CN110489224A - A kind of method and apparatus of task schedule - Google Patents
A kind of method and apparatus of task schedule Download PDFInfo
- Publication number
- CN110489224A CN110489224A CN201810460659.3A CN201810460659A CN110489224A CN 110489224 A CN110489224 A CN 110489224A CN 201810460659 A CN201810460659 A CN 201810460659A CN 110489224 A CN110489224 A CN 110489224A
- Authority
- CN
- China
- Prior art keywords
- task
- performer
- value
- task performer
- amount
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000004364 calculation method Methods 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 7
- 230000005540 biological transmission Effects 0.000 claims description 2
- 238000012216 screening Methods 0.000 claims 2
- 238000012545 processing Methods 0.000 abstract description 37
- 238000011156 evaluation Methods 0.000 abstract description 5
- 238000005516 engineering process Methods 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 15
- 230000006870 function Effects 0.000 description 12
- 230000008569 process Effects 0.000 description 7
- 230000002159 abnormal effect Effects 0.000 description 6
- 230000006854 communication Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 6
- 238000004422 calculation algorithm Methods 0.000 description 5
- 238000004891 communication Methods 0.000 description 5
- 235000013399 edible fruits Nutrition 0.000 description 4
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 241000219793 Trifolium Species 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000005291 magnetic effect Effects 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 241000196324 Embryophyta Species 0.000 description 1
- 241000490229 Eucephalus Species 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 210000003127 knee Anatomy 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000004321 preservation Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
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
- G06F9/50—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
- G06F9/5038—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 considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration
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 invention discloses a kind of method and apparatus of task schedule, are related to field of computer technology.One specific embodiment of this method comprises determining that task performer, sends workload computing and instructs to each task performer；The end value that each task performer is fed back is received, the time needed for determining each task performer feedback result value, to calculate the working efficiency of each task performer；It determines task amount to be processed, according to the working efficiency of each task performer, carries out task distribution.The embodiment is by way of proof of work, the processing capacity of each task performer of accurate evaluation, and according to its processing capacity and task amount to be processed, realizes task smart allocation, accelerate the processing speed of task.
Description
Technical field
The present invention relates to field of computer technology more particularly to a kind of method and apparatus of task schedule.
Background technique
In existing distributed task dispatching system, such as TBSchedule, Clover, realize mean allocation task
Function, and can detecte currently valid actuator quantity, to guarantee that task can be distributed correctly.For example, TBSchedule
Actuator and each actuator task to be dealt with are distinguished by QUEUEID, can guarantee each actuator liberally
Obtain task as much.
In the implementation of the present invention, at least there are the following problems for inventor's discovery prior art:
1) actuator for distributing task at present, despite the machine by artificial distributing uniform performance, but since network is trembled
The performance of the reasons such as dynamic, task quantity, each machine can not be just the same；
2) state and processing capacity that the prior art can not be current according to each actuator carry out smart allocation, cause
For possessing the actuator of higher hardware configuration or free time, the task as many with other actuators can only be obtained, and nothing
Method makes full use of its processing capacity, results in waste of resources.
Summary of the invention
In view of this, the embodiment of the present invention provides a kind of method and apparatus of task schedule, it is at least able to solve existing skill
The process performance that each actuator is not accounted in art leads to the feelings of " high-performance machine idle, low performance machine processing are endless "
Condition.
To achieve the above object, according to an aspect of an embodiment of the present invention, a kind of method of task schedule is provided, is wrapped
It includes: determining task performer, send workload computing and instruct to each task performer；Each task performer is received to be fed back
End value, the time needed for determining each task performer feedback result value, with calculate each task performer work effect
Rate；Wherein, end value is corresponding to workload computing instruction；Task amount to be processed is determined, according to the work of each task performer
Make efficiency, carries out task amount distribution.
Optionally, the time needed for determining each task performer feedback result value, to calculate each task performer
Time needed for working efficiency comprises determining that each task performer feedback result value counts all task performer feedback knots
The total time of fruit value calculate the proof of work value of each task performer；Wherein, proof of work value=all task executions
The total time of device/individual task actuator time；According to the proof of work value of each task performer, all tasks are determined
The amount of work proof value of actuator, calculates the working efficiency of each task performer；Wherein, working efficiency=individual task is held
The proof of work value of row device/all task performers amount of work proof value.
Optionally, the end value that each task performer is fed back is received, determines each task performer feedback result value
The required time further include: the end value that each task performer is fed back is received, based on scheduled proof rule to being fed back
End value verified, reject end value and do not meet the task performer of proof rule；And determine each task performer
Time needed for feedback result value screens the identified time based on scheduled time threshold value, rejects the time beyond predetermined
The task performer of time threshold.
Optionally, it determines task amount to be processed, according to the working efficiency of each task performer, carries out task amount distribution
Include: to receive newly assigned task amount, obtains each task performer currently untreated task amount, obtain task to be processed
Amount；According to task amount to be processed, the working efficiency and current untreated task amount of each task performer, new point is determined
It is distributed in the task amount matched to the task amount of each task performer.
Optionally, sending workload computing and instructing to each task performer includes: the hair when reaching the scheduled update period
Workload computing is sent to instruct to each task performer.
To achieve the above object, according to another aspect of an embodiment of the present invention, a kind of device of task schedule is provided, is wrapped
Include: instruction sending module sends workload computing and instructs to each task performer for determining task performer；Efficiency meter
Module is calculated, the end value fed back for receiving each task performer determines needed for each task performer feedback result value
Time, to calculate the working efficiency of each task performer；Wherein, end value is corresponding to workload computing instruction；Task point
With module, for determining task amount to be processed, according to the working efficiency of each task performer, task amount distribution is carried out.
Optionally, efficiency calculation module is used for: the time needed for determining each task performer feedback result value, statistics
The total time of all task performer feedback result values calculate the proof of work value of each task performer；Wherein, workload
Proof value=all task performers total time/individual task actuator time；According to the work of each task performer
Proof value is measured, the amount of work proof value of all task performers is determined, calculates the working efficiency of each task performer；Its
In, working efficiency=individual task actuator proof of work value/all task performers amount of work proof value.
Optionally, efficiency calculation module is used for: the end value that each task performer is fed back is received, based on scheduled
Proof rule verifies the end value fed back, and rejects the task performer that end value does not meet proof rule；And really
Time needed for fixed each task performer feedback result value screens institute the received time based on scheduled time threshold value,
Reject task performer of the time beyond scheduled time threshold value.
Optionally, task allocating module is used for: being received newly assigned task amount, is obtained each task performer currently not
The task amount of processing obtains task amount to be processed；According to the working efficiency of task amount to be processed, each task performer with
And current untreated task amount, it determines and is distributed in newly assigned task amount to the task amount of each task performer.
Optionally, instruction sending module is used for: when reaching the scheduled update period, being sent workload computing and is instructed to every
A task performer.
To achieve the above object, according to an embodiment of the present invention in another aspect, the electronics for providing a kind of task schedule is set
It is standby.
The electronic equipment of the embodiment of the present invention includes: one or more processors；Storage device, for storing one or more
A program, when one or more of programs are executed by one or more of processors, so that one or more of processing
The method that device realizes any of the above-described task schedule.
To achieve the above object, according to an embodiment of the present invention in another aspect, provide a kind of computer-readable medium,
On be stored with computer program, the method for any of the above-described task schedule is realized when described program is executed by processor.
The scheme of the offer according to the present invention, one embodiment in foregoing invention have the following advantages that or beneficial to effects
Fruit: by way of proof of work, the processing capacity of each task performer of accurate evaluation, and according to its processing capacity and
Task amount to be processed realizes that task smart allocation achievees the purpose that task is quickly handled with realization " able people should do more work ".
Further effect possessed by above-mentioned non-usual optional way adds hereinafter in conjunction with specific embodiment
With explanation.
Detailed description of the invention
Attached drawing for a better understanding of the present invention, does not constitute an undue limitation on the present invention.Wherein:
Fig. 1 is a kind of main flow schematic diagram of the method for task schedule according to an embodiment of the present invention；
Fig. 2 is the distribution schematic diagram of Master host node and task performer under zookeeper；
A kind of Fig. 3 flow diagram of the method for optional task schedule according to an embodiment of the present invention；
Fig. 4 is the flow diagram of the method for another optional task schedule according to an embodiment of the present invention；
Fig. 5 is the flow diagram of the according to an embodiment of the present invention one specifically method of task schedule；
Fig. 6 is a kind of main modular schematic diagram of the device of task schedule according to an embodiment of the present invention；
Fig. 7 is that the embodiment of the present invention can be applied to exemplary system architecture figure therein；
Fig. 8 is adapted for the structural representation for realizing the mobile device of the embodiment of the present invention or the computer system of server
Figure.
Specific embodiment
Below in conjunction with attached drawing, an exemplary embodiment of the present invention will be described, including the various of the embodiment of the present invention
Details should think them only exemplary to help understanding.Therefore, those of ordinary skill in the art should recognize
It arrives, it can be with various changes and modifications are made to the embodiments described herein, without departing from scope and spirit of the present invention.Together
Sample, for clarity and conciseness, descriptions of well-known functions and structures are omitted from the following description.
It should be noted that the embodiment of the present invention is applicable in task allocation scenarios, for example, TBSchedule,
Clover。
Distributed system error-prone, especially processing competition with it is synchronous during, existing solution, usually
Distributed system is coordinated based on reliable coordination system, for example, Zookeeper, and function provided by Zookeeper is extremely
It less include: configuring maintenance, name Service, distributed synchronization, group service etc..
More mature universal in view of zookeeper, present invention is primarily based on zookeeper implementations, and do not do to Zookeeper
Any improvement has only used some functions of Zookeeper.But it should be recognized that it is identical that its function also can be used
Distributed component.
It lays down a definition below to noun according to the present invention:
Proof of work (Proof-of-Work, PoW) is that a kind of corresponding with service is abused with resource or service is blocked to attack
The economic measure hit.User is usually required to carry out some time-consuming complex calculations appropriate, and answer can be quick by service side
Checking computations, time, equipment and the energy consumed using this to ensure to service with resource are made by real demand as Cost of Guarantee
With.
Used time MS: task performer amount of calculation proves the time used, is counted with millisecond.
Actuator: specific execution task applies role, is made of hardware machine and software systems.In the present invention, it holds
Row device is divided into the Master host node actuator of distribution task and handles the task performer of task.
Master host node machine: carrying out job enquiry and that distributes applies role, can merge with actuator code
It splits, is made of hardware machine and software systems.
Node (Master node or Executor node): Zookeeper term can be used for storing for distributed system
The unit of data.Node can be named as Master or Executor, under node can deposit data or newly-built child node, it is generally tree-like
Structure.
Referring to Fig. 1, thus it is shown that a kind of broad flow diagram of the method for task schedule provided in an embodiment of the present invention, including
Following steps:
S101: determining task performer, sends workload computing and instructs to each task performer.
S102: receiving the end value that each task performer is fed back, and determines each task performer feedback result value institute
The time needed, to calculate the working efficiency of each task performer；Wherein, end value is corresponding to workload computing instruction.
S103: determining task amount to be processed, according to the working efficiency of each task performer, carries out task amount distribution.
The present invention is mainly made of two roles: host node Master and actuator Executor, the two roles pass through
It carries out Master election in Zookeeper to determine, to guarantee to only exist a host node in synchronization system.To make system
Sound and stable operation, the present invention are illustrated with there is effective Master host node always.
To avoid subsequent title from obscuring, later for Master host node actuator, abbreviation Master host node；For
The actuator that task is handled under Executor node, abbreviation task performer.
For step S101, election for host node can be in system starting, be also possible to each child node
Detecting that host node is deposited when abnormal, registration request can be the identification information of each actuator, for example, the IP of actuator
Location.Specifically:
1) when system starts for the first time, host node can be elected, Zookeeper can receive the main section of each actuator at this time
Point registration request determines that the actuator that election is succeeded in registration first is main node M aster, remaining election master node failure is held
Row device goes under Executor node to be registered as child node again, so that Master distributes task accordingly；
2) after system starts and elects host node, Zookeeper can be based on predetermined detection period, periodic detection
Whether Master node machine survives, if existing state is abnormal (for example, delay machine, network failure), sends to all actuators
Notice (including this abnormal Master node machine), to re-elect host node Master.
Extremely is resulted in the need for for host node, the host node redefined continues in undertaking the case where re-registering host node
The work of one host node, and its unfinished task amount, can distribute to other task performers and be executed.
In addition, Master host node does not need processing task, therefore for the registration request of host node, wherein without comprising
Workload value, but the IP address for having actuator is included at least, appoint convenient for subsequent Master host node to the distribution of actuator child node
It is used when business.
Registration for task performer machine, registration content are the IP address and its initialization amount PoW of actuator.
And since the child node of Executor is registrable multiple, communicate in normal situation, a possibility that registration failure is lower, if but losing
It loses, then the actuator will not be assigned task.In addition, if can periodically be retransmitted when the actuator detects unregistered success
Registration request reduces the wasting of resources.
Referring specifically to shown in Fig. 2, system is coordinated for zookeeper, TASK_TYPE_1 is task type node, the section
There are Master host node and Executor node under point.A node, content Master can wherein be increased under Master node newly
The IP address of machine；Registrable multiple nodes under Executor node, content are the IP address and PoW value of Executor machine.
IP address information is saved in Master node, is elected as Master to distinguish the actuator of which IP address.Its
The task performer of minor node can inquire Master host node content, periodically to distinguish one's own duty and to re-elect.
Wherein, this re-elect can be detect Master host node exception after notified by Master, be also possible to by actuator
Laggard rob of querying node Master host node disappearance accounts for.
After determining Master host node, by the actuator being registered under Master host node, it is registered in by reading
The identification information of task performer under Executor node obtains task performer list.Master host node is based on IP
Location is communicated with each task performer in actuator list, and notice carries out proof of work calculating or task processing.
Further, for task performer list generated, since the quantity of task performer may increase at any time
Or it reduces.To guarantee follow-up work processing effectively, need to detect currently valid task performer quantity, to guarantee that task can
Correct distribution.
Master host node is sent to the workload computing instruction of each task performer, can be and is sent out with content-form
It send, such as " calculating its hash value top N after increasing variable by random number is 0 ", is a kind of proof of work algorithm, wherein N
Can be by actuator child node quantity self-setting, and any data hash value is all different and completely random, such as:
1) after " 0 " is calculated by MD5 (Message Digest Algorithm MD5, Message Digest 5), hash value
For " CFCD208495D565EF66E7DFF9F98764DA "；
2) to find the hash value that first 1 is 0, then 27 hash value is obtained after being added to 27 since 0
"02E74F10E0327AD868D138F2B4FDD6F0"；
3) to find the hash value that first 2 are 0, then 168 is added to and obtains its hash value
“006F52E9102A8D3BE2FE5614F42BA989”。
For step S102, each task performer starts to calculate after receiving workload computing instruction, and will calculate
End value returns to Master host node.Here end value, be meet the end value of proof of work algorithm requirement, such as
Hash value is " 3636638817772E42B59D74CFF571FBB3 ".
But it should be recognized that task performer, proves algorithm, institute for the different operating amount that Master host node issues
The end value of return is different, but all meets proof of work algorithm requirement.Such as above-mentioned example, if it is desired to the Hash since 0
It is worth the number of front two, task performer then returns to 168 and waits numerical value.
It further, is raising task treatment effeciency, before the working efficiency of calculating task actuator, the main section of Master
Point can dynamically be detected, for the end value and institute's elapsed time that each task performer is returned to hold to task
Row device is screened, specifically:
1) whether correct the end value that each task performer is returned is verified: for example, it is desirable that the front two of hash value is
0, return value is 168 and is verified；But if return value is 169, the hash value that host node calculates 169 is
" 3636638817772E42B59D74CFF571FBB3 " is undesirable, then does not pass through verifying；
2) it verifies each task performer and returns the result the time consumed by value: for example, due to degraded performance, network event
The reasons such as delay machine, cause certain time not calculate end value in barrier, calculating process, these can not returned the result to appointing for value
Business actuator weeds out.
The time that Master host node is returned according to each task performer calculates the work effect of each task performer
Rate rate is referred to as being allocation proportion.For example, 752ms --- first actuator used time when 10 actuators share
10ms, second used time 23ms ... ..., the last one used time 10ms, by determining time reciprocal value and each time reciprocal value
The sum of ratio, for task performer 1, working efficiency rate are as follows:
Rate1=(1/10)/[(1/10)+(1/23)+...+(1/10)].
It, can be according to each task when Master host node is inquired or receives task to be treated for step S103
The working efficiency of actuator and just in untreated task amount, is intelligently allocated:
1) for system initial start up phase, the proof of work of each task performer is 0, at this time can direct root
According to its working efficiency, the task amount accordingly distributed is determined；That is working efficiency * general assignment amount；
2) is had been turned on for system the stage, when receiving new task, there may be partial task actuator, there are also not complete
At task, need task quantity in combination processing to carry out task distribution at this time, such as execute task in each task performer
After 10s, start the distribution of the second subtask, is still the i.e. rate=(task amount in processing+newly distribute task based on working efficiency
Amount)/general assignment amount, therefore newly distribute task amount in task amount=general assignment amount * rate- processing.
Further, some numerical results value may not be integer, can be using the side for rounding up or being rounded downwards
Formula determines task sendout.Distribution for remaining task amount (general assignment amount-distributed task amount), can there are many real
Apply mode:
1) according to the mode to round up, remaining task amount can be distributed to the task execution minimum to working efficiency
Device, to improve overall task treatment effeciency；
2) according to the mode being rounded downwards, remaining task amount can be distributed to the highest task execution of working efficiency
Overall task treatment effeciency equally can be improved in device；
3) performance for being not necessarily based on each machine is ranked up, and directly carries out order-assigned according to task performer list, will
Remaining task amount distributes into table the last one task performer,.
Method provided by above-described embodiment provides a kind of task process performance based on machine and smart allocation task
Thinking realized " able people should do more work " with giving full play to the processing capacity of high performance tasks actuator, and then improve at overall task
Manage speed.
Referring to Fig. 3, a kind of method flow schematic diagram of optional task schedule according to an embodiment of the present invention is shown, is wrapped
Include following steps:
S301: determining task performer, sends workload computing and instructs to each task performer.
S302: receiving the end value that each task performer is fed back, and determines each task performer feedback result value institute
The time needed, the total time of all task performer feedback result values is counted, the proof of work of each task performer is calculated
Value；Wherein, proof of work value=all task performers total time/individual task actuator time.
S303: according to the proof of work value of each task performer, the amount of work card of all task performers is determined
Bright value calculates the working efficiency of each task performer；Wherein, working efficiency=individual task actuator proof of work
Value/all task performers amount of work proof value.
S304: determining task amount to be processed, according to the working efficiency of each task performer, carries out task amount distribution.
In above embodiment, step S301, S304 can distinguish the description of step S101, S103 shown in Figure 1, herein
It repeats no more.
In above embodiment, for step S302, Master host node according to each task performer returned when
Between, the PoW value of each task performer is calculated, for example, 752ms when 10 actuators share, with 752ms divided by each actuator
The time of cost obtains the PoW value of each task performer, it may be assumed that
Wherein, MSNIndicate the used time of task performer N, M is the quantity of effective task performer, PoWNAppoint to calculate gained
The proof of work value of business actuator N.
Further, its PoW value can be safeguarded and (be updated and save) to the time returned for each task performer,
For example, its PoW value is 75.20 after task performer 1 calculates for the first time, then host node needs to save this value, and in this, as
Distribute the foundation of task amount；After second obtains proof of work, PoW value is 5.94, then host node machine needs
5.94 are revised as by 75.20.
Further, proof of work value can be updated according to predetermined period, for example, 10 minutes, 5 minutes,
It can be other times voluntarily configured.For example, the PoW value of first time calculating task actuator 1 be 75.20, but 10 minutes it
Afterwards, its availability is caused to decline since network jitter or other work occupy CPU, its PoW value obtained by second of calculating is
5.94, then the task quantity of lower sub-distribution will be reduced.
For step S303, it is based on gained proof of work value, can determine corresponding working efficiency, it may be assumed that
Wherein, rateNFor the working efficiency of task performer N, ∑ PoW is the cumulative of all task performer PoW values.
Equally, in machine use process, it is understood that there may be machine loss, using the situations such as abnormal causes its working efficiency to drop
It is low.Resulting working efficiency is calculated every time as a result, to be also not quite similar.It, can also be with dimension for the working efficiency being calculated
The mode of shield carries out preservation update, is convenient for subsequent query, improves task and distributes accuracy.
For each task performer, resulting PoW value and corresponding working efficiency can be generated table, be stored in
It is as shown in table 1 below in node:
1 PoW and rate value example of table
It should be noted that since the quantity of task performer may change at any time, and workload computing instruction is not
It can be sent as its quantity changes every time and (avoid server stress excessive), for example, effective task performer number before 10s
Amount is 10, and quantity becomes 20 after 10s, but workload computing instruction is 10min primary.
Task performer quantity is changed, and the case where the proof of work update cycle is not up to, it can be with direct basis
The proof of work value of each task performer, calculates corresponding working efficiency, without relying on the used time, and is convenient for follow-up management,
Reduce server service pressure.
Method provided by above-described embodiment specifically calculates each task performer according to the mode of proof of work
Working efficiency.For task performer quantity change and not up to the proof of work update cycle the case where, can be with direct basis
The proof of work value of each task performer, determines corresponding working efficiency.
Referring to fig. 4, a kind of method flow schematic diagram of optional task schedule according to an embodiment of the present invention is shown, is wrapped
Include following steps:
S401: determining task performer, sends workload computing and instructs to each task performer.
S402: receiving the end value that each task performer is fed back, and determines each task performer feedback result value institute
The time needed, to calculate the working efficiency of each task performer；Wherein, end value is corresponding to workload computing instruction.
S403: receiving newly assigned task amount, obtains each task performer currently untreated task amount, obtains wait locate
The task amount of reason.
S404: according to task amount to be processed, the working efficiency of each task performer and current untreated task
Amount is determined and is distributed in newly assigned task amount to the task amount of each task performer.
In above embodiment, step S401 can be found in the description of step S101 shown in Fig. 1, and step S402 can be found in Fig. 1
The description of shown step S102 also may refer to step S302 shown in Fig. 3 and the description of S303, and details are not described herein.
In above embodiment, for step S403 and S404, for system initial start time, each task execution
Task amount in device processing is 0, therefore can distribute task amount according to working efficiency rate.
System has been run, task performer has started to process the task moment, if receiving newly assigned task, each
Task performer results in the need for distributing different new task amounts due to the task amount difference in processing.Allocation strategy is still based on work
Efficiency carries out, i.e., the task amount ratio after examination distribution is equal to rate value, therefore newly distribution task amount=general assignment amount * rate- does not locate
The task amount of reason, untreated task amount+newly assigned task amount of general assignment amount=total.
For example, with reference to shown in table 2, by taking task performer 1 as an example,
1) for the first sub-distribution, initialization amount is 0, and the task amount thus distributed is 0.25*1000=250
Item；
2) after 10s, 34 is there remains and be not disposed, but carried out a collection of new task again, total quantity is still 1000
Item, the working efficiency at this time based on the task performer 1 determine the sum of distributed task amount and unfinished task amount accounting institute
The task amount for having task performer total is still consistent, i.e. (34+x)/(1000+236)=0.25, is calculated 309, that is, indicate for
This needs to distribute 309-34=275 task.
2 task sendout of table
Method provided by above-described embodiment, provides a kind of mode of specific implementation task distribution, and task performer is worked as
It is preceding whether to have untreated task, task distribution can be carried out by working efficiency, to realize that task distributes intelligent processing.
Referring to Fig. 5, the method flow schematic diagram of according to an embodiment of the present invention one task schedule specifically is shown, is wrapped
Include following steps:
S501: when system starts for the first time or detect current primary node actuator exception, each actuator transmission is received
Host node registration request, electing the actuator that succeeds in registration first is host node actuator, and determines that remaining actuator is to appoint
Business actuator；Wherein, registration request includes at least the identification information of actuator.
S502: it when reaching the scheduled update period, sends workload computing and instructs to each task performer.
S503: the end value that each task performer is fed back is received, based on scheduled proof rule to the knot fed back
Fruit value is verified, and the task performer that end value does not meet proof rule is rejected.
S503 ': the time needed for determining each task performer feedback result value, based on scheduled time threshold value to determining
Time screened, reject the time exceed scheduled time threshold value task performer.
S504: the time needed for determining each task performer feedback result value counts all task performer feedback knots
The total time of fruit value calculate the proof of work value of each task performer；Wherein, proof of work value=all task executions
The total time of device/individual task actuator time.
S505: according to the proof of work value of each task performer, the amount of work card of all task performers is determined
Bright value calculates the working efficiency of each task performer；Wherein, working efficiency=individual task actuator proof of work
Value/all task performers amount of work proof value.
S506: receiving newly assigned task amount, obtains each task performer currently untreated task amount, obtains wait locate
The task amount of reason.
S507: according to task amount to be processed, the working efficiency of each task performer and current untreated task
Amount is determined and is distributed in newly assigned task amount to the task amount of each task performer.
In above embodiment, step S501~S503 ' can be found in the description of step S101 shown in Fig. 1；Step S504,
S505 can distinguish the description of step S302, S303 shown in Figure 3, and step S506, S507 can distinguish step shown in Figure 4
The description of S403, S404, details are not described herein.
It should be noted that the present invention is when configuring actuator, without manually calculating the processing capacity of each machine.
For handling the machine of same generic task, it is generally the case that the almost the same machine of hardware configuration can be configured as far as possible.
But in practical work process, due to the increase of business demand, machine is often repeatedly increased newly, for example, may 3 machines before 1 year
Device may need to increase to regard to enough, after 1 year 10, and due to the missing of machine resources every time newly-increased machine with before
Hardware configuration may not be completely the same.
Currently generally required artificial to guarantee consistent, the workload that distribution is balanced of machine performance in actual operation
Configuration is calculated, and splitting and reorganizing is carried out to hardware, for example, the CPU of 8 cores is invented 42 core CPU, to guarantee performance one
It causes.Method shown in the present invention, the working efficiency based on machine carry out task distribution, are not necessarily to the labour cost when configuring machine and calculate often
The processing capacity of a machine realizes the effect of smart allocation.
Method provided by the embodiment of the present invention, by way of proof of work, each task performer of accurate evaluation
Processing capacity realize task smart allocation, reach realization task and according to its processing capacity and task amount to be processed
Quickly processing.In addition, avoiding task performer by way of regularly updating to proof of work value, working efficiency
There is abnormal, the distribution more situation for causing task treatment effeciency low of task amount.
Referring to Fig. 6, a kind of main modular signal of the device 600 of task schedule provided in an embodiment of the present invention is shown
Figure, comprising:
Instruction sending module 601 sends workload computing and instructs to each task execution for determining task performer
Device；
Efficiency calculation module 602, the end value fed back for receiving each task performer, determines each task execution
Time needed for device feedback result value, to calculate the working efficiency of each task performer；Wherein, end value and workload computing
Instruction is corresponding；
Task allocating module 603, for determining task amount to be processed, according to the working efficiency of each task performer,
Carry out task amount distribution.
In implementing device of the present invention, efficiency calculation module 602 is used for:
Time needed for determining each task performer feedback result value counts all task performer feedback result values
Total time calculate the proof of work value of each task performer；Wherein, proof of work value=all task performers are total
Time/individual task actuator time；
According to the proof of work value of each task performer, the amount of work proof value of all task performers is determined,
Calculate the working efficiency of each task performer；Wherein, working efficiency=individual task actuator proof of work value/all
The amount of work proof value of task performer.
In implementing device of the present invention, efficiency calculation module 602 is used for:
Receive the end value that each task performer is fed back, based on scheduled proof rule to the end value fed back into
The task performer that end value does not meet proof rule is rejected in row verifying；And
Time needed for determining each task performer feedback result value, based on scheduled time threshold value to the received time
It is screened, rejects task performer of the time beyond scheduled time threshold value.
In implementing device of the present invention, task allocating module 603 is used for:
Newly assigned task amount is received, each task performer currently untreated task amount is obtained, obtains to be processed
Task amount；
According to task amount to be processed, the working efficiency and current untreated task amount of each task performer, really
It is distributed in fixed newly assigned task amount to the task amount of each task performer.
In implementing device of the present invention, instruction sending module 601 is used for: when reaching the scheduled update period, sending workload
Computations are to each task performer.
In addition, the specific implementation content of the task scheduling apparatus described in embodiments of the present invention, task described above
It has been described in detail in dispatching method, therefore has no longer illustrated in this duplicate contents.
Device provided by the embodiment of the present invention, by way of proof of work, each task performer of accurate evaluation
Processing capacity realize task smart allocation, reach realization task and according to its processing capacity and task amount to be processed
Quickly processing.In addition, avoiding task performer by way of regularly updating to proof of work value, working efficiency
There is abnormal, the distribution more situation for causing task treatment effeciency low of task amount.
The exemplary of the method for scheduling task or task scheduling apparatus that can apply the embodiment of the present invention is shown referring to Fig. 7
System architecture 700.
As shown in fig. 7, system architecture 700 may include terminal device 701,702,703, network 704 and server 705.
Network 704 between terminal device 701,702,703 and server 705 to provide the medium of communication link.Network 704 can be with
Including various connection types, such as wired, wireless communication link or fiber optic cables etc..
User can be used terminal device 701,702,703 and be interacted by network 704 with server 705, to receive or send out
Send message etc..Various telecommunication customer end applications, such as the application of shopping class, net can be installed on terminal device 701,702,703
(merely illustrative) such as the application of page browsing device, searching class application, instant messaging tools, mailbox client, social platform softwares.
Terminal device 701,702,703 can be the various electronic equipments with display screen and supported web page browsing, packet
Include but be not limited to smart phone, tablet computer, pocket computer on knee and desktop computer etc..
Server 705 can be to provide the server of various services, such as utilize terminal device 701,702,703 to user
The shopping class website browsed provides the back-stage management server (merely illustrative) supported.Back-stage management server can be to reception
To the data such as information query request analyze etc. processing, and by processing result (such as target push information, product letter
Breath -- merely illustrative) feed back to terminal device.
It should be noted that method for scheduling task provided by the embodiment of the present invention is generally executed by server 705, accordingly
Ground, task scheduling apparatus are generally positioned in server 705.
It should be understood that the number of terminal device, network and server in Fig. 7 is only schematical.According to realization need
It wants, can have any number of terminal device, network and server.
Referring to Fig. 8, it illustrates the knots of the computer system 800 for the terminal device for being suitable for being used to realize the embodiment of the present invention
Structure schematic diagram.Terminal device shown in Fig. 8 is only an example, should not function and use scope band to the embodiment of the present invention
Carry out any restrictions.
As shown in figure 8, computer system 800 includes central processing unit (CPU) 801, it can be read-only according to being stored in
Program in memory (ROM) 802 or be loaded into the program in random access storage device (RAM) 803 from storage section 808 and
Execute various movements appropriate and processing.In RAM 803, also it is stored with system 800 and operates required various programs and data.
CPU 801, ROM 802 and RAM 803 are connected with each other by bus 804.Input/output (I/O) interface 805 is also connected to always
Line 804.
I/O interface 805 is connected to lower component: the importation 806 including keyboard, mouse etc.；It is penetrated including such as cathode
The output par, c 807 of spool (CRT), liquid crystal display (LCD) etc. and loudspeaker etc.；Storage section 808 including hard disk etc.；
And the communications portion 809 of the network interface card including LAN card, modem etc..Communications portion 809 via such as because
The network of spy's net executes communication process.Driver 810 is also connected to I/O interface 805 as needed.Detachable media 811, such as
Disk, CD, magneto-optic disk, semiconductor memory etc. are mounted on as needed on driver 810, in order to read from thereon
Computer program be mounted into storage section 808 as needed.
Particularly, disclosed embodiment, the process described above with reference to flow chart may be implemented as counting according to the present invention
Calculation machine software program.For example, embodiment disclosed by the invention includes a kind of computer program product comprising be carried on computer
Computer program on readable medium, the computer program include the program code for method shown in execution flow chart.In
In such embodiment, which can be downloaded and installed from network by communications portion 809, and/or from can
Medium 811 is dismantled to be mounted.When the computer program is executed by central processing unit (CPU) 801, system of the invention is executed
The above-mentioned function of middle restriction.
It should be noted that computer-readable medium shown in the present invention can be computer-readable signal media or meter
Calculation machine readable storage medium storing program for executing either the two any combination.Computer readable storage medium for example can be --- but not
Be limited to --- electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor system, device or device, or any above combination.Meter
The more specific example of calculation machine readable storage medium storing program for executing can include but is not limited to: have the electrical connection, just of one or more conducting wires
Taking formula computer disk, hard disk, random access storage device (RAM), read-only memory (ROM), erasable type may be programmed read-only storage
Device (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory device,
Or above-mentioned any appropriate combination.In the present invention, computer readable storage medium can be it is any include or storage journey
The tangible medium of sequence, the program can be commanded execution system, device or device use or in connection.And at this
In invention, computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal,
Wherein carry computer-readable program code.The data-signal of this propagation can take various forms, including but unlimited
In electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be that computer can
Any computer-readable medium other than storage medium is read, which can send, propagates or transmit and be used for
By the use of instruction execution system, device or device or program in connection.Include on computer-readable medium
Program code can transmit with any suitable medium, including but not limited to: wireless, electric wire, optical cable, RF etc. are above-mentioned
Any appropriate combination.
Flow chart and block diagram in attached drawing are illustrated according to the system of various embodiments of the invention, method and computer journey
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part of one module, program segment or code of table, a part of above-mentioned module, program segment or code include one or more
Executable instruction for implementing the specified logical function.It should also be noted that in some implementations as replacements, institute in box
The function of mark can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are practical
On can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it wants
It is noted that the combination of each box in block diagram or flow chart and the box in block diagram or flow chart, can use and execute rule
The dedicated hardware based systems of fixed functions or operations is realized, or can use the group of specialized hardware and computer instruction
It closes to realize.
Being described in module involved in the embodiment of the present invention can be realized by way of software, can also be by hard
The mode of part is realized.Described module also can be set in the processor, for example, can be described as: a kind of processor packet
Include instruction sending module, efficiency calculation module, task allocating module.Wherein, the title of these modules is not under certain conditions
The restriction to the module itself is constituted, for example, efficiency calculation module is also described as " the working efficiency meter of task performer
Calculate module ".
As on the other hand, the present invention also provides a kind of computer-readable medium, which be can be
Included in equipment described in above-described embodiment；It is also possible to individualism, and without in the supplying equipment.Above-mentioned calculating
Machine readable medium carries one or more program, when said one or multiple programs are executed by the equipment, makes
Obtaining the equipment includes:
It determines task performer, sends workload computing and instruct to each task performer；
The end value that each task performer is fed back is received, when determining needed for each task performer feedback result value
Between, to calculate the working efficiency of each task performer；Wherein, the end value is corresponding to workload computing instruction；
It determines task amount to be processed, according to the working efficiency of each task performer, carries out task amount distribution.
Technical solution according to an embodiment of the present invention, by way of proof of work, each task execution of accurate evaluation
The processing capacity of device, and according to its processing capacity and task amount to be processed, it realizes task smart allocation, it is quick to reach task
The effect of processing.In addition, it is different to avoid task performer appearance by way of regularly updating to proof of work value
Often, the more situation for causing task treatment effeciency low of task amount is distributed.
Above-mentioned specific embodiment, does not constitute a limitation on the scope of protection of the present invention.Those skilled in the art should be bright
It is white, design requirement and other factors are depended on, various modifications, combination, sub-portfolio and substitution can occur.It is any
Made modifications, equivalent substitutions and improvements etc. within the spirit and principles in the present invention, should be included in the scope of the present invention
Within.
Claims (12)
1. a kind of method of task schedule characterized by comprising
It determines task performer, sends workload computing and instruct to each task performer；
The end value that each task performer is fed back is received, the time needed for determining each task performer feedback result value,
To calculate the working efficiency of each task performer；Wherein, the end value is corresponding to workload computing instruction；
It determines task amount to be processed, according to the working efficiency of each task performer, carries out task amount distribution.
2. the method according to claim 1, wherein needed for each task performer feedback result value of the determination
Time, include: to calculate the working efficiency of each task performer
Time needed for determining each task performer feedback result value, count all task performer feedback result values it is total when
Between, calculate the proof of work value of each task performer；Wherein, when proof of work value=all task performers are total
Between/time of individual task actuator；
According to the proof of work value of each task performer, the amount of work proof value of all task performers is determined, calculate
The working efficiency of each task performer；Wherein, working efficiency=individual task actuator proof of work value/all tasks
The amount of work proof value of actuator.
3. the method according to claim 1, wherein the result for receiving each task performer and being fed back
Value, time needed for determining each task performer feedback result value further include:
The end value that each task performer is fed back is received, the end value fed back is tested based on scheduled proof rule
Card rejects the task performer that end value does not meet the proof rule；And
Time needed for determining each task performer feedback result value carries out the identified time based on scheduled time threshold value
Task performer of the time beyond the scheduled time threshold value is rejected in screening.
4. the method according to claim 1, wherein the task amount that the determination is to be processed, according to each task
The working efficiency of actuator, carrying out task amount distribution includes:
Newly assigned task amount is received, each task performer currently untreated task amount is obtained, obtains described to be processed
Task amount；
According to the task amount to be processed, the working efficiency and current untreated task amount of each task performer, really
It is distributed in the fixed newly assigned task amount to the task amount of each task performer.
5. the method according to claim 1, wherein the transmission workload computing is instructed to each task execution
Device includes:
When reaching the scheduled update period, sends the workload computing and instruct to each task performer.
6. a kind of device of task schedule characterized by comprising
Instruction sending module sends workload computing and instructs to each task performer for determining task performer；
Efficiency calculation module, the end value fed back for receiving each task performer determine each task performer feedback
Time needed for end value, to calculate the working efficiency of each task performer；Wherein, the end value and the work meter
It is corresponding to calculate instruction；
Task allocating module, according to the working efficiency of each task performer, carries out task for determining task amount to be processed
Amount distribution.
7. device according to claim 6, which is characterized in that the efficiency calculation module is used for:
Time needed for determining each task performer feedback result value, count all task performer feedback result values it is total when
Between, calculate the proof of work value of each task performer；Wherein, when proof of work value=all task performers are total
Between/time of individual task actuator；
According to the proof of work value of each task performer, the amount of work proof value of all task performers is determined, calculate
The working efficiency of each task performer；Wherein, working efficiency=individual task actuator proof of work value/all tasks
The amount of work proof value of actuator.
8. device according to claim 6, which is characterized in that the efficiency calculation module is used for:
The end value that each task performer is fed back is received, the end value fed back is tested based on scheduled proof rule
Card rejects the task performer that end value does not meet the proof rule；And
Time needed for determining each task performer feedback result value carries out institute the received time based on scheduled time threshold value
Task performer of the time beyond the scheduled time threshold value is rejected in screening.
9. device according to claim 6, which is characterized in that the task allocating module is used for:
Newly assigned task amount is received, each task performer currently untreated task amount is obtained, obtains described to be processed
Task amount；
According to the task amount to be processed, the working efficiency and current untreated task amount of each task performer, really
It is distributed in the fixed newly assigned task amount to the task amount of each task performer.
10. device according to claim 6, which is characterized in that described instruction sending module is used for:
When reaching the scheduled update period, sends the workload computing and instruct to each task performer.
11. a kind of electronic equipment characterized by comprising
One or more processors；
Storage device, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processors are real
Now such as method as claimed in any one of claims 1 to 5.
12. a kind of computer-readable medium, is stored thereon with computer program, which is characterized in that described program is held by processor
Such as method as claimed in any one of claims 1 to 5 is realized when row.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810460659.3A CN110489224A (en) | 2018-05-15 | 2018-05-15 | A kind of method and apparatus of task schedule |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810460659.3A CN110489224A (en) | 2018-05-15 | 2018-05-15 | A kind of method and apparatus of task schedule |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110489224A true CN110489224A (en) | 2019-11-22 |
Family
ID=68545071
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810460659.3A Pending CN110489224A (en) | 2018-05-15 | 2018-05-15 | A kind of method and apparatus of task schedule |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110489224A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112379986A (en) * | 2020-11-16 | 2021-02-19 | 北京字跳网络技术有限公司 | Task processing method and device and electronic equipment |
| CN112486658A (en) * | 2020-12-17 | 2021-03-12 | 华控清交信息科技(北京)有限公司 | Task scheduling method and device for task scheduling |
| CN112948111A (en) * | 2021-02-26 | 2021-06-11 | 北京奇艺世纪科技有限公司 | Task allocation method, device, equipment and computer readable medium |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0862113A2 (en) * | 1997-02-27 | 1998-09-02 | Zuno Limited | Autonomous agent architecture |
| CN101916209A (en) * | 2010-08-06 | 2010-12-15 | 华东交通大学 | Cluster task resource allocation method for multi-core processor |
| CN101957780A (en) * | 2010-08-17 | 2011-01-26 | 中国电子科技集团公司第二十八研究所 | Resource state information-based grid task scheduling processor and grid task scheduling processing method |
| CN102111337A (en) * | 2011-03-14 | 2011-06-29 | 浪潮(北京)电子信息产业有限公司 | Method and system for task scheduling |
| CN104239148A (en) * | 2013-06-06 | 2014-12-24 | 腾讯科技（深圳）有限公司 | Distributed task scheduling method and device |
| CN105049268A (en) * | 2015-08-28 | 2015-11-11 | 东方网力科技股份有限公司 | Distributed computing resource allocation system and task processing method |
| CN105373428A (en) * | 2015-12-09 | 2016-03-02 | 北京奇虎科技有限公司 | Task scheduling method and system |
| CN105487930A (en) * | 2015-12-01 | 2016-04-13 | 中国电子科技集团公司第二十八研究所 | Task optimization scheduling method based on Hadoop |
| CN106126332A (en) * | 2016-06-27 | 2016-11-16 | 北京京东尚科信息技术有限公司 | Distributed timing task scheduling system and method |
| CN106325984A (en) * | 2015-06-30 | 2017-01-11 | 上海证大喜马拉雅网络科技有限公司 | Big data task scheduling apparatus |
| CN106528280A (en) * | 2015-09-15 | 2017-03-22 | 阿里巴巴集团控股有限公司 | Task allocation method and system |
| KR101794696B1 (en) * | 2016-08-12 | 2017-11-07 | 서울시립대학교 산학협력단 | Distributed processing system and task scheduling method considering heterogeneous processing type |
-
2018
- 2018-05-15 CN CN201810460659.3A patent/CN110489224A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0862113A2 (en) * | 1997-02-27 | 1998-09-02 | Zuno Limited | Autonomous agent architecture |
| CN101916209A (en) * | 2010-08-06 | 2010-12-15 | 华东交通大学 | Cluster task resource allocation method for multi-core processor |
| CN101957780A (en) * | 2010-08-17 | 2011-01-26 | 中国电子科技集团公司第二十八研究所 | Resource state information-based grid task scheduling processor and grid task scheduling processing method |
| CN102111337A (en) * | 2011-03-14 | 2011-06-29 | 浪潮(北京)电子信息产业有限公司 | Method and system for task scheduling |
| CN104239148A (en) * | 2013-06-06 | 2014-12-24 | 腾讯科技（深圳）有限公司 | Distributed task scheduling method and device |
| CN106325984A (en) * | 2015-06-30 | 2017-01-11 | 上海证大喜马拉雅网络科技有限公司 | Big data task scheduling apparatus |
| CN105049268A (en) * | 2015-08-28 | 2015-11-11 | 东方网力科技股份有限公司 | Distributed computing resource allocation system and task processing method |
| CN106528280A (en) * | 2015-09-15 | 2017-03-22 | 阿里巴巴集团控股有限公司 | Task allocation method and system |
| CN105487930A (en) * | 2015-12-01 | 2016-04-13 | 中国电子科技集团公司第二十八研究所 | Task optimization scheduling method based on Hadoop |
| CN105373428A (en) * | 2015-12-09 | 2016-03-02 | 北京奇虎科技有限公司 | Task scheduling method and system |
| CN106126332A (en) * | 2016-06-27 | 2016-11-16 | 北京京东尚科信息技术有限公司 | Distributed timing task scheduling system and method |
| KR101794696B1 (en) * | 2016-08-12 | 2017-11-07 | 서울시립대학교 산학협력단 | Distributed processing system and task scheduling method considering heterogeneous processing type |
Non-Patent Citations (2)
| Title |
|---|
| 徐爱萍;吴笛;徐武平;陈军;: "实时多任务异构云计算平台负载均衡算法", 中国科学技术大学学报, no. 03, 15 March 2016 (2016-03-15) * |
| 杨杭;张昕;赵建平;: "基于动态反馈的负载均衡方法研究", 长春理工大学学报(自然科学版), no. 06, 15 December 2017 (2017-12-15) * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112379986A (en) * | 2020-11-16 | 2021-02-19 | 北京字跳网络技术有限公司 | Task processing method and device and electronic equipment |
| CN112379986B (en) * | 2020-11-16 | 2023-12-12 | 北京字跳网络技术有限公司 | Task processing method and device and electronic equipment |
| CN112486658A (en) * | 2020-12-17 | 2021-03-12 | 华控清交信息科技(北京)有限公司 | Task scheduling method and device for task scheduling |
| CN112948111A (en) * | 2021-02-26 | 2021-06-11 | 北京奇艺世纪科技有限公司 | Task allocation method, device, equipment and computer readable medium |
| CN112948111B (en) * | 2021-02-26 | 2023-07-14 | 北京奇艺世纪科技有限公司 | Task allocation method, device, equipment and computer readable medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yang et al. | Joint computation partitioning and resource allocation for latency sensitive applications in mobile edge clouds | |
| Lu et al. | Join-idle-queue: A novel load balancing algorithm for dynamically scalable web services | |
| US9201690B2 (en) | Resource aware scheduling in a distributed computing environment | |
| US9417903B2 (en) | Storage management for a cluster of integrated computing systems comprising integrated resource infrastructure using storage resource agents and synchronized inter-system storage priority map | |
| Tang et al. | Dynamic request redirection and elastic service scaling in cloud-centric media networks | |
| CN103927229A (en) | Scheduling Mapreduce Jobs In A Cluster Of Dynamically Available Servers | |
| CN108874640A (en) | A kind of appraisal procedure and device of clustering performance | |
| CN110162388A (en) | A kind of method for scheduling task, system and terminal device | |
| CN106330987A (en) | Dynamic load balancing method | |
| CN110489224A (en) | A kind of method and apparatus of task schedule | |
| CN110019339A (en) | A kind of data query method and system | |
| CN110166507A (en) | More resource regulating methods and device | |
| JP2012079242A (en) | Composite event distribution device, composite event distribution method and composite event distribution program | |
| Zhang et al. | Heterogeneity aware dominant resource assistant heuristics for virtual machine consolidation | |
| CN102724103A (en) | Proxy server, hierarchical network system and distributed workload management method | |
| CN109614227A (en) | Task resource concocting method, device, electronic equipment and computer-readable medium | |
| CN110377416A (en) | Distributed subregion method for scheduling task and device | |
| CN109413125A (en) | The method and apparatus of dynamic regulation distributed system resource | |
| CN109879126A (en) | A kind of method and system of elevator reservation | |
| Nahir et al. | Workload factoring with the cloud: A game-theoretic perspective | |
| Convolbo et al. | DRASH: A data replication-aware scheduler in geo-distributed data centers | |
| CN108320097A (en) | Method and apparatus for Amount of work | |
| CN108965364A (en) | Resource allocation method, apparatus and system | |
| Garg et al. | Optimal virtual machine scheduling in virtualized cloud environment using VIKOR method | |
| Li | Optimal load distribution for multiple heterogeneous blade servers in a cloud computing environment |
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 |