.) is supplied to task queue node, is convenient for task queue node
Reasonable distribution task.
Task is handled after node gets a new task from task queue node, according to resource index number from data center
It obtains and completes the resource data that the required by task is wanted.Task handles node when carrying out the processing of a certain item task, may generate
The processing of new task, new task is the same with female task, needs to be added to task queue node, while needing to handle task
Required resource data is stored in data center.For example, data processing node needs to complete ancestral task, ancestral task is completed
During produce new task again, then task handles node new task is fed back to task queue node, by task team
New task is added in task queue list by column node, after task processing node completes new task processing, by task
The result that queue nodes complete ancestral task merges, and forms final processing result.
Data center is responsible for the storage of resource data, by the data resource of all tasks according to mission number, resource index
Number and resource data store into database.In order to abundant node data base resource, when task processing node is from the data
It is extracted the resource data of a certain item task in the heart, the corresponding resource index number of the task and resource data are both from data center
It deletes.
Female task refer to it is received from exterior, may be split into multiple subtasks, new post may be generated
The task of business.
In the present invention, when task queue nodes add new task, it is necessary to be stored in the resource data that task processing is related to
Data center, and when task is added to task queue list, it is necessary to additional resource call number and mission number；At task
Reason node obtains task from task queue, and in the task of processing, necessary from data center's acquisition by data directory number
Data, may generate new task in task processes, and the processing of new task is the same with female task, and female task and
The call number of subtask has certain correlation.
When female task splits into multiple subtasks, for the merging (result and data) for completing all subtasks, son need to be completed
Task registration and merging task registration.In in the merging task of registration include all subtasks number and resource index number,
Task queue node will carry out special maintenance to this merging task, that is, need to judge whether subtask completes.
The process flow of one female task A is described below in detail:
(1) according to request, external world mother task A is distributed to certain task queue node by external interface, and task queue node is by A
Two subtask B and subtask C are split into, task queue node stores subtask B and C according to table 1.Wherein times of A
Business number is M001, and corresponding resource index number is 01S23, and the mission number of subtask B is Z001, and resource index number is
The mission number of 11Z01, subtask C are Z002, and resource index number is 11Z02.Simultaneously task queue node by mission number, money
Source call number and resource data store are into database.
1 task queue list of table
(2) state is that the task of " free time " handles node to task queue node request task processing information, task queue
Node handles the performance of node according to task, considers system load balancing, subtask B and C are respectively allocated to different tasks
Handle node.
(3) it is " busy " that the receiving task of the task, which handles node for its status modifier, while task storage being handled to task
In list.
(4) the task processing node of subtask B and C are handled respectively according to the mission number Z001 and Z002 of B and C, is searched
To corresponding resource index 11Z01 and 11Z02, resource data required for processing B and C is obtained from data center, and carry out
The task of subtask B and C are handled.
(5) if in the when generation new task D for handling subtask B, task processing node is needed processing result, new post
Business D and the resource data for handling D needs feed back to task queue node；Task queue node according to task queue list lattice
New task D is added in task queue list by formula, and whether the list information for modifying subtask B (will have subtask by "No"
It is revised as "Yes", the mission number of new task D, resource index number are added in the subtask registration table of subtask B), simultaneously
The corresponding resource data of processing new task D is stored in data center.
(6) task processing node and task queue node carry out the processing of new task D according to step (2), (3), (4).
(7) after new task D processing is completed, task handles node and processing result is fed back to task queue node, together
When by new task D from task processing list in delete, by status modifier be " free time ".Task queue node is by task queue list
In the subtask registration table of middle subtask B whether completion by "No" is revised as "Yes", while merging the processing knot of subtask B
Fruit forms final processing result for B, then by subtask B in the subtask registration table of task A female in task queue list
Whether complete to be revised as "Yes" by "No".
(8) after C processing in subtask is completed, task handles node and processing result is fed back to task queue node, together
When by subtask C from task processing list in delete, by status modifier be " free time ".Task queue node is by task queue list
In the subtask registration table of middle mother task A subtask C whether completion by "No" is revised as "Yes".
(9) task queue node deletes subtask B and C from task queue list, and the processing of subtask B and C
As a result it merges, forms the processing result of female task A.Meanwhile female task A being deleted from task queue list, and will place
Reason result feeds back to external interface, and the processing result of female task is issued from external interface to exterior.
The present invention reduces analogue system towards the big Throughput Asynchronous task processing system of non-coupled streaming that LVC is emulated
Delay, improves communication efficiency, realizes the emulation of LVC integrated high-efficiency.
The content that description in the present invention is not described in detail belongs to the well-known technique of those skilled in the art.
Claims (11)
1. a kind of big Throughput Asynchronous task processing system of non-coupled streaming towards LVC emulation, it is characterised in that: including to external
Mouth, task queue node, task processing node and data center；
External interface: it receives from extraneous female task and resource data, according to the request of task queue node by extraneous female
Business and resource data feed back to task queue node, receive the task processing result of task queue node feeding back and issue outward；
Task queue node: store tasks and resource data, be responsible for the fractionation of task with merge, and assign the task at task
Node is managed, by resource data store to data center；It is fed back to after handling the processing result from task processing node
External interface；The task includes extraneous female task and the new task that task processing node generates；
Task handles node: task and resource data is obtained respectively from task queue node and data center, at task
Reason, feeds back to task queue node for the new task generated in processing result and treatment process；
Data center: storage resource data.
2. a kind of big Throughput Asynchronous task processing system of non-coupled streaming towards LVC emulation according to claim 1,
Be characterized in that: the task queue node include it is multiple, each task queue node according to current load situation judge whether to
External interface sends request.
3. a kind of big Throughput Asynchronous task processing system of non-coupled streaming towards LVC emulation according to claim 2,
Be characterized in that: the task and resource data that task queue node will acquire are numbered, and obtain mission number and resource index number,
Then by mission number, resource index number, whether have subtask, whether be subtask, subtask registration storage to task queue
In list, the mission number and resource index number are corresponded.
4. a kind of big Throughput Asynchronous task processing system of non-coupled streaming towards LVC emulation according to claim 3,
Be characterized in that: the subtask is registered as an individual list；
Task queue node when receiving task, according to resource judgment consumed by task complexity, calculating whether needs
Carry out task fractionation；
When task does not need to split, the subtask is registered as sky table；
When task needs to split, task of all subtasks that the task is included are stored in the subtask register list is compiled
Number, resource index number and whether complete.
5. a kind of big Throughput Asynchronous task processing system of non-coupled streaming towards LVC emulation according to claim 4,
Be characterized in that: task queue node is as follows handled the processing result of task processing node:
When predecessor's business is split, after all subtasks all handle completion, it is responsible for appointing all sons by task queue node
Processing result of being engaged in merges, and obtains the processing result of predecessor's business；
When predecessor is engaged in not being split, the processing result of corresponding task processing node is the processing result formerly held the post of and be engaged in.
6. a kind of big Throughput Asynchronous task processing system of non-coupled streaming towards LVC emulation according to claim 4,
Be characterized in that: task queue node by mission number, resource index number and completes the money wanted of required by task when receiving task
Source data stores in data center together.
7. a kind of big Throughput Asynchronous task processing system of non-coupled streaming towards LVC emulation according to claim 6,
It is characterized in that: when task queue node assigns the task to task processing node, while sending the task to task processing node
Mission number and resource index number.
8. a kind of big Throughput Asynchronous task processing system of non-coupled streaming towards LVC emulation according to claim 7,
Be characterized in that: task processing node has multiple, and each task, which handles node and submits to task queue node, applies, from task queue
Node obtains task, and obtains the resource data that processing task needs from data center according to resource index number, carries out task
Calculation processing；
When the new task not generated in task computation completion and treatment process, processing result is fed back into task queue node；
When task computation complete and generated in treatment process new task when, the new post that will be generated in processing result, treatment process
The resource data that business and new task need feeds back to task queue node；
When task computation does not complete, do not feed back.
9. a kind of big Throughput Asynchronous task processing system of non-coupled streaming towards LVC emulation according to claim 8,
Be characterized in that: the task processing node after task queue node acquisition task, is needing to safeguard a task processing column
Table, it includes: mission number which, which is handled in list, resource index number, whether generates new task, wherein mission number and resource
Call number is directly acquired from task queue node.
10. a kind of big Throughput Asynchronous task processing system of non-coupled streaming towards LVC emulation according to claim 1,
It needs to believe the software and hardware of its own it is characterized by: the task is handled while node is submitted and applied to task queue node
Breath is supplied to task queue node, and the software and hardware information includes computer display card, CPU usage, memory usage and handles up
Amount.
11. a kind of big Throughput Asynchronous task processing system of non-coupled streaming towards LVC emulation according to claim 10,
It is characterized by: the mode that the task queue node assigns the task to task processing node is as follows: task queue node week
Phase property carries out priority ranking according to software and hardware information to the task processing node of application calculating task, appoints to priority is high
Business processing node distributes task, realizes the load balancing of task processing node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811280764.5A CN109214132B (en) | 2018-10-30 | 2018-10-30 | LVC simulation-oriented uncoupled streaming type large-flux asynchronous task processing system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811280764.5A CN109214132B (en) | 2018-10-30 | 2018-10-30 | LVC simulation-oriented uncoupled streaming type large-flux asynchronous task processing system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109214132A true CN109214132A (en) | 2019-01-15 |
| CN109214132B CN109214132B (en) | 2023-06-30 |
Family
ID=64997612
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811280764.5A Active CN109214132B (en) | 2018-10-30 | 2018-10-30 | LVC simulation-oriented uncoupled streaming type large-flux asynchronous task processing system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109214132B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104731663A (en) * | 2015-03-31 | 2015-06-24 | 北京奇艺世纪科技有限公司 | Task processing method and system |
| CN106936899A (en) * | 2017-02-25 | 2017-07-07 | 九次方大数据信息集团有限公司 | The collocation method of distributed statistical analysis system and distributed statistical analysis system |
| CN107632890A (en) * | 2017-08-10 | 2018-01-26 | 北京中科睿芯科技有限公司 | Dynamic node distribution method and system in a kind of data stream architecture |
| CN107707592A (en) * | 2017-01-24 | 2018-02-16 | 贵州白山云科技有限公司 | Task processing method, node and content distributing network |
| CN107743246A (en) * | 2017-01-24 | 2018-02-27 | 贵州白山云科技有限公司 | Task processing method, system and data handling system |
| CN107766129A (en) * | 2016-08-17 | 2018-03-06 | 北京金山云网络技术有限公司 | A kind of task processing method, apparatus and system |
| CN107766572A (en) * | 2017-11-13 | 2018-03-06 | 北京国信宏数科技有限责任公司 | Distributed extraction and visual analysis method and system based on economic field data |
| WO2018121738A1 (en) * | 2016-12-30 | 2018-07-05 | 北京奇虎科技有限公司 | Method and apparatus for processing streaming data task |
-
2018
- 2018-10-30 CN CN201811280764.5A patent/CN109214132B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104731663A (en) * | 2015-03-31 | 2015-06-24 | 北京奇艺世纪科技有限公司 | Task processing method and system |
| CN107766129A (en) * | 2016-08-17 | 2018-03-06 | 北京金山云网络技术有限公司 | A kind of task processing method, apparatus and system |
| WO2018121738A1 (en) * | 2016-12-30 | 2018-07-05 | 北京奇虎科技有限公司 | Method and apparatus for processing streaming data task |
| CN107707592A (en) * | 2017-01-24 | 2018-02-16 | 贵州白山云科技有限公司 | Task processing method, node and content distributing network |
| CN107743246A (en) * | 2017-01-24 | 2018-02-27 | 贵州白山云科技有限公司 | Task processing method, system and data handling system |
| CN106936899A (en) * | 2017-02-25 | 2017-07-07 | 九次方大数据信息集团有限公司 | The collocation method of distributed statistical analysis system and distributed statistical analysis system |
| CN107632890A (en) * | 2017-08-10 | 2018-01-26 | 北京中科睿芯科技有限公司 | Dynamic node distribution method and system in a kind of data stream architecture |
| CN107766572A (en) * | 2017-11-13 | 2018-03-06 | 北京国信宏数科技有限责任公司 | Distributed extraction and visual analysis method and system based on economic field data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109214132B (en) | 2023-06-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105159762B (en) | Heuristic cloud computing method for scheduling task based on Greedy strategy | |
| CN104391918B (en) | The implementation method of distributed networks database query priority management based on equity deployment | |
| CN103034614B (en) | Single task multi-core dispatching method based on critical path and Task Duplication | |
| CN107710200A (en) | System and method for the operator based on hash in parallelization SMP databases | |
| CN105260818B (en) | Mix the on-line optimizing scheduling method with deadline constraint workflow group under cloud environment | |
| CN110348571A (en) | A kind of neural network model training method, device, chip and system | |
| CN101436959A (en) | Method for distributing and scheduling parallel artificial tasks based on background management and control architecture | |
| CN105892996A (en) | Assembly line work method and apparatus for batch data processing | |
| CN112596902A (en) | Task scheduling method and device based on CPU-GPU cooperative computing | |
| CN109522108A (en) | A kind of GPU task scheduling system and method merged based on Kernel | |
| CN109409773A (en) | A kind of earth observation resource dynamic programming method based on Contract Net Mechanism | |
| CN105573850B (en) | Multi-process exchange method, system and server | |
| CN103885826B (en) | Real-time task scheduling implementation method of multi-core embedded system | |
| CN106325976B (en) | A kind of rendering task scheduling processing method and server | |
| CN101271405A (en) | Bidirectional grade gridding resource scheduling method based on QoS restriction | |
| CN105930417B (en) | A kind of big data ETL interactive process platform based on cloud computing | |
| CN110347504A (en) | Many-core computing resource dispatching method and device | |
| CN105631921B (en) | The processing method and processing device of image data | |
| Wang et al. | Task scheduling algorithm based on improved Min-Min algorithm in cloud computing environment | |
| CN107273339A (en) | A kind of task processing method and device | |
| CN105389204A (en) | Multiple-resource partial order scheduling policy | |
| CN109409746A (en) | A kind of production scheduling method and device | |
| CN104537713B (en) | A kind of novel three-dimensional reconfiguration system | |
| US20100004916A1 (en) | Process Analyzer | |
| CN105430074B (en) | Optimization method and system based on the distribution storage of the cloud data of data dependency and visit capacity |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |