CN109960575A - A kind of computing capability sharing method, system and relevant device - Google Patents
A kind of computing capability sharing method, system and relevant device Download PDFInfo
- Publication number
- CN109960575A CN109960575A CN201910233203.8A CN201910233203A CN109960575A CN 109960575 A CN109960575 A CN 109960575A CN 201910233203 A CN201910233203 A CN 201910233203A CN 109960575 A CN109960575 A CN 109960575A
- Authority
- CN
- China
- Prior art keywords
- node device
- subtask
- node
- device group
- result
- 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
- 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
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
- G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
-
- 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
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention provides a kind of computing capability sharing method, system and relevant devices, for improving the efficiency and accuracy rate of shared calculating.It include at least M node device for meeting preset condition in each node device group this method comprises: determining the corresponding node device group in N number of subtask of calculating task;It determines optimal node device in each node device group and optimal node device in each node device group is called to complete corresponding subtask, generate subtask result;Using at least (M-1) a node device except optimal node device in each node device group as check-node equipment, the check-node equipment in each node device group is called to verify the subtask result, if quantity accounting passes through up to check-node equipment verification more than preset ratio, determine that corresponding subtask result verification passes through;Complete calculating task result is generated according to all N number of subtask results passed through that verify.
Description
Technical field
The present invention relates to field of cloud computer technology more particularly to a kind of computing capability sharing methods, system and relevant device.
Background technique
Since the computing capability of individual equipment is to be determined according to the hardware resource of its own, and current single equipment is had
Some hardware resources are relatively limited, and for calculating task huge under current big data scene, single equipment be can be provided
Computing capability it is relatively limited, it is difficult to normally execute calculating task.
In order to ensure the normal execution of calculating task, currently employed mode is to calculate energy by the way that multiple equipment is shared
Complete calculating task is split as multiple subtasks, and multiple subtasks is respectively distributed in a node device respectively by power
Processing, is finally combined into complete calculating task result for the subtask result group that each node device is handled.
In existing schemes, multiple subtasks are distributed into single isolated node equipment and are handled respectively, calculating task it is complete
At need multiple nodes all after the completion of can just provide, server it is difficult to ensure that the calculation force value of multiple nodes be it is duplicate, often
The time-consuming of a calculating task is determined by worst node device completely, if the poor node device of present condition will lead to calculating effect out
Rate is low, and secondly single isolated node equipment processing subtask is it is difficult to ensure that the accuracy calculated.In view of this, it is necessary into one
Step improves computational efficiency and accuracy rate.
Summary of the invention
The main purpose of the present invention is to provide a kind of computing capability sharing method, system and relevant devices, for improving
The shared efficiency and accuracy rate calculated.
To achieve the above object, first aspect present invention provides a kind of computing capability sharing method, it may include:
It determines the corresponding node device group in N number of subtask of calculating task, includes to meet in each node device group
At least M node device of preset condition, wherein N is the positive integer greater than 0, and M is the positive integer greater than 1;
It determines optimal node device in each node device group, and calls optimal node device in each node device group complete
At corresponding subtask, subtask result is generated;
At least M-1 node device except optimal node device in each node device group is set as check-node
It is standby, call the check-node equipment in each node device group to verify the subtask result, if quantity accounting is up in advance
If check-node equipment verification more than ratio passes through, determine that corresponding subtask result verification passes through;
Complete calculating task result is generated according to all N number of subtask results passed through that verify.
Optionally as a kind of possible embodiment, in the embodiment of the present invention, the method can also include:
If quantity accounting does not pass through up to check-node equipment verification more than preset ratio, corresponding subtask knot is determined
Fruit is abnormal, and the node device in corresponding node device group is called to recalculate corresponding subtask.
Optionally as a kind of possible embodiment, in the embodiment of the present invention, the determining calculating task is corresponding N number of
The corresponding node device group in subtask includes:
The calculating task that primary server is sent is received, and the calculating task is divided into N number of subtask, and calculate each
The calculation power demand of subtask;
The calculation power demand information sent to the primary server, the power demand information of calculating includes each subtask
Power demand is calculated, so that the primary server is that corresponding node device group is distributed in each subtask;
At least M node device belonging to receiving in the corresponding N number of node device group in N number of subtask that primary server is sent
Nodal information.
Optionally as a kind of possible embodiment, in the embodiment of the present invention, in each node device group of determination
Optimal node device, comprising:
Each node device nodal information is obtained, the nodal information includes node device hardware performance parameter, stablizes
It is one or more in the accuracy parameter of line duration parameters, network bandwidth performance parameter and history calculated result;
According to the nodal information, at least M node device in each node device group uses preset common recognition mechanism
Select optimal node device;
Or, going out the synthesis of at least M node device according to the default weight calculation of the parameters in the nodal information
Weight determines the comprehensive maximum node device of weight as optimal node device.
Optionally as a kind of possible embodiment, in the embodiment of the present invention, the method can be with further include:
The node device for meeting preset condition is picked out from each node device as standby server, current server event
When barrier, current server is replaced using the standby server.
Optionally as a kind of possible embodiment, in the embodiment of the present invention, the node device includes mobile node
Equipment.
Second aspect of the embodiment of the present invention provides a kind of computing capability sharing means, which is characterized in that described device packet
Memory, processor and bus are included, being stored on the memory can be by the bus transfer to the processor and in institute
The computing capability shared routing run on processor is stated, is realized such as when the computing capability shared routing is executed by the processor
Computing capability sharing method described in any one of claim 1 to 5.
Optionally as a kind of possible embodiment, in the embodiment of the present invention, the computing capability sharing means are group
At the node of CDN network or block chain network.
The third aspect of the embodiment of the present invention provides a kind of computing capability shared system, which is characterized in that the system packet
It includes:
First determination unit, for determining the corresponding node device group in the corresponding N number of subtask of calculating task, each
It include at least M node device for meeting preset condition in node device group, wherein N is the positive integer greater than 0, and M is greater than 1
Positive integer;
Second determination unit, for determining optimal node device in each node device group;
First call unit, it is raw for calling optimal node device in each node device group to complete corresponding subtask
At subtask result;
Second call unit makees at least M-1 node device except optimal node device in each node device group
For check-node equipment, the check-node equipment in each node device group is called to verify the subtask result, if
When quantity accounting passes through up to check-node equipment verification more than preset ratio, determine that corresponding subtask result verification passes through;
Generation unit, for generating complete calculating task result according to all N number of subtask results passed through that verify.
Fourth aspect of the embodiment of the present invention provides a kind of computer readable storage medium, which is characterized in that the calculating
Computing capability shared routing is stored on machine readable storage medium storing program for executing, the computing capability shared routing can be by one or more
It manages device to execute, to realize such as the step in first aspect and any possible computing capability sharing method embodiment of first aspect
Suddenly.
In the embodiment of the present invention, N number of node device group can be distributed for the corresponding N number of subtask of each calculating task, by
Optimal node device completes corresponding subtask calculating in each node device group, can be low to avoid the computational efficiency of subtask
Situation, secondly using at least (M-1) a node device except optimal node device in each node device group as check-node
Equipment verifies sub- task computation result, and only quantity accounting passes through up to check-node equipment verification more than preset ratio
When, determine that corresponding subtask result verification passes through, improves the accuracy rate of calculating.
Detailed description of the invention
Fig. 1 is a kind of one embodiment schematic diagram of computing capability sharing method in the embodiment of the present invention;
Fig. 2 is a kind of another embodiment schematic diagram of computing capability sharing method in the embodiment of the present invention;
Fig. 3 is a kind of one embodiment schematic diagram of computing capability shared system in the embodiment of the present invention;
Fig. 4 is a kind of one embodiment schematic diagram of computing capability sharing means in the embodiment of the present invention.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that described herein, specific examples are only used to explain the present invention, not
For limiting the present invention.Based on the embodiments of the present invention, those of ordinary skill in the art are not before making creative work
Every other embodiment obtained is put, shall fall within the protection scope of the present invention.
Description and claims of this specification and term " first ", " second ", " third ", " in above-mentioned attached drawing
The (if present)s such as four " are to be used to distinguish similar objects, without being used to describe a particular order or precedence order.It should manage
The data that solution uses in this way are interchangeable under appropriate circumstances, so that the embodiments described herein can be in addition to illustrating herein
Or the sequence other than the content of description is implemented.In addition, term " includes " and " having " and their any deformation, it is intended that
Cover it is non-exclusive include, for example, containing the process, method, system, product or equipment of a series of steps or units need not limit
In step or unit those of is clearly listed, but may include be not clearly listed or for these process, methods, produce
The other step or units of product or equipment inherently.
It should be noted that the description for being related to " first ", " second " etc. in the present invention is used for description purposes only, and cannot
It is interpreted as its relative importance of indication or suggestion or implicitly indicates the quantity of indicated technical characteristic.Define as a result, " the
One ", the feature of " second " can explicitly or implicitly include at least one of the features.In addition, the skill between each embodiment
Art scheme can be combined with each other, but must be based on can be realized by those of ordinary skill in the art, when technical solution
Will be understood that the combination of this technical solution is not present in conjunction with there is conflicting or cannot achieve when, also not the present invention claims
Protection scope within.
In existing computing capability secret sharing, N number of son is mainly split as by the calculating task that primary server will acquire
Task, each subtask is distributed to multiple nodes and calculated by primary server later, finally by server to each node
Output result is judged and is summarized merging.Applicant have observed that in currently existing scheme, the fractionation of subtask, output result
Judgement and summarize merging all completed by primary server, this is relatively high to the performance requirement of primary server, when there are more meters
When calculation task, the performance of primary server, which probably becomes, restricts its bottleneck for calculating power, it may appear that distribution of computation tasks delay is led
The node device idle waiting for having computing capability is caused, computing resource is wasted.Even if using the individual node except primary server
Equipment carries out subtask operation, it is found by the applicant that the time-consuming of each calculating task is determined by worst node device completely, if going out
It is low that the poor node device of present condition will lead to computational efficiency, secondly single isolated node equipment processing subtask it is difficult to ensure that
The accuracy of calculating.
In inventive embodiments, N number of node device group can be distributed for the corresponding N number of subtask of calculating task, by each
Optimal node device is completed corresponding subtask and is calculated in node device group, feelings that can be low to avoid the computational efficiency of subtask
Condition, secondly using (M-1) a node device except optimal node device in each node device group as check-node equipment pair
Subtask calculated result is verified, and the accuracy rate of calculating is improved.
In order to make it easy to understand, referring to Fig. 1, one embodiment that the present invention provides a kind of computing capability sharing method can wrap
It includes:
101, the corresponding node device group in N number of subtask of calculating task is determined;
In computing capability shared procedure, all node device needs are direct or indirect to be communicated with primary server
Connection, and the nodal information to primary server itself actively or passively.The nodal information may include node device hardware
Can parameter, stablize online hours parameter, one in the accuracy parameter of network bandwidth performance parameter and history calculated result or
It is multinomial.
Optionally, primary server is after the status information for getting each node device, can pick out meet it is default
The node device of condition serves as staging server.For example, can choose memory greater than 4G, free memory is greater than 10G, online
Duration is more than that the node device of threshold value serves as staging server, and specific selection criteria can carry out reasonably according to actual needs
Setting or modification, herein without limitation.
In the embodiment of the present invention, calculating task directly can be divided into N after getting calculating task by primary server
A subtask, the calculating task that can also be will acquire are sent to staging server, so that staging server draws calculating task
It is divided into N number of subtask, the process that specific calculating task divides is herein without limitation.It can carry out according to actual needs herein rationally
Adjustment, specifically herein without limitation.
After calculating task is divided into N number of subtask, primary server or staging server can be according to calculating tasks pair
The calculation power demand for the N number of subtask answered is that corresponding node device group is distributed in each subtask, includes in each node device group
Meet at least M node device of preset condition, wherein N is positive integer greater than 0, and M is the positive integer greater than 1, specific M with
The value of N can be adjusted reasonably according to actual needs.Staging server can also will calculate power demand and be sent to main service
Device is distributed after the corresponding N number of node device group in N number of subtask by primary server and passes to staging server again, specific to determine
The mode of the corresponding node device group in N number of subtask of calculating task is herein without limitation.
102, it determines optimal node device in each node device group, calls optimal node device in each node device group
Corresponding subtask is completed, subtask result is generated;
For saving computing resource, computational efficiency is improved, after distributing corresponding node device group for subtask, main service
Device or staging server are it needs to be determined that optimal node device, the COMPREHENSIVE CALCULATING of the optimal node device are imitated in each node device group
Rate is higher than the average computation efficiency in corresponding node device, and primary server or staging server can call each node device
Optimal node device completes corresponding subtask in group, generates subtask result.It specifically can be according in each node device group
The nodal information of node device, which is compared, determines optimal node device, and specific method of determination can be according to selected node
The content of information is reasonably arranged, specifically herein without limitation.
103, it is saved at least (M-1) a node device except optimal node device in each node device group as verification
Point device calls the check-node equipment in each node device group to verify sub- task result, if quantity accounting is up in advance
If check-node equipment verification more than ratio passes through, determine that corresponding subtask result verification passes through;
In order to ensure the accuracy rate of calculating, primary server or staging server in the embodiment of the present invention can be by each sections
At least (M-1) a node device in point device group except optimal node device calls each node as check-node equipment
Check-node equipment in equipment group verifies sub- task result, if quantity accounting reaches the check-node of preset ratio or more
When equipment verification passes through, determining that corresponding subtask result is verified by, for example, preset ratio can be set is 50%, 60%
Deng.
Optionally, if quantity accounting does not pass through up to check-node equipment verification more than preset ratio, determine corresponding
Subtask results abnormity calls the node device in corresponding node device group to recalculate corresponding subtask.Primary server
Or staging server can call the optimal node device having determined to re-start the calculating of subtask, can also redefine
New optimal node device re-starts the calculating of subtask again, specifically herein without limitation.
104, complete calculating task result is generated according to all N number of subtask results passed through that verify.
After the subtask result verification of each subtask passes through, primary server or staging server can be according to all
It verifies the N number of subtask result passed through to be combined, generates complete calculating task result.
In the embodiment of the present invention, N number of node device group can be distributed for the corresponding N number of subtask of each calculating task, by
Optimal node device completes corresponding subtask calculating in each node device group, can be low to avoid the computational efficiency of subtask
Situation, secondly using at least (M-1) a node device except optimal node device in each node device group as check-node
Equipment verifies sub- task computation result, and only quantity accounting passes through up to check-node equipment verification more than preset ratio
When, determine that corresponding subtask result verification passes through, improves the accuracy rate of calculating.
On the basis of above-mentioned embodiment shown in FIG. 1, by the computing capability sharing method in the embodiment of the present invention into
Row further instruction, referring to Fig. 2, when carrying out the division of calculating task using staging server, one in the embodiment of the present invention
Another embodiment of kind computing capability sharing method can include:
201, the calculating task that primary server is sent is received, and calculating task is divided into N number of subtask, and calculate each
The calculation power demand of subtask;
In computing capability shared procedure, all node device needs are direct or indirect to be communicated with primary server
Connection, and the nodal information to primary server itself actively or passively.The nodal information may include node device hardware
Can parameter, stablize online hours parameter, one in the accuracy parameter of network bandwidth performance parameter and history calculated result or
It is multinomial.
Primary server can pick out the section for meeting preset condition after the status information for getting each node device
Point device serves as staging server.For example, can choose memory greater than 4G, free memory is greater than 10G, and online hours are more than
The node device of threshold value serves as staging server, and specific selection criteria reasonably can be arranged or be repaired according to actual needs
Change, herein without limitation.
In the embodiment of the present invention, the calculating task that primary server can will acquire after getting calculating task is sent
To staging server, so that calculating task is divided into N number of subtask by staging server, and the calculation power of each subtask is determined
Demand.
202, the calculation power demand information sent to primary server, calculating the calculation power that power demand information includes each subtask needs
It asks;
After calculating task is divided into N number of subtask, staging server can determine that the corresponding N number of son of calculating task is appointed
The calculation power demand of business, and the calculation power demand information sent to primary server calculate the calculation power that power demand information includes each subtask
Demand, so that it is that corresponding node is distributed in each subtask that primary server, which is the corresponding node device group of each subtask distribution,
Equipment group, comprising meeting at least M node device of preset condition in each node device group, wherein N is just whole greater than 0
Number, M are positive integer greater than 1, and the value of specific M and N can be adjusted reasonably according to actual needs, specifically herein not
It limits.
203, at least M node belonging to receiving in the corresponding N number of node device group in N number of subtask that primary server is sent
The nodal information of equipment;
In practice, the case where failure delay machine is likely to occur due to node device, in order to ensure the stabilization of calculating process
Property, it can be that the distribution of each subtask is corresponding according to power demand information is calculated after primary server receives calculation power demand information
Node device group, the node device group kind can be with M node devices, and by the section of node device belonging to each node device group
Point information is sent to corresponding staging server.Wherein, M is the positive integer greater than 1, and specific value can be according to actual needs
It is reasonably adjusted, specifically herein without limitation.
204, it determines optimal node device in each node device group, calls optimal node device in each node device group
Corresponding subtask is completed, subtask result is generated;
To save computing resource, computational efficiency is improved, after distributing corresponding node device group for subtask, is temporarily taken
Device be engaged in it needs to be determined that optimal node device, the COMPREHENSIVE CALCULATING efficiency of the optimal node device are higher than correspondence in each node device group
Node device in average computation efficiency, primary server or staging server can call optimal section in each node device group
Point device completes corresponding subtask, generates subtask result.It specifically can be according to each node device group interior joint equipment
Nodal information, which is compared, determines optimal node device, and specific method of determination can be according to the content of selected nodal information
Reasonably it is arranged, specifically herein without limitation.
Optionally, as a kind of possible embodiment, it is determined in the embodiment of the present invention optimal in each node device group
The process of node device may include:
Each node device nodal information is obtained, when nodal information includes node device hardware performance parameter, stablizes online
It is one or more in the accuracy parameter of long parameter, network bandwidth performance parameter and history calculated result;According to nodal information,
M node device in each node device group selects optimal node device using preset common recognition mechanism.
Specifically, it can be based on preset common recognition mechanism after getting preset node device nodal information, such as
It is compared between the corresponding M node device in each subtask, recommends an optimal node out to carry out the calculating of task jointly, examine
When the standard of amount can be the hardware performance best (memory size, memory space, CPU operational efficiency etc.) of mobile phone, stablize online
Between longest, network bandwidth highest be most stable, accuracy of history calculated result, specific common recognition process is referred to existing implementation
Mode, herein without limitation.
Optionally, as a kind of possible embodiment, it is determined in the embodiment of the present invention optimal in each node device group
The process of node device may include:
Each node device nodal information is obtained, when nodal information includes node device hardware performance parameter, stablizes online
It is one or more in the accuracy parameter of long parameter, network bandwidth performance parameter and history calculated result;According to nodal information
In the default weight calculations of parameters go out the synthesis weight of M node device, determine the comprehensive maximum node device of weight
As optimal node device.
Specifically, staging server can be according to nodal information after getting preset node device nodal information
The default weight of the parameters of China carries out comprehensive weight computing, determines the comprehensive maximum node device of weight as optimal section
Point device.
205, it is saved at least (M-1) a node device except optimal node device in each node device group as verification
Point device calls the check-node equipment in each node device group to verify sub- task result, if quantity accounting is up in advance
If check-node equipment verification more than ratio passes through, determine that corresponding subtask result verification passes through;
In order to ensure the accuracy rate of calculating, primary server or staging server in the embodiment of the present invention can be by each sections
(M-1) a node device in point device group except optimal node device calls each node device as check-node equipment
Check-node equipment in group verifies sub- task result, if quantity accounting reaches the check-node equipment of preset ratio or more
When verification passes through, determine that corresponding subtask result verification passes through.
In practice, although being verified in the present invention to the calculated result of subtask, the same calculating task
Calculated result checking procedure needed for calculation amount be generally less than the task and normally calculate required calculation amount, overall calculating
Efficiency is still enhanced.Mine calculating is dug for example, digging calculation amount needed for the checking procedure of hash value during mine and being often much smaller than
Process.
Optionally, if quantity accounting does not pass through up to check-node equipment verification more than preset ratio, determine corresponding
Subtask results abnormity calls the node device in corresponding node device group to recalculate corresponding subtask.Primary server
Or staging server can call the optimal node device having determined to re-start the calculating of subtask, can also redefine
New optimal node device re-starts the calculating of subtask again, specifically herein without limitation.
206, complete calculating task result is generated according to all N number of subtask results passed through that verify.
In the embodiment of the present invention, N number of node device group can be distributed for the corresponding N number of subtask of each calculating task, by
Optimal node device completes corresponding subtask calculating in each node device group, can be low to avoid the computational efficiency of subtask
Situation, secondly using at least (M-1) a node device except optimal node device in each node device group as check-node
Equipment verifies sub- task computation result, and only quantity accounting passes through up to check-node equipment verification more than preset ratio
When, determine that corresponding subtask result verification passes through, improves the accuracy rate of calculating.Secondly, being calculated without primary server
The division of task and the processing of subtask result, greatly reduce the resource occupation of primary server, further improve shared meter
Calculate efficiency.
In addition, as a preferred embodiment, the node device as staging server includes mobile node equipment.
Mobile node equipment in present embodiment can be specially that mobile phone, laptop, tablet computer etc. have meter
Calculation ability and the equipment with portable mobile.It is widely available in view of current mobile device, there is biggish ownership, and
And mobile device often has certain computing capability, therefore the computing capability that whole mobile device can be provided is relatively
Height, but since in actual use, user causes the vacancy rate of mobile device general the high frequency of replacement of mobile device
All over higher, the waste of available resources often will cause, therefore the use cost of mobile device is relatively low, present embodiment will move
Dynamic node device of the equipment as subtasking can improve the utilization rate to idle available resources relatively, and also can
It reduces computing capability and shares whole executory cost.
In addition, on the basis of a series of above-mentioned embodiments, as a preferred embodiment, node device group
Total amount is greater than N.
It should be noted that due to consideration that in practical applications, the node device in node device group is likely to occur event
The case where hindering delay machine, the availability of the node device group is likely to occur biggish reduction in the case, in order to guarantee for locating
Manage the node device group of subtask can availability with higher, the total amount of the node device group in present embodiment is more than
The whole total amount of subtask, therefore even if a certain node device group is unavailable because of node device delay machine, it also can be by other sections
Point device group substitutes the not available node device group subtasking, and then the opposite entirety for ensuring subtasking is reliable
Property.
On the basis of above-mentioned Fig. 1 and embodiment shown in Fig. 2, primary server or staging server can also be from each nodes
The node device for meeting preset condition is picked out in equipment as standby server, when staging server failure, use is spare
Server replaces staging server, ensures the stability that calculating task is completed.
It is understood that in various embodiments of the present invention, the size of the serial number of above steps is not meant to
Execution sequence it is successive, the execution of each step sequence should be determined by its function and internal logic, without coping with the embodiment of the present invention
Implementation process constitute any restriction.
The computing capability sharing method in the embodiment of the present invention is described in above-described embodiment, referring to Fig. 3, this hair
A kind of one embodiment of computing capability shared system in bright embodiment can include:
First determination unit 301, for determining the corresponding node device group in the corresponding N number of subtask of calculating task,
It include at least M node device for meeting preset condition in each node device group, wherein N is the positive integer greater than 0, and M is big
In 1 positive integer;
Second determination unit 302, for determining optimal node device in each node device group;
First call unit 303, for calling optimal node device in each node device group to complete corresponding subtask,
Generate subtask result;
Second call unit 304 sets at least (M-1) a node except optimal node device in each node device group
It is standby to be used as check-node equipment, call the check-node equipment in each node device group to verify sub- task result, if
When quantity accounting passes through up to check-node equipment verification more than preset ratio, determine that corresponding subtask result verification passes through;
Generation unit 305, for generating complete calculating task knot according to all N number of subtask results passed through that verify
Fruit.
In the embodiment of the present invention, N number of node device group can be distributed for the corresponding N number of subtask of each calculating task, by
Optimal node device completes corresponding subtask calculating in each node device group, can be low to avoid the computational efficiency of subtask
Situation, secondly using at least (M-1) a node device except optimal node device in each node device group as check-node
Equipment verifies sub- task computation result, and only quantity accounting passes through up to check-node equipment verification more than preset ratio
When, determine that corresponding subtask result verification passes through, improves the accuracy rate of calculating.
Be described in detail above for the embodiment of computing capability sharing method, the present invention also provides it is a kind of with
The corresponding computing capability sharing means of this method, the embodiment of the device part are corresponded to each other with the embodiment of method part.
Referring to FIG. 4, Fig. 4 is a kind of structure chart of computing capability sharing means provided in an embodiment of the present invention.
Computing capability sharing means 1 provided in an embodiment of the present invention, including memory 11, processor 12 and bus 13, are deposited
The computing capability shared routing that can be transmitted to processor 12 by bus 13 and run on the processor 12 is stored on reservoir 11,
The operation content such as above-mentioned computing capability sharing method is realized when computing capability shared routing is executed by processor 12.
Specifically, the processor in the present invention can be used to implement following steps:
It determines the corresponding node device group in N number of subtask of calculating task, includes to meet in each node device group
At least M node device of preset condition, wherein N is the positive integer greater than 0, and M is the positive integer greater than 1;
Determine optimal node device in each node device group;
It calls optimal node device in each node device group to complete corresponding subtask, generates subtask result;
At least (M-1) a node device except optimal node device in each node device group is set as check-node
It is standby, call the check-node equipment in each node device group to verify sub- task result, if quantity accounting compares up to default
Or more check-node equipment verification when passing through, determine that corresponding subtask result verification passes through;
Complete calculating task result is generated according to all N number of subtask results passed through that verify.
Optionally, in some embodiments of the present invention, processor can be also used for realizing following steps:
If quantity accounting does not pass through up to check-node equipment verification more than preset ratio, corresponding subtask knot is determined
Fruit is abnormal, and the node device in corresponding node device group is called to recalculate corresponding subtask.
Optionally, in some embodiments of the present invention, processor can be also used for realizing following steps:
The calculating task that primary server is sent is received, and calculating task is divided into N number of subtask, and calculates every height and appoints
The calculation power demand of business;
The calculation power demand information sent to primary server calculates the calculation power demand that power demand information includes each subtask, with
So that primary server is that corresponding node device group is distributed in each subtask;
The section of M node device belonging to receiving in the corresponding N number of node device group in N number of subtask that primary server is sent
Point information.
Optionally, in some embodiments of the present invention, processor can be also used for realizing following steps:
Each node device nodal information is obtained, when nodal information includes node device hardware performance parameter, stablizes online
It is one or more in the accuracy parameter of long parameter, network bandwidth performance parameter and history calculated result;
According to nodal information, M node device in each node device group is selected optimal using preset common recognition mechanism
Node device;
Or, going out the synthesis weight of M node device according to the default weight calculation of the parameters in nodal information, determine
The comprehensive maximum node device of weight is as optimal node device.
Optionally, in some embodiments of the present invention, processor can be also used for realizing following steps:
The node device for meeting preset condition is picked out from each node device as standby server, current server event
When barrier, current server is replaced using standby server.
Optionally, in some embodiments of the present invention, the total amount of node device group is greater than N.
The computing capability sharing means 1 can be the node of composition CDN network or block chain network.
Wherein, memory 11 includes at least a type of readable storage medium storing program for executing, and readable storage medium storing program for executing includes flash memory, hard
Disk, multimedia card, card-type memory (for example, SD or DX memory etc.), magnetic storage, disk, CD etc..Memory 11 exists
It can be the internal storage unit of client, such as the hard disk of the client in some embodiments.Memory 11 is in other realities
The External memory equipment for being also possible to client in example is applied, such as the plug-in type hard disk being equipped in client, intelligent memory card
(Smart Media Card, SMC), secure digital (Secure Digital, SD) card, flash card (Flash Card) etc..Into
One step, memory 11 can also both internal storage units including client or including External memory equipment.Memory 11 is not
It can be only used for storage and be installed on the application software and Various types of data of client, such as the data access based on network library
Code etc. can be also used for temporarily storing the data that has exported or will export.
Processor 12 can be in some embodiments a central processing unit (Central Processing Unit,
CPU), controller, microcontroller, microprocessor or other data processing chips, the program for being stored in run memory 11
Code or processing data, such as execute the data access etc. based on network library.
The bus 13 can be Peripheral Component Interconnect standard (peripheral component interconnect, abbreviation
PCI) bus or expanding the industrial standard structure (extended industry standard architecture, abbreviation EISA)
Bus etc..The bus can be divided into address bus, data/address bus, control bus etc..For convenient for indicating, in Fig. 4 only with one slightly
Line indicates, it is not intended that an only bus or a type of bus.
Further, computing capability sharing means can also include network interface 14, and network interface 14 optionally can wrap
Wireline interface and/or wireless interface (such as WI-FI interface, blue tooth interface) are included, commonly used in setting in the device 1 with other electronics
Communication connection is established between standby.
Optionally, which can also include user interface, and user interface may include display
(Display), input unit such as keyboard (Keyboard), optional user interface can also include standard wireline interface,
Wireless interface.Optionally, in some embodiments, it is aobvious to can be light-emitting diode display, liquid crystal display, touch control type LCD for display
Show that device and OLED (Organic Light-Emitting Diode, Organic Light Emitting Diode) touch device etc..Wherein, display
Can also it is appropriate be known as display screen or display unit, for be shown in the information handled in computing capability sharing means 1 and
For showing visual user interface.
It will be appreciated by persons skilled in the art that the structure shown in Fig. 4 is not constituted to computing capability sharing means 1
It limits, may include perhaps combining certain components or different component layouts than illustrating less perhaps more components.
The present invention also provides a kind of computer readable storage medium, calculating is stored on the computer readable storage medium
When computer program is executed by processor, following steps are may be implemented in machine program:
It determines the corresponding node device group in N number of subtask of calculating task, includes to meet in each node device group
At least M node device of preset condition, wherein N is the positive integer greater than 0, and M is the positive integer greater than 1;
Determine optimal node device in each node device group;
It calls optimal node device in each node device group to complete corresponding subtask, generates subtask result;
At least (M-1) a node device except optimal node device in each node device group is set as check-node
It is standby, call the check-node equipment in each node device group to verify sub- task result, if quantity accounting compares up to default
Or more check-node equipment verification when passing through, determine that corresponding subtask result verification passes through;
Complete calculating task result is generated according to all N number of subtask results passed through that verify.
Optionally, in some embodiments of the present invention, processor can be also used for realizing following steps:
If quantity accounting does not pass through up to check-node equipment verification more than preset ratio, corresponding subtask knot is determined
Fruit is abnormal, and the node device in corresponding node device group is called to recalculate corresponding subtask.
Optionally, in some embodiments of the present invention, processor can be also used for realizing following steps:
The calculating task that primary server is sent is received, and calculating task is divided into N number of subtask, and calculates every height and appoints
The calculation power demand of business;
The calculation power demand information sent to primary server calculates the calculation power demand that power demand information includes each subtask, with
So that primary server is that corresponding node device group is distributed in each subtask;
At least M node device belonging to receiving in the corresponding N number of node device group in N number of subtask that primary server is sent
Nodal information.
Optionally, in some embodiments of the present invention, processor can be also used for realizing following steps:
Each node device nodal information is obtained, when nodal information includes node device hardware performance parameter, stablizes online
It is one or more in the accuracy parameter of long parameter, network bandwidth performance parameter and history calculated result;
According to nodal information, at least M node device in each node device group is selected using preset common recognition mechanism
Optimal node device;
Or, go out the synthesis weight of at least M node device according to the default weight calculation of the parameters in nodal information,
Determine the comprehensive maximum node device of weight as optimal node device.
Optionally, in some embodiments of the present invention, processor can be also used for realizing following steps:
The node device for meeting preset condition is picked out from each node device as standby server, current server event
When barrier, current server is replaced using standby server.
Optionally, in some embodiments of the present invention, the total amount of node device group is greater than N.
In the above-described embodiments, can come wholly or partly by software, hardware, firmware or any combination thereof real
It is existing.When implemented in software, it can entirely or partly realize in the form of a computer program product.
The computer program product includes one or more computer instructions.Load and execute on computers the meter
When calculation machine program instruction, entirely or partly generate according to process or function described in the embodiment of the present invention.The computer can
To be general purpose computer, special purpose computer, computer network or other programmable devices.The computer instruction can be deposited
Storage in a computer-readable storage medium, or from a computer readable storage medium to another computer readable storage medium
Transmission, for example, the computer instruction can pass through wired (example from a web-site, computer, server or data center
Such as coaxial cable, optical fiber, Digital Subscriber Line (DSL)) or wireless (such as infrared, wireless, microwave) mode to another website
Website, computer, server or data center are transmitted.The computer readable storage medium can be computer and can deposit
Any usable medium of storage either includes that the data storages such as one or more usable mediums integrated server, data center are set
It is standby.The usable medium can be magnetic medium, (for example, floppy disk, hard disk, tape), optical medium (for example, DVD) or partly lead
Body medium (such as solid state hard disk Solid State Disk (SSD)) etc.
It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description,
The specific work process of device and unit, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
In several embodiments provided by the present invention, it should be understood that disclosed system, device and method can be with
It realizes by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the unit
It divides, only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units or components
It can be combined or can be integrated into another system, or some features can be ignored or not executed.Another point, it is shown or
The mutual coupling, direct-coupling or communication connection discussed can be through some interfaces, the indirect coupling of device or unit
It closes or communicates to connect, can be electrical property, mechanical or other forms.
The above is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair
Equivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skills
Art field, is included within the scope of the present invention.
Claims (10)
1. a kind of computing capability sharing method characterized by comprising
Determine the corresponding node device group in N number of subtask of calculating task, it is default comprising meeting in each node device group
At least M node device of condition, wherein N is the positive integer greater than 0, and M is the positive integer greater than 1;
It determines optimal node device in each node device group, and calls optimal node device completion pair in each node device group
The subtask answered generates subtask result;
Using at least M-1 node device except optimal node device in each node device group as check-node equipment, adjust
The subtask result is verified with the check-node equipment in each node device group, if quantity accounting reaches preset ratio
When above check-node equipment verification passes through, determine that corresponding subtask result verification passes through;
Complete calculating task result is generated according to all N number of subtask results passed through that verify.
2. the method according to claim 1, wherein the method also includes:
If quantity accounting does not pass through up to check-node equipment verification more than preset ratio, determine that corresponding subtask result is different
Often, the node device in corresponding node device group is called to recalculate corresponding subtask.
3. according to the method described in claim 2, it is characterized in that, the corresponding N number of subtask difference of the determining calculating task
Corresponding node device group includes:
The calculating task that primary server is sent is received, and the calculating task is divided into N number of subtask, and calculates every height and appoints
The calculation power demand of business;
The calculation power demand information sent to the primary server, it is described to calculate the calculation power that power demand information includes each subtask
Demand, so that the primary server is that corresponding node device group is distributed in each subtask;
The section of at least M node device belonging to receiving in the corresponding N number of node device group in N number of subtask that primary server is sent
Point information.
4. according to the method in any one of claims 1 to 3, which is characterized in that in each node device group of determination
Optimal node device, comprising:
Each node device nodal information is obtained, when the nodal information includes node device hardware performance parameter, stablizes online
It is one or more in the accuracy parameter of long parameter, network bandwidth performance parameter and history calculated result;
According to the nodal information, at least M node device in each node device group is selected using preset common recognition mechanism
Optimal node device;
Or, go out the synthesis weight of at least M node device according to the default weight calculation of the parameters in the nodal information,
Determine the comprehensive maximum node device of weight as optimal node device.
5. according to the method described in claim 4, it is characterized by further comprising:
The node device for meeting preset condition is picked out from each node device as standby server, current server failure
When, current server is replaced using the standby server.
6. according to the method described in claim 5, it is characterized in that, the node device includes mobile node equipment.
7. a kind of computing capability sharing means, which is characterized in that described device includes memory, processor and bus, described to deposit
It is shared that the computing capability that can be run to the processor and on the processor by the bus transfer is stored on reservoir
Program realizes such as calculating described in any one of claim 1 to 5 when the computing capability shared routing is executed by the processor
Ability sharing method.
8. device according to claim 7, which is characterized in that described device is composition CDN network or block chain network
Node.
9. a kind of computing capability shared system, which is characterized in that the system comprises:
First determination unit, for determining the corresponding node device group in the corresponding N number of subtask of calculating task, each node
It include at least M node device for meeting preset condition in equipment group, wherein N is the positive integer greater than 0, and M is just whole greater than 1
Number;
Second determination unit, for determining optimal node device in each node device group;
First call unit generates son for calling optimal node device in each node device group to complete corresponding subtask
Task result;
Second call unit, using at least M-1 node device except optimal node device in each node device group as school
Node device is tested, calls the check-node equipment in each node device group to verify the subtask result, if quantity
When accounting passes through up to check-node equipment verification more than preset ratio, determine that corresponding subtask result verification passes through;
Generation unit, for generating complete calculating task result according to all N number of subtask results passed through that verify.
10. a kind of computer readable storage medium, which is characterized in that be stored with calculating energy on the computer readable storage medium
Power shared routing, the computing capability shared routing can be executed by one or more processor, with realize as claim 1 to
Computing capability sharing method described in any one of 6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910233203.8A CN109960575B (en) | 2019-03-26 | 2019-03-26 | Computing capacity sharing method, system and related equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910233203.8A CN109960575B (en) | 2019-03-26 | 2019-03-26 | Computing capacity sharing method, system and related equipment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109960575A true CN109960575A (en) | 2019-07-02 |
CN109960575B CN109960575B (en) | 2023-09-15 |
Family
ID=67024960
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910233203.8A Active CN109960575B (en) | 2019-03-26 | 2019-03-26 | Computing capacity sharing method, system and related equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109960575B (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110995614A (en) * | 2019-11-05 | 2020-04-10 | 华为技术有限公司 | Computing power resource allocation method and device |
CN111432159A (en) * | 2020-03-19 | 2020-07-17 | 深圳市鹏创软件有限公司 | Computing task processing method, device and system and computer readable storage medium |
CN111444010A (en) * | 2020-01-19 | 2020-07-24 | 南京邮电大学 | Consensus method based on computing resource computing power certification |
CN112261141A (en) * | 2020-10-23 | 2021-01-22 | 移康智能科技(上海)股份有限公司 | Local Internet of things computing power detection system based on block chain technology |
CN113542316A (en) * | 2020-04-13 | 2021-10-22 | 展讯半导体(南京)有限公司 | Computing power sharing method and related equipment |
CN113535343A (en) * | 2020-04-15 | 2021-10-22 | 展讯半导体(南京)有限公司 | Computing power sharing method based on network scheduling and related product |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2013077787A1 (en) * | 2011-11-23 | 2013-05-30 | Telefonaktiebolaget Lm Ericsson (Publ) | Method and apparatus for distributed processing tasks |
CN103701894A (en) * | 2013-12-25 | 2014-04-02 | 浙江省公众信息产业有限公司 | Method and system for dispatching dynamic resource |
CN103970611A (en) * | 2014-05-05 | 2014-08-06 | 神华集团有限责任公司 | Task processing method based on computer cluster |
EP2819011A2 (en) * | 2013-06-27 | 2014-12-31 | Tata Consultancy Services Limited | Task execution by idle resources in grid computing system |
CN108268318A (en) * | 2016-12-30 | 2018-07-10 | 华为技术有限公司 | A kind of method and apparatus of distributed system task distribution |
CN108491263A (en) * | 2018-03-02 | 2018-09-04 | 珠海市魅族科技有限公司 | Data processing method, data processing equipment, terminal and readable storage medium storing program for executing |
CN108540568A (en) * | 2018-04-23 | 2018-09-14 | 移康智能科技(上海)股份有限公司 | Computing capability sharing method and smart machine |
CN108958935A (en) * | 2018-06-28 | 2018-12-07 | 阿瑞思科技(成都)有限责任公司 | The method and its system of distributed computing are realized based on mobile terminal |
CN109298946A (en) * | 2018-10-19 | 2019-02-01 | 上海达家迎信息科技有限公司 | Digging mine method, cloud platform, device, equipment and medium based on block chain |
-
2019
- 2019-03-26 CN CN201910233203.8A patent/CN109960575B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2013077787A1 (en) * | 2011-11-23 | 2013-05-30 | Telefonaktiebolaget Lm Ericsson (Publ) | Method and apparatus for distributed processing tasks |
EP2819011A2 (en) * | 2013-06-27 | 2014-12-31 | Tata Consultancy Services Limited | Task execution by idle resources in grid computing system |
CN103701894A (en) * | 2013-12-25 | 2014-04-02 | 浙江省公众信息产业有限公司 | Method and system for dispatching dynamic resource |
CN103970611A (en) * | 2014-05-05 | 2014-08-06 | 神华集团有限责任公司 | Task processing method based on computer cluster |
CN108268318A (en) * | 2016-12-30 | 2018-07-10 | 华为技术有限公司 | A kind of method and apparatus of distributed system task distribution |
CN108491263A (en) * | 2018-03-02 | 2018-09-04 | 珠海市魅族科技有限公司 | Data processing method, data processing equipment, terminal and readable storage medium storing program for executing |
CN108540568A (en) * | 2018-04-23 | 2018-09-14 | 移康智能科技(上海)股份有限公司 | Computing capability sharing method and smart machine |
CN108958935A (en) * | 2018-06-28 | 2018-12-07 | 阿瑞思科技(成都)有限责任公司 | The method and its system of distributed computing are realized based on mobile terminal |
CN109298946A (en) * | 2018-10-19 | 2019-02-01 | 上海达家迎信息科技有限公司 | Digging mine method, cloud platform, device, equipment and medium based on block chain |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110995614A (en) * | 2019-11-05 | 2020-04-10 | 华为技术有限公司 | Computing power resource allocation method and device |
CN110995614B (en) * | 2019-11-05 | 2022-04-05 | 华为技术有限公司 | Computing power resource allocation method and device |
CN111444010A (en) * | 2020-01-19 | 2020-07-24 | 南京邮电大学 | Consensus method based on computing resource computing power certification |
CN111444010B (en) * | 2020-01-19 | 2022-10-14 | 南京邮电大学 | Consensus method based on computing resource computing power certification |
CN111432159A (en) * | 2020-03-19 | 2020-07-17 | 深圳市鹏创软件有限公司 | Computing task processing method, device and system and computer readable storage medium |
CN111432159B (en) * | 2020-03-19 | 2022-05-17 | 深圳市鹏创软件有限公司 | Computing task processing method, device and system and computer readable storage medium |
CN113542316A (en) * | 2020-04-13 | 2021-10-22 | 展讯半导体(南京)有限公司 | Computing power sharing method and related equipment |
CN113535343A (en) * | 2020-04-15 | 2021-10-22 | 展讯半导体(南京)有限公司 | Computing power sharing method based on network scheduling and related product |
CN112261141A (en) * | 2020-10-23 | 2021-01-22 | 移康智能科技(上海)股份有限公司 | Local Internet of things computing power detection system based on block chain technology |
Also Published As
Publication number | Publication date |
---|---|
CN109960575B (en) | 2023-09-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109873868A (en) | A kind of computing capability sharing method, system and relevant device | |
CN109960575A (en) | A kind of computing capability sharing method, system and relevant device | |
US10733026B2 (en) | Automated workflow selection | |
Rahman et al. | Adaptive workflow scheduling for dynamic grid and cloud computing environment | |
US8881168B2 (en) | Scheduling a parallel job in a system of virtual containers | |
US9805140B2 (en) | Striping of directed graphs and nodes with improved functionality | |
US9703610B2 (en) | Extensible centralized dynamic resource distribution in a clustered data grid | |
US11966792B2 (en) | Resource processing method of cloud platform, related device, and storage medium | |
US20110191781A1 (en) | Resources management in distributed computing environment | |
CN107645407B (en) | Method and device for adapting QoS | |
CN110162388A (en) | A kind of method for scheduling task, system and terminal device | |
US20150133214A1 (en) | Video encoding based on areas of interest | |
CN108415772B (en) | Container-based resource adjustment method, device and medium | |
US9471387B2 (en) | Scheduling in job execution | |
CN109618174A (en) | A kind of live data transmission method, device, system and storage medium | |
JP2013186770A (en) | Data processing device | |
CN112600761A (en) | Resource allocation method, device and storage medium | |
CN115033340A (en) | Host selection method and related device | |
CN114356547A (en) | Low-priority blocking method and device based on processor virtualization environment | |
CN107872405A (en) | Distributed bandwidth allocation and regulation | |
CN109002286A (en) | Data asynchronous processing method and device based on synchronous programming | |
CN113347238A (en) | Message partitioning method, system, device and storage medium based on block chain | |
CN108964904A (en) | Group cipher method for managing security, device, electronic equipment and storage medium | |
Bessai et al. | Business process scheduling strategies in cloud environments with fairness metrics | |
CN114090247A (en) | Method, device, equipment and storage medium for processing data |
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 |