CN107229511A - Cluster task equalization scheduling method, device, storage medium and electronic equipment - Google Patents
Cluster task equalization scheduling method, device, storage medium and electronic equipment Download PDFInfo
- Publication number
- CN107229511A CN107229511A CN201710330278.9A CN201710330278A CN107229511A CN 107229511 A CN107229511 A CN 107229511A CN 201710330278 A CN201710330278 A CN 201710330278A CN 107229511 A CN107229511 A CN 107229511A
- Authority
- CN
- China
- Prior art keywords
- task
- node
- time
- allocated
- cluster
- 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
- 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
-
- 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]
- G06F9/5083—Techniques for rebalancing the load in a distributed system
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
This disclosure relates to which a kind of cluster task equalization scheduling method, device, storage medium and electronic equipment, this method include：When listening to first and being dispatched to of about fixed time point, scheduling lock is seized；After the scheduling lock is successfully seized, the corresponding task to be allocated of the described first scheduling is distributed to each node in the cluster according to default task allocation rule；Other nodes in the cluster are notified to perform the task of the first node distribution.Can solve the problem that task distribution it is unbalanced, and task lock expense it is big the problem of, reach cause task distribution it is more balanced, reduction system burden effect.
Description
Technical field
This disclosure relates to load-balancing technique field, in particular it relates to a kind of cluster task equalization scheduling method, device,
Storage medium and electronic equipment.
Background technology
At present, in data acquisition (integrated) business, business needs are met frequently by timing acquiring, such as with 10
Second, 1 minute, 5 minutes, 10 minutes, data acquisition program started to gather number on the time point of agreement as timing acquiring frequency
According to, in business be also referred to as timed task.
In the prior art, after timed task is triggered, all nodes that can perform task seize scheduling lock, get tune
Corresponding task is placed in shared resource pond by the node of degree lock, such as in shared task queue, regularly using competitive way
Lock is got, its competitive way is that all nodes seize task lock, and the node for first getting lock takes the task away, and performs this
Business, after the task has been performed, node release task lock, and appointed again by above-mentioned all nodes by competitive way acquisition
Business lock, is performed repeatedly with this.
It can be seen that, aforesaid way is that, by multiple nodes perform task by competitive way, therefore the distribution of task is uneven
Weighing apparatus, and by the way of competition task lock, when the node of participation competition is more, the expense of task lock is also bigger, can increase
System burden.
The content of the invention
The purpose of the disclosure is to provide a kind of cluster task equalization scheduling method, device, storage medium and electronic equipment, uses
Distribute unbalanced in the task of solution, and task locks the problem of expense is big.
To achieve these goals, in a first aspect, the disclosure provides a population task balance dispatching method, applied to cluster
In any node, methods described includes：
When listening to first and being dispatched to of about fixed time point, scheduling lock is seized；
After the scheduling lock is successfully seized, by the corresponding task to be allocated of the described first scheduling according to default task point
Each node in the cluster is distributed to rule；
Other nodes in the cluster are notified to perform the task of distribution.
Optionally, it is described to distribute to the corresponding task to be allocated of the described first scheduling according to default task allocation rule
Each node in the cluster, including：
The corresponding all tasks to be allocated of described first scheduling are distributed in the cluster according to Principle of Average Allocation
Each node；
In the task queue that the task to be allocated for distributing to each node is added to each node, wherein,
One task queue of each node correspondence.
Optionally, it is described to distribute to the corresponding task to be allocated of the described first scheduling according to default task allocation rule
Each node in the cluster, including：
The task queue information of all nodes in the cluster is obtained, task queue information is included in the cluster
The distribution task that there is currently in the task queue of each node；Wherein, one task queue of each node correspondence；
Determine that each having distributed required by task of the task takes in the task queue of each node；
It is time-consuming according to each having distributed required by task of the task, when determining that the residue of all nodes performs total
Between, and each node the remaining execution time, it is described it is remaining perform total time be all nodes will it is all described in
Distributed tasks carrying complete required total time, remaining execution time of each node for each node by oneself
All in task queue described have distributed tasks carrying the complete required time；
Determine that the task of each required by task to be allocated in the corresponding all tasks to be allocated of first scheduling takes；
Taken according to the task of each required by task to be allocated, determine the total time-consuming of all tasks to be allocated；
All nodes are determined according to the remaining execution total time, and the total time-consuming of all tasks to be allocated
Perform the total time of required by task；
It is defined as the task time of each node distribution according to the number of nodes of the total time and the cluster；
According to the task time for each node distribution, remaining execution time of each node and described every
The task of individual required by task to be allocated is taken as each node distribution task to be allocated；
In the task queue that the task to be allocated for distributing to each node is added to each node.
Optionally, each task consumption for having distributed required by task in the task queue for determining each node
When, including：
When the number of times m that first task was once performed is more than or equal to n, the first task is obtained in first n times execution
When the time-consuming task as needed for the first task of average task take, the first task is any in the cluster
Any in the task queue of node described has distributed task；
When the number of times m that the first task was once performed is less than n, then the first task is obtained in first m times execution
When the time-consuming task as needed for the first task of average task take；
When the first task is not carried out out-of-date, obtain that default task set in advance is time-consuming to be used as the first task institute
Needing for task takes.
Optionally, each required by task to be allocated determined in the corresponding all tasks to be allocated of first scheduling
Task take, including：
When the number of times m that the second task was once performed is more than or equal to n, second task is obtained in first n times execution
When the time-consuming task as second required by task of average task take, second task is the described first scheduling correspondence
All tasks to be allocated in any task to be allocated；
When the number of times m that second task was once performed is less than n, then second task is obtained in first m times execution
When the time-consuming task as second required by task of average task take；
When second task is not carried out out-of-date, obtain that default task set in advance is time-consuming to be used as the second task institute
Needing for task takes, or, when second task is not carried out out-of-date, second task is distributed according to Principle of Average Allocation
To any node in the cluster.
Optionally, when the basis is the task time of each node distribution, the remaining execution of each node
Between and the task of each required by task to be allocated take as each node distribution task to be allocated, including：
The remaining execution time of the first node will be subtracted for the task time that first node is distributed, obtain described first
The task time to be allocated of node, the first node is any node in the cluster；
Taken according to the task of each required by task to be allocated, be defined as to be allocated of the first node distribution
Business so that the task of all required by task to be allocated of first node distribution takes treating point for sum and the first node
With task time matching.
Second aspect, the disclosure also provides a population task balance dispatching device, applied to any node in cluster, institute
Stating device includes：
Module is monitored in scheduling, for when listening to first and being dispatched to of about fixed time point, seizing scheduling lock；
Task allocating module, it is for after the scheduling lock is successfully seized, the described first scheduling is corresponding to be allocated
Each node in the cluster is distributed in business according to default task allocation rule；
Task notifications module, for notifying other nodes in the cluster to perform the task of distribution.
Optionally, the task allocating module, including：
First distribution sub module, for dispatching corresponding all tasks to be allocated according to Principle of Average Allocation by described first
Distribute to each node in the cluster；
Queue management submodule, for the task to be allocated for distributing to each node to be added into each node
Task queue in, wherein, one task queue of each node correspondence.
Optionally, the task allocating module, including：
Queuing message acquisition submodule, the task queue information for obtaining all nodes in the cluster, task team
Column information includes the distribution task that there is currently in the task queue of each node in the cluster；Wherein, it is described every
One task queue of individual node correspondence；
Task takes the first determination sub-module, each described in the task queue for determining each node to have distributed
The task of required by task takes；
Remaining time determination sub-module, for being taken according to each having distributed required by task of the task, it is determined that described
The remaining of all nodes performs total time, and each node the remaining execution time, it is described remaining to perform total time and be
All nodes by it is all it is described distributed tasks carrying complete required total time, the remaining execution time of each node
For each node by distributed described in all in the task queue of oneself tasks carrying it is complete needed for time；
Task takes the second determination sub-module, every in the corresponding all tasks to be allocated of first scheduling for determining
The task of individual required by task to be allocated takes；
Task total time-consuming determination sub-module, for being taken according to the task of each required by task to be allocated, determines institute
State the total time-consuming of all tasks to be allocated；
Node total time determining module, for remaining performing total time, and all tasks to be allocated according to described
Total time-consuming determine that all nodes perform the total time of required by task；
Node tasks time distribution sub module, for being defined as institute according to the number of nodes of the total time and the cluster
State the task time of each node distribution；
Second distribution sub module, for according to for task time of each node distribution, each node it is surplus
The task of remaining execution time and each required by task to be allocated is taken as each node distribution task to be allocated；
Queue management submodule, for the task to be allocated for distributing to each node to be added into each node
Task queue in.
Optionally, the task takes the first determination sub-module, is used for：
When the number of times m that first task was once performed is more than or equal to n, the first task is obtained in first n times execution
When the time-consuming task as needed for the first task of average task take, the first task is any in the cluster
Any in the task queue of node described has distributed task；
When the number of times m that the first task was once performed is less than n, then the first task is obtained in first m times execution
When the time-consuming task as needed for the first task of average task take；
When the first task is not carried out out-of-date, obtain that default task set in advance is time-consuming to be used as the first task institute
Needing for task takes.
Optionally, task takes the second determination sub-module, is used for：
When the number of times m that the second task was once performed is more than or equal to n, second task is obtained in first n times execution
When the time-consuming task as second required by task of average task take, second task is the described first scheduling correspondence
All tasks to be allocated in any task to be allocated；
When the number of times m that second task was once performed is less than n, then second task is obtained in first m times execution
When the time-consuming task as second required by task of average task take；
When second task is not carried out out-of-date, obtain that default task set in advance is time-consuming to be used as the second task institute
Needing for task takes, or, when second task is not carried out out-of-date, second task is distributed according to Principle of Average Allocation
To any node in the cluster.
Optionally, second distribution sub module, is used for：
The remaining execution time of the first node will be subtracted for the task time that first node is distributed, obtain described first
The task time to be allocated of node, the first node is any node in the cluster；
Taken according to the task of each required by task to be allocated, be defined as to be allocated of the first node distribution
Business so that the task of all required by task to be allocated of first node distribution takes treating point for sum and the first node
With task time matching.
The third aspect, the disclosure also provides a kind of computer-readable recording medium, is stored thereon with computer program, described
The step of method described in first aspect being realized when computer program is executed by processor.
Fourth aspect, a kind of electronic equipment, including：Computer-readable recording medium described in the third aspect；And one
Individual or multiple processors, for performing the computer program in the computer-readable recording medium.
Cluster task equalization scheduling method, device, storage medium and electronic equipment that the disclosure is provided, when listening to first
When being dispatched to of about fixed time point, each node in cluster seizes scheduling lock, after the scheduling lock is successfully seized, seizes
The corresponding task to be allocated of described first scheduling is distributed to the collection by the node for dispatching lock according to default task allocation rule
Each node in group, notifies other nodes in the cluster to perform the task of distribution.By above-mentioned technical proposal, perform and appoint
The node of business no longer competes task lock, but after any one node seizes scheduling lock, by currently of about fixed time point the
The corresponding task to be allocated of one scheduling distributes to each node in cluster according to default task allocation rule, by each node
Come together execution task, it is unbalanced so as to solve task distribution, and task lock expense it is big the problem of, reach so that task
Distribution is more balanced, reduces the effect of system burden.
Other feature and advantage of the disclosure will be described in detail in subsequent embodiment part.
Brief description of the drawings
Accompanying drawing is, for providing further understanding of the disclosure, and to constitute a part for specification, with following tool
Body embodiment is used to explain the disclosure together, but does not constitute limitation of this disclosure.In the accompanying drawings：
Fig. 1 is a kind of cluster task equalization scheduling method according to an exemplary embodiment；
Fig. 2 is a kind of method for allocating tasks according to embodiment illustrated in fig. 1；
Fig. 3 is another method for allocating tasks according to embodiment illustrated in fig. 1；
Fig. 4 is a kind of block diagram of cluster task balance dispatching device according to an exemplary embodiment；
Fig. 5 is a kind of block diagram of task allocating module according to embodiment illustrated in fig. 4；
Fig. 6 is the block diagram of another task allocating module according to embodiment illustrated in fig. 4；
Fig. 7 is the block diagram of a kind of electronic equipment according to an exemplary embodiment；
Fig. 8 is the block diagram of another electronic equipment according to an exemplary embodiment.
Embodiment
It is described in detail below in conjunction with accompanying drawing embodiment of this disclosure.It should be appreciated that this place is retouched
The embodiment stated is merely to illustrate and explained the disclosure, is not limited to the disclosure.
Fig. 1 is a kind of cluster task equalization scheduling method according to an exemplary embodiment, as shown in figure 1, the party
Method is applied to any node in cluster, and this method includes：
Step 101, when listening to first and being dispatched to of about fixed time point, scheduling lock is seized.
Example, a series of set equivalent to timed tasks is dispatched, scheduling can trigger a series of when reaching timing point
Task, in the scheduling queue of a cluster, one or more scheduling can be included, the node of all execution tasks all can
Monitor the timing of the scheduling in scheduling queue.When some listened in scheduling queue is dispatched to of about fixing time, institute
The node for having execution task seizes scheduling lock by competitive way.
Step 102, after scheduling lock is successfully seized, by the corresponding task to be allocated of the first scheduling according to default task point
Each node in cluster is distributed to rule.
Example, in the present embodiment, the node of task is performed after scheduling lock is successfully seized, determining for task is no longer monitored
When, also no longer competition task is locked, but by robbing the node that account for dispatching lock by all tasks to be allocated of the scheduling according to default
Task allocation rule distribute to including each node in the cluster including the node.The task allocation rule can be average
Distribute or distributed according to need according to the load of each node.
Step 103, other nodes in cluster are notified to perform the task of distribution.
Due to, each perform task node, no longer monitor task timing, therefore rob account for scheduling lock node will
All tasks of the scheduling are distributed to after each node in cluster, in addition it is also necessary to the node notice collection that account for scheduling lock is robbed by this
Other nodes in group, to activate the worker thread of other each nodes so that each node performs just having distributed for task.
, wherein it is desired to which explanation, in the present embodiment, each node in cluster correspond to a task queue, rob
Account for the node of scheduling lock can be added to the task for each node distribution in each corresponding task queue of node.Thus
It can be seen that, each node no longer shares a task queue, task need not be also performed by way of competition again, so that task
Distribution it is more balanced.
The cluster task equalization scheduling method that the disclosure is provided, when listening to first and being dispatched to of about fixed time point,
Each node in cluster is seized scheduling and locked, and after the scheduling lock is successfully seized, the node for seizing scheduling lock is dispatched first
Corresponding task to be allocated distributes to each node in cluster according to default task allocation rule, notifies other in cluster
Node performs the task of distribution.By above-mentioned technical proposal, the node for performing task no longer competes task lock, but at any one
Node is seized after scheduling lock, and are dispatched into corresponding task to be allocated according to default the current time point first for reaching agreement
Business allocation rule distributes to each node in cluster, is come together execution task by each node, so as to solve task point
With unbalanced, and task lock expense it is big the problem of, reach and make it that task distribution is more balanced, the effect of reduction system burden.
Fig. 2 is a kind of method for allocating tasks according to embodiment illustrated in fig. 1, as shown in Fig. 2 above-mentioned step 102
Described in after the scheduling lock is successfully seized, the first corresponding task to be allocated of scheduling is distributed according to default task and advised
Each node in cluster is then distributed to, be may comprise steps of：
Step 1021a, the corresponding all tasks to be allocated of the first scheduling are distributed in cluster according to Principle of Average Allocation
Each node.
Exemplary, Principle of Average Allocation can be by the sum divided by collection of the corresponding all tasks to be allocated of the first scheduling
Node total number in group, calculates the number of tasks that each node should be distributed, and then appoints according to the number of tasks for each node distribution
Business.Or, average weighted mode can be used, for example, can set the weight of the stronger node of disposal ability is high by one
A bit, by the weight of the weaker node of disposal ability set it is lower, so as to the sum according to all tasks to be allocated, with
And the weight of each node is each node distribution task.
Step 1022a, the task to be allocated for distributing to each node is added in the task queue of each node, its
In, one task queue of each node correspondence.
Fig. 3 is another method for allocating tasks according to embodiment illustrated in fig. 1, as shown in Figure 3, it is contemplated that performs and appoints
May the also having last distribution in the task queue of each node of business of the task has been not carried out, and each node is remaining have been divided
Quantity with task is different, and is additionally contemplates that and has either distributed task, the new task that still will be distributed, each task institute
The time of consumption may be also different, therefore based on the consideration of some above-mentioned factor, institute is successfully being seized in above-mentioned step 102
State after scheduling lock, the corresponding task to be allocated of the first scheduling is distributed to according to default task allocation rule each in cluster
Node, may comprise steps of：
Step 1021b, obtains the task queue information of all nodes in cluster, and task queue information is included in cluster
Each node task queue in the distribution task that there is currently；Wherein, one task queue of each node correspondence.
In the present embodiment, the task of having distributed can be understood as being distributed during last distribution task for task, to be allocated
Business refers to the corresponding task of the first scheduling, will distribute to each node but also unallocated.Due to task execution when according to point
What the order with priority was performed, and because node needs to consume the regular hour in the task of execution, being not carried out for task
Meeting temporal persistence waits a upper tasks carrying just to be performed after finishing, therefore perform speed when node in task queue
In the case that degree is unable to catch up with the dispensing rate of task, there may be the task of last distribution when the task of distribution.Therefore, exist
When being node distribution task every time, it is also desirable to consider these time-consuming demands for having distributed task.
Step 1022b, it is determined that each having distributed required by task of the task takes in the task queue of each node.
Example, above-mentioned steps 1022b can include following three kinds of situations：
When the number of times m that first task was once performed be more than or equal to n when, obtain first task first n time execution when
The average time-consuming task as needed for first task of task takes, in task queue of the first task for any node in cluster
Any distributed task；
When the number of times m that first task was once performed is less than n, then first task being averaged when performing for first m times is obtained
The time-consuming task as needed for first task of task takes；
When first task is not carried out out-of-date, obtain that default task set in advance is time-consuming to be used as the task needed for first task
It is time-consuming.
Example, n can take 3, i.e., when first task was once performed, and the number of times performed is when being more than 3, then by this
The time that one task is spent when performing for nearest 3 times averages, and is consumed as the task needed for the first task of this prediction
When, when the number of times that first task was once performed is discontented 3 times, then performed several times, and just took this time spent several times to be averaged
It is worth and is taken as the task needed for the first task of this prediction, if be not carried out before, that is just appointed with default acquiescence
Business is time-consuming to be taken as its task.
Step 1023b, it is time-consuming according to each having distributed required by task of the task, when determining that the residue of all nodes performs total
Between, and each node the remaining execution time, residue, which is performed, to be all nodes to have distributed tasks carrying complete by all total time
Required total time, the remaining execution time of each node appoints all distributed in the task queue of oneself for each node
Business has performed the required time.
Example, it is assumed that there is node A, node B, node C and node D totally 4 nodes in cluster, its interior joint A's appoints
Task has been distributed present in business queue for task 1, task 2, task has been distributed present in node B task queue for task
3rd, task 4 and task 5, in the absence of task of having distributed in node C task queues, have been distributed present in node D task queue
Task is task 6.Then, all nodes it is remaining perform be equal to total time task 1, task 2, task 3, task 4, task 5 and
The task of task 6 takes sum, and the node A remaining execution time is equal to task 1, the task of task 2 and takes sum, node B's
The task that the remaining execution time is equal to task 3, task 4 and task 5 takes sum, and the node C remaining execution time is zero, node
4 remaining execution time is that the task of task 6 takes.
Step 1024b, determines appointing for each required by task to be allocated in the corresponding all tasks to be allocated of the first scheduling
Business is time-consuming.
Example, above-mentioned steps 1024b can include following three kinds of situations with step 1022b similarly：
When the number of times m that the second task was once performed be more than or equal to n when, obtain first task first n time execution when
The average time-consuming task as the second required by task of task takes, and second task is that the first scheduling is corresponding all to be allocated
Any task in business；
When the number of times m that the second task was once performed is less than n, then first task being averaged when performing for first m times is obtained
The time-consuming task as the second required by task of task takes；
When the second task is not carried out out-of-date, obtain that default task set in advance is time-consuming to be used as the task needed for first task
It is time-consuming.
Example, n can take 3, i.e., when the second task was once performed, and the number of times performed is when being more than 3, then by this
The time that two tasks are spent when performing for nearest 3 times averages, and is consumed as the task of the second required by task of this prediction
When, when the number of times that the second task was once performed is discontented 3 times, then performed several times, and just took this time spent several times to be averaged
It is worth and is taken as the task of the second required by task of this prediction, if be not carried out before, that is just appointed with default acquiescence
Business is time-consuming to be taken as its task.
In addition, in another implementation, when second task be not carried out it is out-of-date, by second task according to average mark
Any node in the cluster is distributed to principle.It can be understood as：By the corresponding all tasks to be allocated of the first scheduling
According to the processing different with the two classes progress being not carried out of executed, wherein for the task to be allocated of executed, by
Task when performing before be referred to takes, therefore holds several times before the task to be allocated can be calculated according to preceding method
The task that average task during row takes the required by task to be allocated when being distributed as this subtask takes；For what is be not carried out
Task to be allocated, due to can not be for reference history take, therefore the task to be allocated that these can be not carried out according to
Principle of Average Allocation distributes to each node in cluster, and wherein Principle of Average Allocation is referred to the method shown in Fig. 2, no longer
Repeat.
Step 1025b, it is time-consuming according to the task of each required by task to be allocated, determine total consumption of all tasks to be allocated
When.
Example, it is assumed that the first corresponding task to be allocated of scheduling is 6~task of task 15, then first obtain task 6~appoint
The task of each required by task takes in business 15, then the task of 6~task of calculating task 15 takes sum, is the total time-consuming.
Step 1026b, total time is performed according to residue, and the total time-consuming of all tasks to be allocated determines that all nodes are held
The total time of row required by task.
Wherein it is possible to remaining perform calculated in step 1023b is designated as lv total time, can be by step 1025b
The total time-consuming of all tasks to be allocated calculated is designated as wv, then all nodes perform the total time T of required by tasktotal=lv+
wv.I.e. all nodes are performed has distributed task and the required by task that will be distributed the total time of required by task for currently all
Task takes sum.
Step 1027b, is defined as the task time of each node distribution according to the number of nodes of the total time and cluster.
Example, be each node distribution task time can use mean allocation principle, you can with by this it is total when
Between TtotalDivided by the number of nodes of cluster, the task time of each node distribution is obtained, av can be designated as.
Step 1028b, according to the task time for each node distribution, remaining execution time of each node and each
The task of required by task to be allocated is taken as each node distribution task to be allocated.
Wherein, the remaining execution time of each node can be designated as nlv, due to being wrapped in the task time of each node distribution
It is contained and performs in oneself task queue all having distributed required by task of the tasks and taken, with the task time of each node distribution
This part-time is subtracted, the remaining time is exactly the time for being used to perform will distributing for task that can be to the node distribution,
Task time to be allocated can be referred to as, fv is designated as, then fv=av-nlv.Then according to the fv of each node, and treat point
The task for each required by task to be allocated matched somebody with somebody takes, you can it is determined that what is matched with the task time to be allocated of each node treats
Distribution task.
Example, by taking first node as an example, first node is any node in cluster, and above-mentioned steps 1028b can include
Following steps：
First, the remaining execution time of first node will be subtracted for the task time that first node is distributed, first segment is obtained
The task time to be allocated of point；
Secondly, it is time-consuming according to the task of each required by task to be allocated, it is defined as the task to be allocated of first node distribution,
So that the task of all required by task to be allocated of first node distribution takes the task time to be allocated of sum and first node
Matching, it can be understood as to the time-consuming sum of task of all required by task to be allocated of first node distribution and treating for the first node
Distribution task time is equal, or takes sum and the first segment to the task that first node distributes all required by task to be allocated
The error of the task time to be allocated of point is less than certain threshold value.
Step 1029b, the task to be allocated for distributing to each node is added in the task queue of each node.
In addition, it is noted that can also be former by the method shown in above-mentioned step 1021b-1029b and mean allocation
Then it is combined, for example, for all task of distribution and task to be allocated, if once performing for task, can adopt
It is allocated with the method shown in step 1021b-1029b, the task that residue was not carried out can use Principle of Average Allocation
It is allocated, Principle of Average Allocation refers to step 1021a-1022a, repeats no more.
Fig. 4 is a kind of block diagram of cluster task balance dispatching device according to an exemplary embodiment, such as Fig. 4 institutes
Show, applied to any node in cluster, the cluster task balance dispatching device 400 includes：
Module 410 is monitored in scheduling, for when listening to first and being dispatched to of about fixed time point, seizing scheduling lock；
Task allocating module 420, for after scheduling lock is successfully seized, by the first corresponding task to be allocated of scheduling according to
Default task allocation rule distributes to each node in cluster；
Task notifications module 430, for notifying other nodes in cluster to perform the task of distribution.
Optionally, Fig. 5 is a kind of block diagram of task allocating module according to embodiment illustrated in fig. 4, as shown in figure 5,
The task allocating module 420, can include：
First distribution sub module 4201, for dispatching corresponding all tasks to be allocated according to Principle of Average Allocation by first
Distribute to each node in cluster；
Queue management submodule 4202, for the task to be allocated for distributing to each node to be added into appointing for each node
It is engaged in queue, wherein, one task queue of each node correspondence.
Or, in another implementation, Fig. 6 is another task distribution mould according to embodiment illustrated in fig. 4
The block diagram of block, as shown in fig. 6, the task allocating module 420, can include：
Queuing message acquisition submodule 4203, the task queue information for obtaining all nodes in cluster, task team
Column information includes the distribution task that there is currently in the task queue of each node in cluster；Wherein, each node pair
Answer a task queue；
Task takes the first determination sub-module 4204, and each task has been distributed in the task queue for determining each node
Required task takes；
Remaining time determination sub-module 4205, for being taken according to each having distributed required by task of the task, it is determined that all
The remaining of node performs total time, and each node the remaining execution time, it is all nodes by institute that residue, which performs total time,
Have and distributed tasks carrying complete required total time, remaining execution time of each node is each node by the task team of oneself
All in row have distributed tasks carrying the complete required time；
Task takes the second determination sub-module 4206, every in the corresponding all tasks to be allocated of the first scheduling for determining
The task of individual required by task to be allocated takes；
Task total time-consuming determination sub-module 4207, for being taken according to the task of each required by task to be allocated, determines institute
There is the total time-consuming of task to be allocated；
Node total time determining module 4208, for performing total time according to residue, and all tasks to be allocated is total
It is time-consuming to determine the total time that all nodes perform required by task；
Node tasks time distribution sub module 4209, for being defined as each section according to the number of nodes of total time and cluster
The task time of point distribution；
Second distribution sub module 4210, for being held according to the task time for each node distribution, the residue of each node
The task of row time and each required by task to be allocated is taken as each node distribution task to be allocated；
Queue management submodule 4211, for the task to be allocated for distributing to each node to be added into appointing for each node
It is engaged in queue.
Optionally, task takes the first determination sub-module 4204, is used for：
When the number of times m that first task was once performed be more than or equal to n when, obtain first task first n time execution when
The average time-consuming task as needed for first task of task takes, in task queue of the first task for any node in cluster
Any distributed task；
When the number of times m that first task was once performed is less than n, then first task being averaged when performing for first m times is obtained
The time-consuming task as needed for first task of task takes；
When first task is not carried out out-of-date, obtain that default task set in advance is time-consuming to be used as the task needed for first task
It is time-consuming.
Optionally, task takes the second determination sub-module 4206, is used for：
When the number of times m that the second task was once performed is more than or equal to n, second task is obtained when performing for first n times
The time-consuming task as second required by task of average task take, second task first dispatches corresponding needed by this
Any task to be allocated in distribution task；
When the number of times m that second task was once performed be less than n when, then obtain second task first m time execution when
The average time-consuming task as second required by task of task takes；
When second task is not carried out out-of-date, obtain that default task set in advance is time-consuming to be used as second required by task
Task takes, or, when second task is not carried out out-of-date, second task is distributed into the cluster according to Principle of Average Allocation
In any node.
Optionally, the second distribution sub module 4210, is used for：
The remaining execution time of first node will be subtracted for the task time that first node is distributed, obtain treating for first node
Distribute task time, the first node is any node in cluster；
It is time-consuming according to the task of each required by task to be allocated, it is defined as the task to be allocated of first node distribution so that
The task of all required by task to be allocated of first node distribution takes sum and matched with the task time to be allocated of first node.
The cluster task balance dispatching device that the disclosure is provided, when listening to first and being dispatched to of about fixed time point,
Each node in cluster is seized scheduling and locked, and after the scheduling lock is successfully seized, the node for seizing scheduling lock is dispatched first
Corresponding task to be allocated distributes to each node in cluster according to default task allocation rule, notifies other in cluster
Node performs the task of first node distribution.By above-mentioned technical proposal, the node for performing task no longer competes task lock, but
After any one node seizes scheduling lock, corresponding task to be allocated currently will be dispatched according to pre- of about fixed time point first
If task allocation rule distribute to each node in cluster, come together execution task by each node, so as to solve
Task distribution it is unbalanced, and task lock expense it is big the problem of, reach cause task distribution it is more balanced, reduction system burden effect
Really.
Fig. 7 is the block diagram of a kind of electronic equipment 700 according to an exemplary embodiment.As shown in fig. 7, the electronics is set
Standby 700 can include：Processor 701, memory 702, multimedia groupware 703, input/output (I/O) interface 704, Yi Jitong
Believe component 705.
Wherein, processor 701 is used for the integrated operation for controlling the electronic equipment 700, equal with the cluster task for completing above-mentioned
All or part of step in weighing apparatus dispatching method.Memory 702 is used to store various types of data to support to set in the electronics
Standby 700 operation, these data can for example include any application program or method for being used to operate on the electronic equipment 700
Instruction, and related data of application program, such as contact data, the message of transmitting-receiving, picture, audio, video etc..Should
Memory 702 can be realized by any kind of volatibility or non-volatile memory device or combinations thereof, such as static
Random access memory (Static Random Access Memory, abbreviation SRAM), Electrically Erasable Read Only Memory
(Electrically Erasable Programmable Read-Only Memory, abbreviation EEPROM), erasable programmable
Read-only storage (Erasable Programmable Read-Only Memory, abbreviation EPROM), programmable read only memory
(Programmable Read-Only Memory, abbreviation PROM), and read-only storage (Read-Only Memory, referred to as
ROM), magnetic memory, flash memory, disk or CD.Multimedia groupware 703 can include screen and audio-frequency assembly.Wherein
Screen for example can be touch-screen, and audio-frequency assembly is used to export and/or input audio signal.For example, audio-frequency assembly can include
One microphone, microphone is used to receive external audio signal.The audio signal received can be further stored in storage
Device 702 is sent by communication component 705.Audio-frequency assembly also includes at least one loudspeaker, for exports audio signal.I/O
Interface 704 is provides interface between processor 701 and other interface modules, other above-mentioned interface modules can be keyboard, mouse,
Button etc..These buttons can be virtual push button or entity button.Communication component 705 is used for the electronic equipment 700 and other
Wired or wireless communication is carried out between equipment.Radio communication, such as Wi-Fi, bluetooth, near-field communication (Near Field
Communication, abbreviation NFC), 2G, 3G or 4G, or one or more of combinations in them, therefore the corresponding communication
Component 705 can include：Wi-Fi module, bluetooth module, NFC module.
In one exemplary embodiment, electronic equipment 700 can be by one or more application specific integrated circuits
(Application Specific Integrated Circuit, abbreviation ASIC), digital signal processor (Digital
Signal Processor, abbreviation DSP), digital signal processing appts (Digital Signal Processing Device,
Abbreviation DSPD), PLD (Programmable Logic Device, abbreviation PLD), field programmable gate array
(Field Programmable Gate Array, abbreviation FPGA), controller, microcontroller, microprocessor or other electronics member
Part realization, the cluster task equalization scheduling method above-mentioned for performing.
In a further exemplary embodiment, a kind of computer-readable recording medium including programmed instruction, example are additionally provided
Such as include the memory 702 of programmed instruction, said procedure instruction can be performed with completion by the processor 701 of electronic equipment 700
The cluster task equalization scheduling method stated.
Fig. 8 is the block diagram of another electronic equipment 800 according to an exemplary embodiment.For example, electronic equipment 800
It may be provided in a server.Reference picture 8, electronic equipment 800 includes processor 822, its quantity can be one or more,
And memory 832, for storing the computer program that can be performed by processor 822.The computer journey stored in memory 832
Sequence can include it is one or more each correspond to the module of one group of instruction.In addition, processor 822 can be configured
To perform the computer program, to perform above-mentioned cluster task equalization scheduling method.
In addition, electronic equipment 800 can also include power supply module 826 and communication component 850, the power supply module 826 can be with
It is configured as performing the power management of electronic equipment 800, the communication component 850 can be configured as realizing electronic equipment 800
Communication, for example, wired or wireless communication.In addition, the electronic equipment 800 can also include input/output (I/O) interface 858.Electricity
Sub- equipment 800 can be operated based on the operating system for being stored in memory 832, such as Windows ServerTM, Mac OS
XTM, UnixTM, LinuxTM etc..
In a further exemplary embodiment, a kind of computer-readable recording medium including programmed instruction, example are additionally provided
Such as include the memory 832 of programmed instruction, said procedure instruction can be performed with completion by the processor 822 of electronic equipment 800
The cluster task equalization scheduling method stated.
The preferred embodiment of the disclosure is described in detail above in association with accompanying drawing, still, the disclosure is not limited to above-mentioned reality
The detail in mode is applied, in the range of the technology design of the disclosure, a variety of letters can be carried out with technical scheme of this disclosure
Monotropic type, these simple variants belong to the protection domain of the disclosure.
It is further to note that each particular technique feature described in above-mentioned embodiment, in not lance
In the case of shield, can be combined by any suitable means, in order to avoid unnecessary repetition, the disclosure to it is various can
The combination of energy no longer separately illustrates.
In addition, can also be combined between a variety of embodiments of the disclosure, as long as it is without prejudice to originally
Disclosed thought, it should equally be considered as disclosure disclosure of that.
Claims (10)
1. a kind of cluster task equalization scheduling method, it is characterised in that applied to any node in cluster, methods described bag
Include：
When listening to first and being dispatched to of about fixed time point, scheduling lock is seized；
After the scheduling lock is successfully seized, the corresponding task to be allocated of the described first scheduling is distributed according to default task and advised
Then distribute to each node in the cluster；
Other nodes in the cluster are notified to perform the task of distribution.
2. according to the method described in claim 1, it is characterised in that described to press the corresponding task to be allocated of the described first scheduling
Each node in the cluster is distributed to according to default task allocation rule, including：
The corresponding all tasks to be allocated of described first scheduling are distributed to according to Principle of Average Allocation each in the cluster
Node；
In the task queue that the task to be allocated for distributing to each node is added to each node, wherein, it is described
One task queue of each node correspondence.
3. according to the method described in claim 1, it is characterised in that described to press the corresponding task to be allocated of the described first scheduling
Each node in the cluster is distributed to according to default task allocation rule, including：
The task queue information of all nodes in the cluster is obtained, task queue information includes each in the cluster
The distribution task that there is currently in the task queue of node；Wherein, one task queue of each node correspondence；
Determine that each having distributed required by task of the task takes in the task queue of each node；
It is time-consuming according to each having distributed required by task of the task, determine that the remaining of all nodes performs total time, with
And the remaining execution time of each node, it is described remaining to perform total time to be all nodes described distributed all
Tasks carrying complete required total time, remaining execution time of each node is each node by the task team of oneself
All in row described have distributed tasks carrying the complete required time；
Determine that the task of each required by task to be allocated in the corresponding all tasks to be allocated of first scheduling takes；
Taken according to the task of each required by task to be allocated, determine the total time-consuming of all tasks to be allocated；
Determine that all nodes are performed according to the total time-consuming of the remaining execution total time, and all tasks to be allocated
The total time of required by task；
It is defined as the task time of each node distribution according to the number of nodes of the total time and the cluster；
Described treat according to the task time for each node distribution, the remaining execution time of each node and each
The task of distribution required by task is taken as each node distribution task to be allocated；
In the task queue that the task to be allocated for distributing to each node is added to each node.
4. method according to claim 3, it is characterised in that each in the task queue of determination each node
Having distributed required by task of the task takes, including：
When the number of times m that first task was once performed is more than or equal to n, the first task is obtained in first n time execution
The average time-consuming task as needed for the first task of task takes, and the first task is any node in the cluster
Task queue in any described distributed task；
When the number of times m that the first task was once performed be less than n when, then obtain the first task first m time execution when
The average time-consuming task as needed for the first task of task takes；
When the first task is not carried out out-of-date, default task set in advance is obtained time-consuming as needed for the first task
Task takes.
5. method according to claim 3, it is characterised in that determination first scheduling is corresponding all to be allocated
The task of each required by task to be allocated in task takes, including：
When the number of times m that the second task was once performed is more than or equal to n, second task is obtained in first n time execution
The average time-consuming task as second required by task of task takes, and second task is the corresponding institute of the described first scheduling
There is any task to be allocated in task to be allocated；
When the number of times m that second task was once performed be less than n when, then obtain second task first m time execution when
The average time-consuming task as second required by task of task takes；
When second task is not carried out out-of-date, obtain that default task set in advance is time-consuming to be used as second required by task
Task takes, or, when second task is not carried out out-of-date, second task is distributed into institute according to Principle of Average Allocation
State any node in cluster.
6. method according to claim 3, it is characterised in that when the basis is the task of each node distribution
Between, the task of the remaining execution time of each node and each required by task to be allocated taken as each section
Point distribution task to be allocated, including：
The remaining execution time of the first node will be subtracted for the task time that first node is distributed, obtain the first node
Task time to be allocated, the first node be the cluster in any node；
Taken according to the task of each required by task to be allocated, be defined as the task to be allocated of the first node distribution,
So that the task of all required by task to be allocated of first node distribution takes the to be allocated of sum and the first node
Task time matches.
7. a kind of cluster task balance dispatching device, it is characterised in that applied to any node in cluster, described device bag
Include：
Module is monitored in scheduling, for when listening to first and being dispatched to of about fixed time point, seizing scheduling lock；
Task allocating module, for after the scheduling lock is successfully seized, the corresponding task to be allocated of the described first scheduling to be pressed
Each node in the cluster is distributed to according to default task allocation rule；
Task notifications module, for notifying other nodes in the cluster to perform the task of distribution.
8. device according to claim 7, it is characterised in that the task allocating module, including：
Queuing message acquisition submodule, the task queue information for obtaining all nodes in the cluster, task queue letter
Breath includes the distribution task that there is currently in the task queue of each node in the cluster；Wherein, each section
Point one task queue of correspondence；
Task takes the first determination sub-module, each described in the task queue for determining each node to have distributed task
Required task takes；
Remaining time determination sub-module, for being taken according to each having distributed required by task of the task, is determined described all
The remaining of node performs total time, and each node the remaining execution time, it is described that remaining to perform total time be described
All nodes by it is all it is described distributed tasks carrying complete required total time, the remaining execution time of each node is institute
Stating each node described has distributed tasks carrying the complete required time all in the task queue of oneself；
Task takes the second determination sub-module, for determining each treating in the corresponding all tasks to be allocated of first scheduling
The distributing required by task of the task takes；
Task total time-consuming determination sub-module, for being taken according to the task of each required by task to be allocated, determines the institute
There is the total time-consuming of task to be allocated；
Node total time determining module, for according to it is described it is remaining perform total time, and all tasks to be allocated is total
It is time-consuming to determine the total time that all nodes perform required by task；
Node tasks time distribution sub module, it is described every for being defined as according to the number of nodes of the total time and the cluster
The task time of individual node distribution；
Second distribution sub module, for being held according to the task time for each node distribution, the residue of each node
The task of row time and each required by task to be allocated is taken as each node distribution task to be allocated；
Queue management submodule, for the task to be allocated for distributing to each node to be added into appointing for each node
It is engaged in queue.
9. a kind of computer-readable recording medium, is stored thereon with computer program, it is characterised in that the computer program quilt
The step of method any one of claim 1-6 is realized during computing device.
10. a kind of electronic equipment, it is characterised in that including：
Computer-readable recording medium described in claim 9；And
One or more processor, for performing the computer program in the computer-readable recording medium.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710330278.9A CN107229511B (en) | 2017-05-11 | 2017-05-11 | Cluster task balanced scheduling method and device, storage medium and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710330278.9A CN107229511B (en) | 2017-05-11 | 2017-05-11 | Cluster task balanced scheduling method and device, storage medium and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107229511A true CN107229511A (en) | 2017-10-03 |
| CN107229511B CN107229511B (en) | 2020-07-03 |
Family
ID=59934347
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710330278.9A Active CN107229511B (en) | 2017-05-11 | 2017-05-11 | Cluster task balanced scheduling method and device, storage medium and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107229511B (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109062681A (en) * | 2018-08-24 | 2018-12-21 | 郑州云海信息技术有限公司 | A kind of execution method, system, device and the storage medium of periodic cycle task |
| CN109343941A (en) * | 2018-08-14 | 2019-02-15 | 阿里巴巴集团控股有限公司 | Task processing method, device, electronic equipment and computer readable storage medium |
| CN110275770A (en) * | 2018-03-15 | 2019-09-24 | 阿里巴巴集团控股有限公司 | Task balance dispatching method, system, node and electronic equipment |
| CN110618855A (en) * | 2018-12-25 | 2019-12-27 | 北京时光荏苒科技有限公司 | Task allocation method and device, electronic equipment and storage medium |
| CN110633135A (en) * | 2019-08-15 | 2019-12-31 | 中国平安财产保险股份有限公司 | Asynchronous task allocation method and device, computer equipment and storage medium |
| CN111782626A (en) * | 2020-08-14 | 2020-10-16 | 工银科技有限公司 | Task allocation method and device, distributed system, electronic device and medium |
| CN112395075A (en) * | 2019-08-15 | 2021-02-23 | 阿里巴巴集团控股有限公司 | Resource processing method and device and resource scheduling system |
| CN112948106A (en) * | 2020-09-07 | 2021-06-11 | 深圳市明源云科技有限公司 | Task allocation method and device |
| CN113256125A (en) * | 2021-05-28 | 2021-08-13 | 西安交通大学 | Heterogeneous robot cluster task allocation method, system, equipment and storage medium |
| CN113722060A (en) * | 2021-07-30 | 2021-11-30 | 浪潮电子信息产业股份有限公司 | Alarm task management method, device and related components |
| CN113934535A (en) * | 2021-10-11 | 2022-01-14 | 广东科诺勘测工程有限公司 | Mass point cloud data processing method, device, server and system |
| CN114780217A (en) * | 2022-04-19 | 2022-07-22 | 湖南长银五八消费金融股份有限公司 | Task scheduling method and device, computer equipment and medium |
| CN116723225A (en) * | 2023-06-16 | 2023-09-08 | 广州银汉科技有限公司 | Automatic allocation method and system for game tasks |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103793272A (en) * | 2013-12-27 | 2014-05-14 | 北京天融信软件有限公司 | Periodical task scheduling method and periodical task scheduling system |
| US20150205633A1 (en) * | 2013-05-24 | 2015-07-23 | Google Inc. | Task management in single-threaded environments |
| CN105100259A (en) * | 2015-08-18 | 2015-11-25 | 北京京东尚科信息技术有限公司 | Distributed timed task execution method and system |
| CN105700943A (en) * | 2015-12-31 | 2016-06-22 | 上海百事通信息技术股份有限公司 | Distributed task scheduling method and system |
| WO2016205845A1 (en) * | 2015-06-23 | 2016-12-29 | Mahorka Diethard | Device, method, and computer program product for secure data communication |
-
2017
- 2017-05-11 CN CN201710330278.9A patent/CN107229511B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150205633A1 (en) * | 2013-05-24 | 2015-07-23 | Google Inc. | Task management in single-threaded environments |
| CN103793272A (en) * | 2013-12-27 | 2014-05-14 | 北京天融信软件有限公司 | Periodical task scheduling method and periodical task scheduling system |
| WO2016205845A1 (en) * | 2015-06-23 | 2016-12-29 | Mahorka Diethard | Device, method, and computer program product for secure data communication |
| CN105100259A (en) * | 2015-08-18 | 2015-11-25 | 北京京东尚科信息技术有限公司 | Distributed timed task execution method and system |
| CN105700943A (en) * | 2015-12-31 | 2016-06-22 | 上海百事通信息技术股份有限公司 | Distributed task scheduling method and system |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110275770A (en) * | 2018-03-15 | 2019-09-24 | 阿里巴巴集团控股有限公司 | Task balance dispatching method, system, node and electronic equipment |
| CN110275770B (en) * | 2018-03-15 | 2023-09-22 | 阿里巴巴集团控股有限公司 | Task balanced scheduling method, system, node and electronic equipment |
| CN109343941B (en) * | 2018-08-14 | 2023-02-21 | 创新先进技术有限公司 | Task processing method and device, electronic equipment and computer readable storage medium |
| CN109343941A (en) * | 2018-08-14 | 2019-02-15 | 阿里巴巴集团控股有限公司 | Task processing method, device, electronic equipment and computer readable storage medium |
| CN109062681A (en) * | 2018-08-24 | 2018-12-21 | 郑州云海信息技术有限公司 | A kind of execution method, system, device and the storage medium of periodic cycle task |
| CN110618855A (en) * | 2018-12-25 | 2019-12-27 | 北京时光荏苒科技有限公司 | Task allocation method and device, electronic equipment and storage medium |
| CN110633135A (en) * | 2019-08-15 | 2019-12-31 | 中国平安财产保险股份有限公司 | Asynchronous task allocation method and device, computer equipment and storage medium |
| CN112395075A (en) * | 2019-08-15 | 2021-02-23 | 阿里巴巴集团控股有限公司 | Resource processing method and device and resource scheduling system |
| CN111782626A (en) * | 2020-08-14 | 2020-10-16 | 工银科技有限公司 | Task allocation method and device, distributed system, electronic device and medium |
| CN112948106A (en) * | 2020-09-07 | 2021-06-11 | 深圳市明源云科技有限公司 | Task allocation method and device |
| CN112948106B (en) * | 2020-09-07 | 2024-05-31 | 深圳市明源云科技有限公司 | Task allocation method and device |
| CN113256125A (en) * | 2021-05-28 | 2021-08-13 | 西安交通大学 | Heterogeneous robot cluster task allocation method, system, equipment and storage medium |
| CN113256125B (en) * | 2021-05-28 | 2023-06-30 | 西安交通大学 | Heterogeneous robot cluster task allocation method, system, equipment and storage medium |
| CN113722060A (en) * | 2021-07-30 | 2021-11-30 | 浪潮电子信息产业股份有限公司 | Alarm task management method, device and related components |
| CN113934535A (en) * | 2021-10-11 | 2022-01-14 | 广东科诺勘测工程有限公司 | Mass point cloud data processing method, device, server and system |
| CN114780217A (en) * | 2022-04-19 | 2022-07-22 | 湖南长银五八消费金融股份有限公司 | Task scheduling method and device, computer equipment and medium |
| CN116723225A (en) * | 2023-06-16 | 2023-09-08 | 广州银汉科技有限公司 | Automatic allocation method and system for game tasks |
| CN116723225B (en) * | 2023-06-16 | 2024-05-17 | 广州银汉科技有限公司 | Automatic allocation method and system for game tasks |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107229511B (en) | 2020-07-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107229511A (en) | Cluster task equalization scheduling method, device, storage medium and electronic equipment | |
| Chen et al. | Resource sharing of a computing access point for multi-user mobile cloud offloading with delay constraints | |
| US11582166B2 (en) | Systems and methods for provision of a guaranteed batch | |
| US8276143B2 (en) | Dynamic scheduling of application tasks in a distributed task based system | |
| CN104243405B (en) | A kind of request processing method, apparatus and system | |
| CN105335229B (en) | Scheduling method and device of service resources | |
| CN108563493A (en) | A kind of event-driven method, device, equipment and readable storage medium storing program for executing | |
| KR20130127508A (en) | Method and apparatus of smart power management for mobile communication terminals using power thresholds | |
| CN105589750A (en) | CPU (Central Processing Unit) resource scheduling method and server | |
| US8799474B2 (en) | Apparatus and method to allocate limited resources | |
| Liu et al. | Security-aware resource allocation for mobile cloud computing systems | |
| CN105637841B (en) | The data classification synchronous for adaptability | |
| CN107506240A (en) | Background application management-control method, device, storage medium and electronic equipment | |
| CN106790092A (en) | Remote procedure call services end control system and method | |
| Saha et al. | Exploring the fairness and resource distribution in an apache mesos environment | |
| CN110837401A (en) | Hierarchical processing method and device for java thread pool | |
| CN109471711A (en) | A kind of method and device of task processing | |
| CN108563200A (en) | A kind of Job Scheduling method and device of the multiple target based on ant group algorithm | |
| CN116185623A (en) | Task allocation method and device, electronic equipment and storage medium | |
| Aldhalaan et al. | Near-optimal allocation of vms from iaas providers by saas providers | |
| CN108900607A (en) | A kind of processing method, device and the server of SMB agreement request | |
| CN104636205B (en) | A kind of method and apparatus that task is seized | |
| CN109426561A (en) | A kind of task processing method, device and equipment | |
| CN105474173B (en) | For reducing the parallel network application scheduling of power consumption | |
| CN107563732A (en) | A kind of method, apparatus and electric terminal for controlling locker |
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 |