CN106155770A - Method for scheduling task and electronic equipment - Google Patents
Method for scheduling task and electronic equipment Download PDFInfo
- Publication number
- CN106155770A CN106155770A CN201510145555.XA CN201510145555A CN106155770A CN 106155770 A CN106155770 A CN 106155770A CN 201510145555 A CN201510145555 A CN 201510145555A CN 106155770 A CN106155770 A CN 106155770A
- Authority
- CN
- China
- Prior art keywords
- node
- virtualization
- task
- calculating
- burst
- 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
Landscapes
- Debugging And Monitoring (AREA)
- Hardware Redundancy (AREA)
Abstract
The present invention provides a kind of method for scheduling task and electronic equipment.Described method includes: receives task to be processed, and is multiple task bursts by described Task-decomposing；The plurality of task burst is distributed to the respective virtualization in described virtualization computing cluster and calculates node, respective virtualization calculate node and the tasks carrying of distribution is calculated operation；Receiving and calculate, from respective virtualization, the indication signal that node timing returns, described indication signal indicates the whether normal operating of described calculating node；When not receiving indication signal from specific virtualization calculating node, determining that this specific virtualization calculates node is malfunctioning node, and determines whether to meet predetermined condition；And when a predetermined condition is satisfied, wait that this specific virtualization calculates node and recovers the calculating operation to continue executing with allocated task burst.
Description
Technical field
The application relates to method for scheduling task and electronic equipment.
Background technology
MapReduce is to operate on physical machine cluster at the beginning of design, it is contemplated that unreliable to physical machine
Property, MapReduce has carried out the design of a lot of fault-tolerance in task scheduling process.And it is the most a lot
MapReduce application all operates on virtual platform, virtual platform and traditional physical machine collection faciation
Ratio, it is provided that preferably guaranteed reliability.For the original task scheduling of this situation MapReduce
Method is accomplished by being correspondingly improved.
For this reason, it may be desirable to provide a kind of method for scheduling task and electronic equipment, it can calculate for virtualization
Platform preferably scheduler task.
Summary of the invention
According to embodiments of the present invention, it is provided that a kind of method for scheduling task, it is applied to virtualize computing cluster
System, described method includes:
Receive task to be processed, and be multiple task bursts by described Task-decomposing；
The plurality of task burst is distributed to the respective virtualization in described virtualization computing cluster and calculates joint
Point, is calculated node by respective virtualization and the tasks carrying of distribution is calculated operation；
Calculating node from respective virtualization and receive the indication signal that timing returns, described indication signal indicates institute
State calculating node whether normal operating；
When not receiving indication signal from specific virtualization calculating node, determine this specific virtualization meter
Operator node is malfunctioning node, and determines whether to meet predetermined condition；
When a predetermined condition is satisfied, wait that this specific virtualization calculates node and recovers to distribute to continue executing with
Task burst calculating operation.
Preferably, it is determined whether meet predetermined condition and also include:
Estimate to re-execute the meter needed for distributing to the first task burst of this specific virtualization calculating node
Evaluation time；
The recovery time that the calculating time of estimation calculates node with the virtualization pre-set is compared；
If the calculating time estimated is more than described recovery time, it is determined that meet predetermined condition.
Preferably, needed for estimating to re-execute the task burst distributing to this specific virtualization calculating node
The calculating time also include:
Obtain the information of the unfinished portion distributing to the task burst that this specific virtualization calculates node；
Information according to the unfinished portion obtained estimates to re-execute the unfinished portion institute of this task burst
The calculating time needed.
Preferably, described method also includes:
When being unsatisfactory for predetermined condition, by the task burst distributing to this specific virtualization calculating node
Unfinished portion distribute to other virtualization calculate node.
Preferably, described method also includes:
According to apart from order, from closely to far distributing to the task burst of this specific virtualization calculating node
In unfinished portion distribute to other virtualization calculate node.
Preferably, described method also includes:
Receive at the task burst of each return of the respective virtualization calculating node performing to calculate operation
Reason result；And
Assemble all of result to generate result to be exported.
According to another embodiment of the present invention, it is provided that a kind of electronic equipment, including:
Task-decomposing module, is configured to receive task to be processed, and is multiple by described Task-decomposing
Business burst；
Task allocating module, is configured to distribute to the plurality of task burst virtualize in computing cluster
Respective virtualization calculates node, respective virtualization calculate node and the tasks carrying of distribution is calculated operation；
Control module, is configured to calculate node from respective virtualization and receives the indication signal that timing returns, institute
State indication signal and indicate the whether normal operating of described calculating node,
Wherein, when not receiving indication signal from specific virtualization calculating node, described control module
Determining that this specific virtualization calculates node is malfunctioning node, and determines whether to meet predetermined condition；And
When a predetermined condition is satisfied, described control module waits that this specific virtualization calculates node and recovers to continue
The continuous calculating operation performing allocated task burst.
Preferably, described control module is further configured to:
Estimate to re-execute the meter needed for distributing to the first task burst of this specific virtualization calculating node
Evaluation time；
The recovery time that the calculating time of estimation calculates node with the virtualization pre-set is compared；
If the calculating time estimated is more than described recovery time, it is determined that meet predetermined condition.
Preferably, described control module is further configured to:
Obtain the information of the unfinished portion distributing to the task burst that this specific virtualization calculates node；
Information according to the unfinished portion obtained estimates to re-execute the unfinished portion institute of this task burst
The calculating time needed.
Preferably, described task allocating module is further configured to
When being unsatisfactory for predetermined condition, by the task burst distributing to this specific virtualization calculating node
Unfinished portion distribute to other virtualization calculate node.
Preferably, described task allocating module is further configured to:
According to apart from order, from closely to far distributing to the task burst of this specific virtualization calculating node
In unfinished portion distribute to other virtualization calculate node.
Preferably, described control module is further configured to
Receive at the task burst of each return of the respective virtualization calculating node performing to calculate operation
Reason result；And
Assemble all of result to generate result to be exported.
Therefore, method for scheduling task according to embodiments of the present invention and electronic equipment, it is possible to for virtualization
Calculate platform preferably scheduler task.
Accompanying drawing explanation
Fig. 1 is the flow chart illustrating method for scheduling task according to a first embodiment of the present invention；
Fig. 2 is the flow chart illustrating method for scheduling task according to a second embodiment of the present invention；And
Fig. 3 is the functional configuration block diagram illustrating electronic equipment according to a third embodiment of the present invention.
Detailed description of the invention
Below, method for scheduling task according to embodiments of the present invention will be described in detail with reference to the attached drawings and electronics sets
Standby.
<first embodiment>
Below, method for scheduling task according to a first embodiment of the present invention will be described in detail with reference to Fig. 1.
Method for scheduling task is applied to virtualize computing cluster system according to embodiments of the present invention.
Described method 100 includes:
Step S101: receive task to be processed, and be multiple task bursts by described Task-decomposing；
Step S102: the plurality of task burst is distributed to the corresponding void in described virtualization computing cluster
Planization calculates node, respective virtualization calculate node and the tasks carrying of distribution is calculated operation；
Step S103: receive the indication signal that timing returns, described indication signal indicates described calculating node
Whether normal operating；
Step S104: when not receiving indication signal from specific virtualization calculating node, determine this spy
It is malfunctioning node that fixed virtualization calculates node, and determines whether to meet predetermined condition；
Step S105: when a predetermined condition is satisfied, waits that this specific virtualization calculates node and recovers to continue
Perform the calculating operation of allocated task burst.
Specifically, in step S101, such as, received by task management node (JobTracker) and want
The task (Job) processed, then, is multiple task bursts (Task) by described Task-decomposing.Such as,
Receive long data block to be processed, this long data block is divided into multiple small data block, say, that be every
Individual task burst corresponds to a small data block.
Then, in step s 102, the plurality of task burst is distributed to described virtualization computing cluster
In respective virtualization calculate node (TaskTracker), by respective virtualization calculate node distribution is appointed
Business performs to calculate operation (that is, Mpa/Reduce task).
Specifically, virtualization computing cluster includes that multiple scattered virtualization calculates node (that is, virtual machine).
In step s 102, virtualization meter is distributed to by dividing each task burst obtained in step S101
The virtualization calculated in cluster calculates node, corresponding virtualization calculate node and hold the task of distribution
Row calculates operation.It should be noted that distribution task burst can be appointing in virtualization computing cluster
One virtualization calculates node.
Then, in step s 103, receive and calculate, from respective virtualization, the indication signal that node timing returns,
Described indication signal indicates the whether normal operating of described calculating node.
Such as, the virtualization calculating node timing performing to calculate operation calculates joint to the virtualization of distribution task
Point sends heartbeat signal.The virtualization that this heartbeat signal indicates this execution to calculate operation is just calculating nodal operation
Often.
Then, in step S104, when not receiving indication signal from specific virtualization calculating node,
Determining that this specific virtualization calculates node is malfunctioning node.It is to say, without from specific virtualization
Calculate node and receive heartbeat signal, then may determine that this specific virtualization node current failure, i.e. do not have
Heart beating.Then, it is determined that this virtualization calculates whether node meets predetermined condition.
Then, in step S105, when a predetermined condition is satisfied, wait that this specific virtualization calculates node
Recover the calculating to continue executing with allocated task burst to operate.
Because when when virtualizing operation in computing cluster system, virtual machine can be the completeest
Full recovery to the state (including the data on internal memory and network) before the machine of delaying, now in virtual machine before run
All Task can continue executing with, re-execute without on other machine.
It is to say, it is different from the scheme of prior art, when determining that performing to calculate the virtualization thought calculates
During one malfunctions, it not deleting immediately and distributing to this calculating node of task, but first judge whether
Meet predetermined condition, and when a predetermined condition is satisfied, do not delete the task of distributing to this calculating node.
On the contrary, when a predetermined condition is satisfied, wait that this specific virtualization calculates node and recovers to divide to continue executing with
The calculating operation of the task burst joined.
Therefore, method for scheduling task according to embodiments of the present invention, at the virtualization meter performing calculating task
When operator node breaks down, it is possible to judge satisfied predetermined according to oneself state, it is judged that wait this calculating node
Restart rather than by distribution of computation tasks to other node such that it is able to calculate platform for virtualization more preferable
Ground scheduler task, greatly improves computational efficiency.
<the second embodiment>
Below, method for scheduling task according to a second embodiment of the present invention will be described in detail with reference to Fig. 2.
Described method 200 includes:
Step S201: receive task to be processed, and be multiple task bursts by described Task-decomposing；
Step S202: the plurality of task burst is distributed to the corresponding void in described virtualization computing cluster
Planization calculates node, respective virtualization calculate node and the tasks carrying of distribution is calculated operation；
Step S203: receive the indication signal that timing returns, described indication signal indicates described calculating node
Whether normal operating；
Step S204: when not receiving indication signal from specific virtualization calculating node, determine this spy
It is malfunctioning node that fixed virtualization calculates node；
Step S205: estimate re-execute distribute to this specific virtualization calculate node first task divide
The calculating time needed for sheet, the calculating time of estimation and the virtualization pre-set are calculated the recovery of node
Time compares；
Step S206: when the calculating time estimated is more than described recovery time, wait this specific virtualization
Calculate node and recover the calculating operation to continue executing with allocated task burst.
Specifically, in step s 201, task to be processed (Job) is received, then, by described task
It is decomposed into multiple task burst (Task).Such as, receive long data block to be processed, by this long data block
It is divided into multiple small data block, say, that each task burst corresponds to a small data block.
Then, in step S202, the plurality of task burst is distributed to described virtualization computing cluster
In respective virtualization calculate node, by respective virtualization calculate node to distribution tasks carrying calculate behaviour
Make.
Specifically, virtualization computing cluster includes that multiple scattered virtualization calculates node (that is, virtual machine).
In step S202, distribute to virtualization meter by dividing each task burst obtained in step s 201
The virtualization calculated in cluster calculates node, corresponding virtualization calculate node and hold the task of distribution
Row calculates operation.It should be noted that distribution task burst can be appointing in virtualization computing cluster
One virtualization calculates node.
Then, in step S203, receive and calculate, from respective virtualization, the indication signal that node timing returns,
Described indication signal indicates the whether normal operating of described calculating node.
Such as, the virtualization calculating node timing performing to calculate operation calculates joint to the virtualization of distribution task
Point sends heartbeat signal.The virtualization that this heartbeat signal indicates this execution to calculate operation is just calculating nodal operation
Often.
Then, in step S204, when not receiving indication signal from specific virtualization calculating node,
Determining that this specific virtualization calculates node is malfunctioning node.It is to say, without from specific virtualization
Calculate node and receive heartbeat signal, then may determine that this specific virtualization node current failure, i.e. do not have
Heart beating.
Then, in step S205, estimate to re-execute and distribute to this specific virtualization calculating node
The calculating time needed for first task burst, the calculating time of estimation is calculated with the virtualization pre-set
The recovery time of node compares.
Specifically, in one embodiment, it is determined whether meet predetermined condition and also include:
Estimate to re-execute the meter needed for distributing to the first task burst of this specific virtualization calculating node
Evaluation time；The recovery time that the calculating time of estimation calculates node with the virtualization pre-set is compared；
If the calculating time estimated is more than described recovery time, it is determined that meet predetermined condition.
The first of this specific virtualization calculating node is distributed to it is to say, first estimate to re-execute
The calculating time needed for business burst, for example, it is assumed that calculated node by another virtualization to re-execute this
The calculating time needed for first task burst is 5 minutes.
Then, read the virtualization pre-set and calculate the recovery time of node, for example, it is assumed that each void
Planization calculates and is set to 1 minute the recovery time of node.By the calculating time of estimation and the void pre-set
Planization calculates and compares the recovery time of node, it may be determined that the calculating time (5 minutes) of estimation is more than extensive
The multiple time (1 minute).At this point it is possible to determine and meet this predetermined condition.
Then, in step S206, when the calculating time estimated is more than described recovery time, waiting should
Specific virtualization calculates node and recovers the calculating operation to continue executing with allocated task burst.
It is to say, it is different from the scheme of prior art, when determining that the virtualization performing calculating task calculates
During one malfunctions, it not deleting immediately and distributing to this calculating node of task, but first judge whether
Meet predetermined condition, and when a predetermined condition is satisfied, do not delete the task of distributing to this calculating node.
On the contrary, when a predetermined condition is satisfied, wait that this specific virtualization calculates node and recovers to divide to continue executing with
The calculating operation of the task burst joined.
On the other hand, when being unsatisfactory for predetermined condition, say, that when the calculating time estimated is less than institute
When stating recovery time, can by distribute to this specific virtualization calculate node task burst in the completeest
One-tenth part is distributed to other virtualization and is calculated node.
In one embodiment, estimate to re-execute the task of distributing to this specific virtualization calculating node
The calculating time needed for burst also includes:
Obtain the information of the unfinished portion distributing to the task burst that this specific virtualization calculates node；
Information according to the unfinished portion obtained estimates to re-execute the unfinished portion institute of this task burst
The calculating time needed.
It is to say, distribute to this specific first task virtualizing calculating node estimating to re-execute
During calculating time needed for burst, it is not necessary to estimate to re-execute needed for whole first task bursts time
Between, but can first obtain the unfinished portion distributing to the task burst that this specific virtualization calculates node
Information.
Such as, calculate, according to this virtualization of caching in virtualized computing system, the task that node is complete
The degree (such as, 40%) of burst, it may be determined that unfinished portion is 60%.
Then, estimate to re-execute being not fully complete of this task burst according to the information of the unfinished portion obtained
The calculating time needed for part.That is, determined that the time needed for the calculating of this unfinished portion is 5 points
Clock * 60%, is 3 minutes.
Therefore, re-executed the calculating time needed for the unfinished portion of this task burst by estimation, can
By more accurately determine be to wait for this calculating node recover or this first task burst is distributed in terms of other
Operator node.In this embodiment, because the time needed for completing remainder is 3 minutes, still above
Restart the time needed for this calculating node, i.e. 1 minute.
Therefore, continue waiting for this calculating node to recover.
On the other hand, virtual management platform detect specific virtualization calculate one malfunctions time,
This specific virtualization can be migrated and calculate state and the data of node, and start to recover this virtualization calculating
Node.
By the time this virtualization calculates after node is successfully recovered, and this when can will appear from fault is specific virtual
State and the Data Migration of changing calculating node are returned, and make this virtualization calculate node from breaking down
Time state continue to start to perform calculating, and send heartbeat signal to host node.
Host node is after receiving heartbeat signal, and redefining this calculating node is normal condition.
In one embodiment, described method also includes:
According to apart from order, from closely to far distributing to the task burst of this specific virtualization calculating node
In unfinished portion distribute to other virtualization calculate node.
Specifically, when the unfinished portion of task burst being distributed to other virtualization and calculating node, press
According to order from back to front, the unfinished portion of this task burst is distributed to other virtualization and calculates node.
In one embodiment, described method also includes:
Receive at the task burst of each return of the respective virtualization calculating node performing to calculate operation
Reason result；And
Assemble all of result to generate result to be exported.
It is to say, each virtualization calculating node is after completing calculating task, the calculating of oneself is tied
Fruit returns to the host node of distribution task, is then assembled all of result by host node defeated to generate
The result gone out, i.e. for the result of calculation of the task of distribution.
Therefore, method for scheduling task according to embodiments of the present invention, at the virtualization meter performing calculating task
When operator node breaks down, it is possible to judge that being to wait for this calculating node restarts maybe by calculating according to oneself state
Task distributes to other node such that it is able to calculate platform preferably scheduler task for virtualization, significantly
Ground improves computational efficiency.
<the 3rd embodiment>
Below, with reference to Fig. 3, electronic equipment according to a third embodiment of the present invention will be described.
Electronic equipment 300 according to a third embodiment of the present invention includes:
Task-decomposing module 301, is configured to receive task to be processed, and is many by described Task-decomposing
Individual task burst；
Task allocating module 302, is configured to distribute to the plurality of task burst virtualize computing cluster
In respective virtualization calculate node, by respective virtualization calculate node to distribution tasks carrying calculate behaviour
Make；
Control module 303, is configured to receive the indication signal that timing returns, and described indication signal indicates institute
State calculating node whether normal operating,
Wherein, when not receiving indication signal from specific virtualization calculating node, described control module
303 determine that this specific virtualization calculating node is malfunctioning node, and determine whether to meet predetermined condition；With
And
When a predetermined condition is satisfied, described control module 303 waits that this specific virtualization calculates node and recovers
To continue executing with the calculating operation of allocated task burst.
Preferably, described control module 303 is further configured to:
Estimate to re-execute the meter needed for distributing to the first task burst of this specific virtualization calculating node
Evaluation time；
The recovery time that the calculating time of estimation calculates node with the virtualization pre-set is compared；
If the calculating time estimated is more than described recovery time, it is determined that meet predetermined condition.
Preferably, described control module 303 is further configured to:
Obtain the information of the unfinished portion distributing to the task burst that this specific virtualization calculates node；
Information according to the unfinished portion obtained estimates to re-execute the unfinished portion institute of this task burst
The calculating time needed.
Preferably, described task allocating module 302 is further configured to
When being unsatisfactory for predetermined condition, by the task burst distributing to this specific virtualization calculating node
Unfinished portion distribute to other virtualization calculate node.
Preferably, described task allocating module 302 is further configured to:
According to apart from order, from closely to far distributing to the task burst of this specific virtualization calculating node
In unfinished portion distribute to other virtualization calculate node.
Preferably, described control module 303 is further configured to
Receive at the task burst of each return of the respective virtualization calculating node performing to calculate operation
Reason result；And
Assemble all of result to generate result to be exported.
Therefore, method for scheduling task according to embodiments of the present invention and electronic equipment, it is possible to for virtualization
Calculate platform preferably scheduler task.
Therefore, electronic equipment according to embodiments of the present invention, calculate joint in the virtualization performing calculating task
When point breaks down, it is possible to judge that being to wait for this calculating node restarts the task that maybe will calculate according to oneself state
Distribute to other node such that it is able to calculate platform preferably scheduler task for virtualization, greatly carry
Computationally efficient.
It should be noted that above embodiment is intended merely as example, the invention is not restricted to such showing
Example, but various change can be carried out.
Such as, this electronic equipment can be to include multiple physical machine cluster with computing capability.
It should be noted that in this manual, term " include ", " comprising " or its any other become
Body is intended to comprising of nonexcludability, so that include the process of a series of key element, method, article
Or equipment not only includes those key elements, but also includes other key elements being not expressly set out, or
Also include the key element intrinsic for this process, method, article or equipment.There is no more restriction
In the case of, statement " including ... " key element limited, it is not excluded that in the mistake including described key element
Journey, method, article or equipment there is also other identical element.
Finally, in addition it is also necessary to explanation, above-mentioned a series of process not only include with order described here by
The process that time series performs, and include the place performed parallel or respectively rather than in chronological order
Reason.
Through the above description of the embodiments, those skilled in the art is it can be understood that arrive this
Bright can add the mode of required hardware platform by software and realize, naturally it is also possible to all be come by hardware
Implement.Based on such understanding, technical scheme background technology is contributed whole or
Part can embody with the form of software product, and this computer software product can be stored in storage and be situated between
In matter, such as ROM (read only memory)/RAM (random access memory), magnetic disc, CD etc., bag
Include some instructions with so that a computer equipment (can be personal computer, server, or net
Network equipment etc.) perform each embodiment of the present invention or the method described in some part of embodiment.
Above the present invention is described in detail, the specific case principle to the present invention used herein
And embodiment is set forth, the method that the explanation of above example is only intended to help to understand the present invention
And core concept；Simultaneously for one of ordinary skill in the art, according to the thought of the present invention,
All will change in detailed description of the invention and range of application, in sum, this specification content should not
It is interpreted as limitation of the present invention.
Claims (12)
1. a method for scheduling task, is applied to virtualize computing cluster system, and described method includes:
Receive task to be processed, and be multiple task bursts by described Task-decomposing；
The plurality of task burst is distributed to the respective virtualization in described virtualization computing cluster and calculates joint
Point, is calculated node by respective virtualization and the tasks carrying of distribution is calculated operation；
Receiving and calculate, from respective virtualization, the indication signal that node timing returns, described indication signal indicates institute
State calculating node whether normal operating；
When not receiving indication signal from specific virtualization calculating node, determine this specific virtualization meter
Operator node is malfunctioning node, and determines whether to meet predetermined condition；And
When a predetermined condition is satisfied, wait that this specific virtualization calculates node and recovers to distribute to continue executing with
Task burst calculating operation.
The most the method for claim 1, wherein determine whether that meeting predetermined condition also includes:
Estimate to re-execute the meter needed for distributing to the first task burst of this specific virtualization calculating node
Evaluation time；
The recovery time that the calculating time of estimation calculates node with the virtualization pre-set is compared；
If the calculating time estimated is more than described recovery time, it is determined that meet predetermined condition.
3. method as claimed in claim 2, wherein, estimates to re-execute that to distribute to this specific virtual
The calculating time needed for calculating the task burst of node of changing also includes:
Obtain the information of the unfinished portion distributing to the task burst that this specific virtualization calculates node；
Information according to the unfinished portion obtained estimates to re-execute the unfinished portion institute of this task burst
The calculating time needed.
4. the method for claim 1, also includes:
When being unsatisfactory for predetermined condition, by the task burst distributing to this specific virtualization calculating node
Unfinished portion distribute to other virtualization calculate node.
5. the method for claim 1, also includes:
According to apart from order, from closely to far distributing to the task burst of this specific virtualization calculating node
In unfinished portion distribute to other virtualization calculate node.
6. the method for claim 1, also includes:
Receive the process calculating each task burst that node returns from the respective virtualization performing to calculate operation
Result；And
Assemble all of result to generate result to be exported.
7. an electronic equipment, including:
Task-decomposing module, is configured to receive task to be processed, and is multiple by described Task-decomposing
Business burst；
Task allocating module, is configured to distribute to the plurality of task burst virtualize in computing cluster
Respective virtualization calculates node, respective virtualization calculate node and the tasks carrying of distribution is calculated operation；
Control module, is configured to calculate node from respective virtualization and receives the indication signal that timing returns, institute
State indication signal and indicate the whether normal operating of described calculating node,
Wherein, when not receiving indication signal from specific virtualization calculating node, described control module
Determining that this specific virtualization calculates node is malfunctioning node, and determines whether to meet predetermined condition；And
When a predetermined condition is satisfied, described control module waits that this specific virtualization calculates node and recovers to continue
The continuous calculating operation performing allocated task burst.
8. electronic equipment as claimed in claim 7, wherein, described control module is further configured to:
Estimate to re-execute the meter needed for distributing to the first task burst of this specific virtualization calculating node
Evaluation time；
The recovery time that the calculating time of estimation calculates node with the virtualization pre-set is compared；
If the calculating time estimated is more than described recovery time, it is determined that meet predetermined condition.
9. method as claimed in claim 8, wherein, described control module is further configured to:
Obtain the information of the unfinished portion distributing to the task burst that this specific virtualization calculates node；
Information according to the unfinished portion obtained estimates to re-execute the unfinished portion institute of this task burst
The calculating time needed.
10. electronic equipment as claimed in claim 9, wherein, described task allocating module is joined further
It is set to
When being unsatisfactory for predetermined condition, by the task burst distributing to this specific virtualization calculating node
Unfinished portion distribute to other virtualization calculate node.
11. electronic equipments as claimed in claim 7, wherein, described task allocating module is joined further
It is set to:
According to apart from order, from closely to far distributing to the task burst of this specific virtualization calculating node
In unfinished portion distribute to other virtualization calculate node.
12. electronic equipments as claimed in claim 7, wherein, described control module is further configured to
Receive at the task burst of each return of the respective virtualization calculating node performing to calculate operation
Reason result；And
Assemble all of result to generate result to be exported.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510145555.XA CN106155770B (en) | 2015-03-30 | 2015-03-30 | Method for scheduling task and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510145555.XA CN106155770B (en) | 2015-03-30 | 2015-03-30 | Method for scheduling task and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106155770A true CN106155770A (en) | 2016-11-23 |
| CN106155770B CN106155770B (en) | 2019-11-26 |
Family
ID=57340307
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510145555.XA Active CN106155770B (en) | 2015-03-30 | 2015-03-30 | Method for scheduling task and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106155770B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106850354A (en) * | 2017-02-22 | 2017-06-13 | 郑州云海信息技术有限公司 | A kind of processing method and processing device of Single Point of Faliure |
| CN111818159A (en) * | 2020-07-08 | 2020-10-23 | 腾讯科技（深圳）有限公司 | Data processing node management method, device, equipment and storage medium |
| CN114691241A (en) * | 2022-04-19 | 2022-07-01 | 中煤航测遥感集团有限公司 | Task execution method and device, electronic equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6041425A (en) * | 1996-09-03 | 2000-03-21 | Hitachi, Ltd. | Error recovery method and apparatus in a computer system |
| US7900206B1 (en) * | 2004-03-31 | 2011-03-01 | Symantec Operating Corporation | Information technology process workflow for data centers |
| CN103019889A (en) * | 2012-12-21 | 2013-04-03 | 曙光信息产业(北京)有限公司 | Distributed file system and failure processing method thereof |
| CN103051691A (en) * | 2012-12-12 | 2013-04-17 | 华为技术有限公司 | Subarea distribution method, device and distributed type storage system |
| CN103259832A (en) * | 2012-12-24 | 2013-08-21 | 中国科学院沈阳自动化研究所 | Cluster resource control method for achieving dynamic load balance, fault diagnosis and failover |
| CN103440160A (en) * | 2013-08-15 | 2013-12-11 | 华为技术有限公司 | Virtual machine recovering method and virtual machine migration method , device and system |
| CN104461752A (en) * | 2014-11-21 | 2015-03-25 | 浙江宇视科技有限公司 | Two-level fault-tolerant multimedia distributed task processing method |
-
2015
- 2015-03-30 CN CN201510145555.XA patent/CN106155770B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6041425A (en) * | 1996-09-03 | 2000-03-21 | Hitachi, Ltd. | Error recovery method and apparatus in a computer system |
| US7900206B1 (en) * | 2004-03-31 | 2011-03-01 | Symantec Operating Corporation | Information technology process workflow for data centers |
| CN103051691A (en) * | 2012-12-12 | 2013-04-17 | 华为技术有限公司 | Subarea distribution method, device and distributed type storage system |
| CN103019889A (en) * | 2012-12-21 | 2013-04-03 | 曙光信息产业(北京)有限公司 | Distributed file system and failure processing method thereof |
| CN103259832A (en) * | 2012-12-24 | 2013-08-21 | 中国科学院沈阳自动化研究所 | Cluster resource control method for achieving dynamic load balance, fault diagnosis and failover |
| CN103440160A (en) * | 2013-08-15 | 2013-12-11 | 华为技术有限公司 | Virtual machine recovering method and virtual machine migration method , device and system |
| CN104461752A (en) * | 2014-11-21 | 2015-03-25 | 浙江宇视科技有限公司 | Two-level fault-tolerant multimedia distributed task processing method |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106850354A (en) * | 2017-02-22 | 2017-06-13 | 郑州云海信息技术有限公司 | A kind of processing method and processing device of Single Point of Faliure |
| CN111818159A (en) * | 2020-07-08 | 2020-10-23 | 腾讯科技（深圳）有限公司 | Data processing node management method, device, equipment and storage medium |
| CN111818159B (en) * | 2020-07-08 | 2024-04-05 | 腾讯科技（深圳）有限公司 | Management method, device, equipment and storage medium of data processing node |
| CN114691241A (en) * | 2022-04-19 | 2022-07-01 | 中煤航测遥感集团有限公司 | Task execution method and device, electronic equipment and storage medium |
| CN114691241B (en) * | 2022-04-19 | 2024-01-19 | 中煤航测遥感集团有限公司 | Task execution method, device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106155770B (en) | 2019-11-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9244751B2 (en) | Estimating a performance parameter of a job having map and reduce tasks after a failure | |
| CN106796540A (en) | For the active fault recovery model of Distributed Calculation | |
| CN106326002B (en) | Resource scheduling method, device and equipment | |
| CN105871603B (en) | A kind of the real time streaming data processing fail recovery and method of data grids based on memory | |
| CN104063293B (en) | A kind of data back up method and stream calculation system | |
| CN104794015B (en) | A kind of real-time stream calculation flow velocity perceives elasticity and performs tolerant system | |
| CN104035836B (en) | Automatic disaster-tolerant recovery method and system in cluster searching platform | |
| JP6424632B2 (en) | Load calculation method, load calculation program and load calculation device | |
| CN109471770A (en) | A kind of method for managing system and device | |
| CN108633311A (en) | A kind of method, apparatus and control node of the con current control based on call chain | |
| CN107273182B (en) | Method and system for dynamically expanding virtual machine resources | |
| CN115328663A (en) | Method, device, equipment and storage medium for scheduling resources based on PaaS platform | |
| CN103677990B (en) | Dispatching method, device and the virtual machine of virtual machine real-time task | |
| CN103823748B (en) | Partitioning software reliability analyzing method based on stochastic Petri net | |
| CN106155770A (en) | Method for scheduling task and electronic equipment | |
| CN103677756A (en) | System and method for synchronizing processor instruction execution | |
| CN108958975A (en) | A kind of method, device and equipment controlling data resume speed | |
| CN107844406A (en) | Method for detecting abnormality and system, service terminal, the memory of distributed system | |
| CN114064217B (en) | OpenStack-based node virtual machine migration method and device | |
| CN106991070A (en) | Real-time computing technique and device | |
| CN108647007B (en) | Computing system and chip | |
| US9329916B2 (en) | Method for determining availability of a software application using composite hidden Markov model | |
| JP6560630B2 (en) | Network evaluation system, unavailability calculation method, and program | |
| CN106452939A (en) | Method for assessing availability of computing resource of cloud computing system in consideration of redundant fault-tolerant recovery | |
| CN111886605A (en) | Processing for multiple input data sets |
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 |