CN109597685A

CN109597685A - Method for allocating tasks, device and server

Info

Publication number: CN109597685A
Application number: CN201811157251.5A
Authority: CN
Inventors: 吴轲
Original assignee: Alibaba Group Holding Ltd
Current assignee: Advanced New Technologies Co Ltd; Advantageous New Technologies Co Ltd
Priority date: 2018-09-30
Filing date: 2018-09-30
Publication date: 2019-04-09
Anticipated expiration: 2038-09-30
Also published as: CN109597685B

Abstract

Present description provides a kind of method for allocating tasks, device and servers.Wherein, which includes: to obtain the operation data of multiple processing nodes in first time period, the use data of task processing data and relied on resource；According to above-mentioned data, the task bearing capacity of target processing node is determined；To target processing node distribution second time period with the matched data processing task of task bearing capacity.In this specification embodiment, by obtain and using handle in first time period the relied on resource of node using data, handle the operation data and task processing data of node, subtly determine the specific tasks bearing capacity of processing node, the data processing task of second time period is distributed to processing node further according to task bearing capacity, so as to combination processing node, and handle the concrete condition for the system resource that node is relied on, accurately, task reasonably is distributed for processing node, the utilization rate of resource is improved, so that data processing is efficient, stablizes.

Description

Method for allocating tasks, device and server

Technical field

This specification belongs to Internet technical field more particularly to a kind of method for allocating tasks, device and server.

Background technique

In distributed data processing system, it will usually multiple processing nodes be distributed with.Wherein, above-mentioned processing node difference It is responsible for the data processing task of processing system distribution, so as to carry out batch processing to data processing task to be processed.

Existing method for allocating tasks is to carry out pressure test to the processing capacity of the processing node in system in advance mostly, According to test result determine processing node processing speed (such as processing node completion per second data processing task number Amount) and discount ratio；Further according to the processing speed and discount ratio of above-mentioned processing node, pinned task distribution is calculated Amount, so can according to above-mentioned fixation task sendout at regular intervals, to processing node distribute at data to be processed Reason task.

When by the data processing task of above method allocation processing node, often can only according to based on static test institute really Fixed pinned task sendout distributes data processing task to be processed for processing node.However specifically carrying out data processing When, resource service condition, the running environment of system entirety, and the own situation etc. of processing node is all in dynamic change (such as resource workable for current system tail off or system in certain several processing node temporarily can not work normally Deng), above-mentioned variation can all impact the processing capacity of the processing specific data processing task of node.Therefore, according to existing Method according to fixed task sendout based on determined by static test come the data processing task of allocation processing node, often It is not accurate enough, reasonable, it is easy to lead to not effectively utilize system resource there are error, or even the shakiness for causing system to run It is fixed.Therefore, a kind of more accurate, reasonable method for allocating tasks is needed, accurately data processing can be distributed for processing node and appoint Business, thus the utilization rate of the resource improved, so that data processing is efficient, stablizes.

Summary of the invention

This specification is designed to provide a kind of method for allocating tasks, device and server, to combine system and processing The concrete condition of node accurately, reasonably distributes task for processing node, thus the utilization rate of the resource improved, so that data Processing is efficient, stablizes.

A kind of method for allocating tasks, device and the server that this specification provides are achieved in that

A kind of method for allocating tasks, comprising: obtain the operation data of multiple processing nodes, task processing in first time period The use data of data and relied on resource；At the operation data of processing nodes multiple in the first time period, task The use data for managing data and relied on resource determine the task bearing capacity of target processing node；Node is handled to the target Distribute second time period with the matched data processing task of task bearing capacity.

A kind of task allocation apparatus, comprising: module is obtained, for obtaining the operation of multiple processing nodes in first time period The use data of data, task processing data and relied on resource；Determining module, for according to multiple in the first time period The operation data of node, the use data of task processing data and relied on resource are handled, determines the task of target processing node Bearing capacity；Distribution module, for distributing the matched with the task bearing capacity of second time period to target processing node Data processing task.

A kind of server, including processor and for the memory of storage processor executable instruction, the processor It is realized when executing described instruction and obtains the operation data, task processing data of multiple processing nodes in first time period and relied on The use data of resource；According in the first time period it is multiple processing nodes operation datas, task processing data and institute according to The use data for relying resource determine the task bearing capacity of target processing node；Distributed for the second time to target processing node Section with the matched data processing task of task bearing capacity.

A kind of computer readable storage medium, is stored thereon with computer instruction, and described instruction is performed realization and obtains The operation data of multiple processing nodes, the use data of task processing data and relied on resource in first time period；According to institute The operation data of multiple processing nodes in first time period, the use data of task processing data and relied on resource are stated, is determined The task bearing capacity of target processing node；To the target processing node distribution second time period with the task bearing capacity The data processing task matched.

A kind of method for allocating tasks, device and the server that this specification provides, by obtaining and utilizing first time period In it is multiple processing the relied on resources of node using data, handle node operation data and task processing data, subtly The specific tasks bearing capacity for determining processing node distributes the data of second time period further according to task bearing capacity to processing node Processing task so as to the service condition of system entirety resource during integrated data processing and handles the specific of node itself The factors such as situation, the data processing task accurately, being reasonably adapted for processing node distribution, to improve the utilization rate of resource, Enable processing node efficiently, steadily carry out data processing.

Detailed description of the invention

In order to illustrate more clearly of this specification embodiment or technical solution in the prior art, below will to embodiment or Attached drawing needed to be used in the description of the prior art is briefly described, it should be apparent that, the accompanying drawings in the following description is only The some embodiments recorded in this specification, for those of ordinary skill in the art, in not making the creative labor property Under the premise of, it is also possible to obtain other drawings based on these drawings.

Fig. 1 is a kind of embodiment of the structure composition of the system of the method for allocating tasks provided using this specification embodiment Schematic diagram；

Fig. 2 is in a Sample Scenario, using the method for allocating tasks allocation processing section of this specification embodiment offer A kind of schematic diagram of embodiment of the data processing task of point；

Fig. 3 is in a Sample Scenario, using the method for allocating tasks allocation processing of this specification embodiment offer A kind of schematic diagram of embodiment of the data processing task of node；

Fig. 4 is a kind of schematic diagram of embodiment of the process for the method for allocating tasks that this specification embodiment provides；

Fig. 5 is a kind of schematic diagram of embodiment of the structure for the server that this specification embodiment provides；

Fig. 6 is a kind of schematic diagram of embodiment of the structure for the task allocation apparatus that this specification embodiment provides.

Specific embodiment

In order to make those skilled in the art more fully understand the technical solution in this specification, below in conjunction with this explanation Attached drawing in book embodiment is clearly and completely described the technical solution in this specification embodiment, it is clear that described Embodiment be only this specification a part of the embodiment, instead of all the embodiments.The embodiment of base in this manual, Every other embodiment obtained by those of ordinary skill in the art without making creative efforts, all should belong to The range of this specification protection.

In view of existing method for allocating tasks is when determining the task amount for distributing to processing node, often do not account for (such as the service condition of system resource, handles the operating condition of node, and processing node is tired for system and processing node concrete condition Untreated task amount of product etc.) influence to the processing capacity of processing node, but source is determined by way of static test The processing speed of node is managed, i.e., pressure test is carried out to the processing capacity of processing node in advance, source is determined according to test result The processing speed of node is managed, and then determines fixed task sendout, the task sendout according still further to above-mentioned fixation is processing Node distributes each processing task.

But the processing actual processing capacity of node will receive system environments, used system locating for processing node Resource, and the influence of processing node situation itself.For example, shake, which occurs, for the downstream treatment systems that processing node is relied on to drop The processing of reduction process node can wait.Therefore, it is relied on based on existing method and passes through pinned task sendout determined by static test Data processing task is distributed, it is often not accurate enough, lead to not effectively utilize system resource, or even cause the shakiness of system It is fixed.

For above situation, this specification proposes that available and utilization is able to reflect out system resource (i.e. multiple processing The resource relied on) service condition and handle node itself concrete condition (including handle node operating condition and upper one The task disposition of a period) data, determine that current system resource and place can be suitble to further according to above-mentioned data The task bearing capacity for managing node situation reasonably gives processing node distribution data processing task according still further to above-mentioned task bearing capacity. So as to accurately, reasonably distribute task for processing node, improve resource according to the concrete condition of system and processing node Utilization rate, enable processing node efficiently, steadily carry out data processing.

This specification embodiment provides a kind of method for allocating tasks, and the method for allocating tasks can be applied to wrap Include: dispatch server, sensing module, processor cluster system architecture in.

It can be shown refering to fig. 1.The processor cluster be specifically as follows include multiple processors set.Wherein, Each of processor cluster processor can be understood as a processing node, for handling the number to be processed distributed According to processing task.The sensing module specifically can be set in above system framework, for (such as every every a period Every 30 seconds) resource that the processor in cluster relied on (such as CPU, the number of system entirety was handled in acquisition upper period According to resources such as libraries) use data, and processing cluster in each processor individual operation data and task processing data, and Above-mentioned data are sent to dispatch server.The dispatch service implement body can be used for a received upper period Above-mentioned data handled, analyzed, determine the whole resource of current system and the carrying out practically feelings of processor individual Condition；And based on the above situation, in the case where taking into account the requirement of resource utilization and operation stability, next period processing is calculated The task bearing capacity (the maximum task amount for the data processing task that i.e. next period processor can be handled) of device, further according to Above-mentioned task bearing capacity, distributes the data processing task of next period reasonably for processor.

In the present embodiment, the server can have data operation, storage and network interaction function for one Electronic equipment；Or run in the electronic equipment, it is that data processing, storage and network interaction provide support Software.Do not limit the quantity of the server specifically in the present embodiment.The server can be a server, It can be several servers, alternatively, the server cluster formed by several servers.

In the present embodiment, the processor can interpret computer instruction for one kind, and can be carried out specific The integrated circuit modules of data processing.For example, above-mentioned processor can be central processing unit (CPU) etc..By multiple above-mentioned processing Device combination, available above-mentioned processor cluster, for carrying out the data processing of batch.

In the present embodiment, the sensing module is specifically as follows one kind and is laid in above system building, for supervising Survey and acquire the service condition for the system resource that specified processor is relied on and operating condition, the task of processor itself The acquisition equipment of the related datas such as disposition.For the concrete form type of above-mentioned sensing module, this specification is not construed as limiting.

In a Sample Scenario, the method for allocating tasks that can be provided using this specification embodiment is processor cluster In each processor distribute data processing task to be processed of next period.

In this Sample Scenario, it is contemplated that each specific processing capacity of processor is to be processed device in processor cluster The task disposition of the service condition and processor self-operating situation of the system resource relied on and a upper period What equal many factors influenced.Therefore, when it is implemented, can monitor and collect by default sensing module in systems The use data of processing section the relied on resource of device, processor in a upper period in a upper period (i.e. first time period) In cluster the operation data of each processor and in a upper period processor task handle data.

Wherein, the relied on resource of above-mentioned processor specifically can be to be made in carrying out data handling procedure for processor , system resource in addition to processor itself.For example, it may be processor system of calling when carrying out data processing CPU, or access system database, the also either I/O interface of system etc. that uses.For above-mentioned processor institute according to Concrete type this specification of resource is relied to be not construed as limiting.The use number of processing section the relied on resource of device in an above-mentioned upper period According to specifically can be understood as characterizing the system resource entirety service condition in a upper period in addition to processor itself Supplemental characteristic.Specifically, the use data of above-mentioned the relied on resource of processor may include: system in a upper period The utilization rate of CPU, the amount of access of system database in a upper period, in a upper period Installed System Memory occupancy Deng.Certainly, it should be noted that the use data of above-mentioned cited the relied on resource of processor are that one kind is schematically said It is bright, when it is implemented, can be selected that other suitable supplemental characteristics is used to be relied on as above-mentioned processor according to concrete scene The use data of resource.In this regard, this specification is not construed as limiting.

The operation data of processor specifically can be understood as characterizing a upper period in an above-mentioned upper period The supplemental characteristic of the operating condition of middle processor.Specifically, the operation data of processor can wrap in an above-mentioned upper period Include: (such as period carries out data processing when institute to the resource data of processor on each processor in a upper period Occupy the process resource amount of processor itself), in a upper period processor operating status (such as normal operation or fortune Row is abnormal) etc..Certainly, it should be noted that the operation data of above-mentioned cited processor is that one kind schematically illustrates, tool When body is implemented, it can be selected to use other operation datas of the suitable supplemental characteristic as above-mentioned processor according to concrete scene. In this regard, this specification is not construed as limiting.

When the task processing data of processor specifically can be understood as characterizing upper one in an above-mentioned upper period Between in section the disposition of processor data processing task supplemental characteristic.Specifically, processor in an above-mentioned upper period Task processing data when may include: the quantity allotted of the data processing task of processor in a upper period, is one upper Between the quantity of the unfinished data processing task of processor in section, in a upper period processor processing failure data processing The quantity of task, handling duration of processor processes data processing task etc. in a upper period.Certainly, it needs to illustrate It is that the task processing data of above-mentioned cited processor are that one kind schematically illustrates, when it is implemented, can be according to specific Scene selects that other suitable supplemental characteristics is used to handle data as the task of above-mentioned processor.In this regard, this specification is not made It limits.

It is laid in systems in advance when it is implemented, sensing module can be, and can be with each processing in system Other resource modules in device and system in addition to above-mentioned processor carry out data interaction.Specifically, above-mentioned sensing module can To obtain other resource modules of each processor and system in addition to processors in processor cluster by data interaction Log recording；And then above-mentioned log recording can be parsed, it obtains handling section in a specific above-mentioned upper period The relied on resource of device used data, the operation data of processor, processor in a upper period in a upper period Task handles data etc..

Sensing module can be sent above-mentioned data by wired or wireless mode after having acquired above-mentioned data To dispatch server.Dispatch server can be analyzed after receiving above-mentioned data by data, determine a period The task disposition of middle processor, and the processing of the task based on a upper period, the operating condition and system of processor In resource service condition in addition to processors；And then can in summary a variety of situation information, taking into account system overall operation While stabilization, determine that each processor handles task amount in the maximum data for handling completion of next period, i.e., The task bearing capacity of processor；Further according to processor distribution and the task bearing capacity that above-mentioned task bearing capacity is in processor cluster Processor to be processed data processing task of the matched data processing task as next period, so that processor Cluster can relatively efficiently utilize system resource, improve the data-handling efficiency of processor cluster entirety.

Specifically, can be as shown in fig.2, dispatch server can be first according to the operation of processor in a upper period Data, the task of processor handles data in a upper period, determines the place for the processing requirement for meeting data processing task Manage device.

Wherein, the processor of the above-mentioned processing requirement for meeting data processing task specifically can be understood as normally transporting Row, and the processor of new data processing task (i.e. the data processing task of subsequent time period) can also be received and processed.

When it is implemented, for example, dispatch server can according to the operation data of processor in a upper period, first from It is normal multiple processors as first processor that multiple processors in processor cluster, which filter out operating status,；Further according to The task of processor handles data in a upper period, the unfinished number of further screening processor from first processor Be less than or equal to the processor of remaining task amount threshold as second processor according to the quantity of processing task, i.e., described in meet data The processor of the processing requirement of processing task.Certainly, it should be noted that be determined for compliance with data processing task cited by above-mentioned The mode of processor of processing requirement be that one kind schematically illustrates, the improper restriction to this specification should not be constituted.

Further, dispatch server can be according to the use number of the relied on resource of processor in the upper period Handled data according to the task of processor in, upper period, determine in next period (i.e. second time period) to point The quantity for the data processing task matched.

Specifically, dispatch server can be according to use data of the relied on resource of processor in a upper period and upper The task of processor handles data in one period, determines the task total amount distributed based on a upper period processor And the data processing task amount actually accomplished, and to complete to remove processor in the data processing task system of a upper period The service condition of the relied on resource of processor in addition；Further according to identified above situation information and next period Specific duration, under the premise of taking into account system run all right, completion can be handled by calculating next period processor The quantity of data processing task, i.e., above-mentioned data processing task quantity to be allocated.

For example, can according in a upper period in processor cluster the data processing task of each processor distribution Quantity calculates the task total amount that a period processor cluster is distributed；Again by comparing a upper period processor The amount of access of the system database in resource relied on, and the database access amount based on determined by system overall stability Threshold value, whether the task total amount for determining that a period processor cluster is distributed suitable, and based on above-mentioned task it is total Whether amount system being capable of stable operation.It may further be according to the amount of access and database of system database in a upper period The comparison result of the threshold value of amount of access, the task total amount distributed upper period processor cluster are adjusted, obtain Task total amount adjusted is as assignable data processing task amount of next period.

For example, if in a upper period threshold value of the amount of access and database amount of access of system database comparison knot Fruit was that the amount of access of system database in a upper period is greater than the threshold value of database access amount, then can be in the upper time On the basis of the task total amount that section processor cluster is distributed, task amount to be allocated is suitably reduced, to guarantee system entirety It is stable；If the comparison result of the threshold value of the amount of access and database amount of access of system database in a upper period It was that the amount of access of system database in a upper period is much smaller than the threshold value of database access amount, then it can be in the upper time On the basis of the task total amount that section processor cluster is distributed, suitably increases task amount to be allocated, be to further increase The utilization rate for resource of uniting, to improve the data-handling efficiency of processor cluster entirety.

Processor and next period of the dispatch server in the processing requirement for determining to meet data processing task In data processing task to be allocated quantity after, may further be according to the processing requirement for meeting data processing task The quantity of processor, data processing task to be allocated in next period, the average task amount of computation processor are made For the task bearing capacity (i.e. the task bearing capacity of target processor) of each processor in the processor cluster.Specifically, It can be by the quantity of data processing task to be allocated in above-mentioned next period divided by the data processing task that meets The quantity of the processor of processing requirement, task bearing capacity of the obtained average value as processor.

Wherein, the task bearing capacity of above-mentioned processor specifically can be understood as integrating in a period processor institute according to The operating condition and task disposition of bad resource entirety service condition and processor itself, that is determined is being handled The maximum data processing completed can be handled under the premise of device cluster integrally keeps stable operation, in processor next period Task amount.

In determining processor cluster after the task bearing capacity of processor, dispatch server can be obtained from storage medium The data processing task for meeting the quantity of data processing task to be allocated in above-mentioned next period is taken, and according to processor Task bearing capacity, it is average using above-mentioned data processing task distribute to each processor in processor cluster as it is next when Between section data processing task to be processed, wherein the data that each processor in above-mentioned processor cluster distributes The task bearing capacity of processing task and processor matches.

Each processor in server cluster is after receiving distributed data processing task, next period meeting Specific data processing is carried out to the data processing task distributed respectively, data processing to be processed is appointed so as to reach Business carries out the effect of batch processing.While processor processes data handles task, processor can also be carried out with sensing module Data interaction is felt so as to which the operation data of processor, task processing data are fed back to sensing module in time in time Know module while can also monitor and the use data of resource that Acquisition Processor is relied on.

In another Sample Scenario, dispatch server according to the operation data of processor in the said one period, on The task processing data of processor are after being determined for compliance with the processor of the processing requirement of data processing task in one period It can be directed to the concrete condition of each processor, more subtly distribute data processing task, it can also be in elder generation's statistical system The quantity of the processor of the processing requirement for meeting data processing task.The processing for meeting data processing task is detected again It is required that the quantity of processor whether be less than preset processor amount threshold；Wherein, above-mentioned processor amount threshold can root It is determined according to the overall performance of system resource.It is less than place in the quantity of the processor of the processing requirement for meeting data processing task In the case where managing device amount threshold, the operation data, task of multiple processors number can be handled from an above-mentioned upper period It using the task processing data for extracting target processor in a upper period in data and is relied on according to relied on resource The use data of resource；Further according to the task processing data and relied on resource of target processor in the upper period Using data, targetedly according to the concrete condition of target processor, the task bearing capacity of the target processor is determined.And It is not using the average task amount of processor in processor cluster as the task bearing capacity of target processor.In this way can more added with Individual instances of the effect ground according to processor, the other systems resource relied on using processor and processor, so that at data The distribution of reason task more rationally, accurately, further increases data-handling efficiency.

Specifically, for example, refering to shown in Fig. 3.The processing that dispatch server meets data processing task described in the determination is wanted In the case that the quantity for the processor asked is less than processor amount threshold, relied on according to processor in the upper period Resource the task processing data of processor and met the processing of data processing task and wants using data, in a upper period The concrete condition for the processor asked determines the next of No. 1 processor for the single concrete condition of each processor respectively The maximum data processing task amount that a period can complete is 5, i.e., the task bearing capacity of No. 1 processor is 5；No. 2 processors The maximum data processing task amount that next period can complete is 3, i.e., the task bearing capacity of No. 2 processors is 3；No. N processing The task bearing capacity of device is 7 etc..And then the task of different processor can be distributed and be treated with a certain discrimination, for example, can be handled to No. 1 Device distributes 5 data processing tasks；3 data processing tasks are distributed to No. 2 processors, are distributed at 7 data to N processor Reason task.

By above-mentioned Sample Scenario as it can be seen that this specification provide method for allocating tasks, due to by obtain and utilize first The relied on resource of processor handles data, fine, conjunction using data, the operation data of processing node and task in period The specific tasks bearing capacity for determining to reason processor, the data of second time period are distributed further according to task bearing capacity to processor Processing task is accurately place so as to the concrete condition for the other systems resource that combination processing device and processor are relied on Manage device distribute task, improve the utilization rate of resource, enable a processor to efficiently, steadily carry out data processing.

As shown in fig.4, this specification embodiment provides a kind of method for allocating tasks, wherein this method specific implementation When, may include the following contents:

S41: the operation data, task processing data and relied on resource of multiple processing nodes in first time period are obtained Use data.

In the present embodiment, above-mentioned first time period specifically can be understood as some historical designated time period.Tool Body, above-mentioned first time period can refer to nearest with the time interval of currently (or next period of task to be allocated) A period, for example, a upper period；It may also mean that as the case may be and require, that selects from history expires One period of sufficient preset requirement, for example, the handled data processing task chosen from history and currently pending A data processing task situation similar period.Certainly, above-mentioned cited first time period is intended merely to preferably say Bright this specification embodiment.The improper restriction to this specification should not be constituted.

In the present embodiment, above system specifically can be understood as including multiple distributed type assemblies systems for handling nodes System.Wherein, distributed deployment has multiple processing nodes in above system.Above-mentioned processing node is mutually indepedent, is each responsible for handling The data processing task distributed.Specifically, system can lead to when system receives a large amount of data processing task to be processed Dispatch server is crossed according to certain sendout, at regular intervals, is gradually distributed to above-mentioned data processing task in system Each processing node.Node is handled after receiving distributed data processing task, respectively distributed data processing is appointed Business carries out specific data processing, so that system can carry out batch processing to data processing task to be processed, to improve number According to treatment effeciency, resource utilization is improved.

In the present embodiment, above-mentioned processing node can specifically refer to have the functions such as certain data operation, storage Electronic equipment, for example, above-mentioned processing node specifically can be processor, server etc..It may also mean that and run on above-mentioned electricity In sub- equipment, the software of support is provided for data processing, storage etc., for example, above-mentioned processing node, which specifically can also be, corresponds to certain The application program etc. of a little data processing tasks.For handling the concrete form of node, this specification is not construed as limiting.

In the present embodiment, the use data that the relied on resource of node is handled in above-mentioned first time period can specifically manage Solution is the supplemental characteristic for characterizing the system resource entirety service condition in first time period in addition to processor itself.Specifically , above-mentioned the relied on resource of processing node may include: the utilization rate of system CPU, first in first time period using data The amount of access of system database, occupancy of Installed System Memory etc. in a upper period in period.Certainly, it needs to illustrate It is that the use data of above-mentioned cited the relied on resource of processing node are that one kind schematically illustrates, when it is implemented, can be with It is selected that other suitable supplemental characteristics is used to use data as above-mentioned the relied on resource of processing node according to concrete scene.It is right This, this specification is not construed as limiting.

In the present embodiment, the operation data that node is handled in above-mentioned first time period specifically can be understood as table Levy the supplemental characteristic that the operating condition of node is handled in first time period.Specifically, handling node in above-mentioned first time period Operation data may include: in first time period handle node resource data (such as each processing node first time period into The stock number of occupied processing node itself when row data processing), the operating status that handles node in first time period (such as transports Row is normal or is operating abnormally) etc..Certainly, it should be noted that the operation data of above-mentioned cited processing node is one Kind schematically illustrates, when it is implemented, can be selected to use other suitable supplemental characteristics as above-mentioned place according to concrete scene Manage the operation data of node.In this regard, this specification is not construed as limiting.

In the present embodiment, the task processing data that node is handled in above-mentioned first time period specifically can be understood as using The supplemental characteristic of the disposition of processing node data processing task in characterization first time period.Specifically, when above-mentioned first Between the task processing data of node are handled in section may include: that the distribution of the data processing task of node is handled in first time period Node processing is handled in the quantity, first time period of the unfinished data processing task of processing node in quantity, first time period Handling duration of node processing data processing task etc. is handled in the quantity of data processing task of failure, first time period.When So, it should be noted that the task processing data of above-mentioned cited processing node are that one kind schematically illustrates, specific implementation When, it can be selected that other suitable supplemental characteristics is used to handle data as the task of above-mentioned processing node according to concrete scene. In this regard, this specification is not construed as limiting.

In the present embodiment, the use data of processing the relied on resource of node, first in above-mentioned acquisition first time period The operation data of node is handled in period, the handling node in first time period of the task handles data, when it is implemented, can be with It include: that processing the relied on resource of node in above-mentioned first time period is acquired and obtained by default sensing module in systems Task processing data etc. of node are handled in operation data, first time period using processing node in data, first time period Data.

In the present embodiment, it should be noted that by processing the relied on resource of node in above-mentioned first time period Handle data using handling the operation data of node in data, first time period, handling node in first time period for task It is analyzed and processed, can determine the disposition of the data processing task that system is distributed in first time period and right Answer the disposition of above-mentioned data processing task, processing node individual, to handle the specific feelings for the system resource that node is relied on Condition.And then subsequent the case where capable of can referring to above-mentioned identified first time period information, taking into account operation stability and money Under the premise of source utilization rate, the task allocation plan of each processing node can be subtly determined.Certainly, it needs to illustrate Being that above scheme only lists obtains the use data of processing the relied on resource of node in first time period, in first time period It handles the operation data of node, handle the data such as the task processing data of node in first time period, when it is implemented, according to tool The application scenarios of body and requirement can also obtain other historical datas in addition to above-mentioned cited data.In this regard, this explanation Book is not construed as limiting.

S43: according to the operation data of processing nodes multiple in the first time period, task processing data and money is relied on The use data in source determine the task bearing capacity of target processing node.

In the present embodiment, above-mentioned target processing node specifically can be understood as any one in multiple processing nodes Handle node.Above-mentioned task bearing capacity specifically can be understood as comprehensive (relied on based on node is handled in the first time period Resource using data, the operation data that handles node, handle determined by the task processing data of node) in first time period Handle node individual instances (operating condition and task disposition including handling node), and the money that processing node is relied on The overall condition in source, determined under the premise of keeping stable operation, processing node is (i.e. current next in second time period A period) in can handle completion maximum data handle task amount.

In the present embodiment, it is above-mentioned according to the processing operation datas of nodes multiple in the first time period, task at The use data for managing data and relied on resource determine the task bearing capacity of target processing node, when it is implemented, dispatch service Device can be according to the operation using processing node in data, first time period for handling the relied on resource of node in first time period Task of node is handled in data, first time period handles data, determines in first time period, at the data distributed The disposition of reason task and the data processing task completed, processing node individual, and the system that processing node is relied on The situation data of resource entirety；Further according to above situation data, in system of taking into account, the operation stability and resource of processing node Under the premise of utilization rate, the task amount that can be done that node is handled in second time period is determined according to certain strategy, i.e., The task bearing capacity of target processing node.

S45: Xiang Suoshu target handle node distribution second time period with the matched data processing of task bearing capacity Task.

In the present embodiment, above-mentioned second time specifically can be understood as a period of difference first time period. For example, it may be next period, or some following designated time period etc..

In the present embodiment, it is above-mentioned to the target processing node distribution second time period with the task bearing capacity Matched data processing task, when it is implemented, dispatch server is obtaining second time period data processing to be processed times After business, above-mentioned second time period data processing task to be processed can be distributed according to each processing node tasks bearing capacity To processing node each in system.Wherein, the data processing task and handle node that each processing node distributes Task carrying is flux matched.For example, the task bearing capacity of No. 1 processing node is 3, then the second time of No. 1 processing node is distributed to The data processing task amount of section can be 3, or the task amount less than 3.Pass through each processing node of the above-mentioned task method of salary distribution The data processing task for the second time period distributed be meet processing node individual, and processing node relied on be Resource overall condition of uniting, therefore, based on the data processing task of above-mentioned distribution, system resource and system can be effectively utilized In processing node improve data-handling efficiency so as to stablize and efficiently carry out data processing.

Therefore the method for allocating tasks that this specification provides, due to by obtaining and using in first time period That manages the relied on resource of node handles data using data, the operation data of processing node and task, subtly determines source The specific tasks bearing capacity for managing node is appointed further according to task bearing capacity to the data processing of processing node distribution second time period Business, so as to according to processing node individual, and the concrete condition of system resource entirety that processing node is relied on, rationally, standard Really for processing node distribute task, improve the utilization rate of resource, enable multiple processing nodes efficiently, steadily carry out data Processing.

In one embodiment, the use data that the relied on resource of node is handled in the first time period specifically can be with Including at least one of: the amount of access of system database in the utilization rate, first time period of system CPU in first time period, The occupancy etc. of Installed System Memory in first time period.Certainly, it should be noted that above-mentioned cited processing node relies on money The use data in source are that one kind schematically illustrates, when it is implemented, can be selected using other suitably according to concrete scene Use data of the supplemental characteristic as above-mentioned the relied on resource of processing node.In this regard, this specification is not construed as limiting.

In one embodiment, the operation data for node being handled in the first time period includes at least one of: The resource data of node is handled in first time period, handles operating status of node etc. in first time period.Certainly, it needs to illustrate , the operation data of above-mentioned cited processing node is that one kind schematically illustrates, when it is implemented, can be according to specific Scene selects the operation data for using other suitable supplemental characteristics as above-mentioned processing node.In this regard, this specification does not limit It is fixed.

In one embodiment, in the first time period handle node task processing data include it is following at least it One: handling what node did not completed in the quantity allotted, first time period of the data processing task of processing node in first time period The quantity of the data processing task of processing node processing failure, first time in the quantity of data processing task, first time period The handling duration etc. of node processing data processing task is handled in section.Certainly, it should be noted that above-mentioned cited processing section The task processing data of point are that one kind schematically illustrates, when it is implemented, can be selected according to concrete scene using other conjunctions Suitable supplemental characteristic handles data as the task of above-mentioned processing node.In this regard, this specification is not construed as limiting.

In one embodiment, handled in above-mentioned acquisition first time period the relied on resource of node use data, the The operation data of node is handled in one period, the handling node in first time period of the task handles data, when specific implementation, can To include the following contents: acquiring the multiple processing node in first time period by default sensing module in systems Operation data, task processing data and the institute of processing node in first time period when managing distributed data processing task according to Rely the use data of resource.

In the present embodiment, above-mentioned sensing module specifically can be understood as it is a kind of be laid in system in advance, Neng Gouyu Other resource modules in each processor and system in system in addition to above-mentioned processing node carry out data interaction, adopt Collect the device or equipment of related situation data.Specifically, above-mentioned sensing module can be obtained each in cluster by data interaction The log recording of other resource modules of a processing node and system in addition to handling node；And then it can be to above-mentioned log Record is parsed, and is obtained the use data of processing the relied on resource of node in first time period, is handled section in first time period Situations such as operation data of point, task of node is handled in first time period handle data data.

In one embodiment, above-mentioned operation data, task according to processing nodes multiple in the first time period The use data for handling data and relied on resource determine the task bearing capacity of target processing node, when it is implemented, can wrap Include the following contents:

S1: data are handled according to operation data, the task of processing nodes multiple in the first time period, determine multiple symbols Close the processing node of the processing requirement of data processing task；

S2: handling data according to use data, the task of processing the relied on resources of node multiple in the first time period, Determine the quantity of data processing task to be allocated in second time period；

S3: according in the processing node of the multiple processing requirement for meeting data processing task, the second time period The quantity of data processing task to be allocated, the average task amount of calculation processing node, as appointing for target processing node Business bearing capacity.

In the present embodiment, the processing node of the above-mentioned processing requirement for meeting data processing task, it is specific to be understood that For that can operate normally, and the disposition of the data processing task based on history, and processing node individual operating condition and The case where handling node relied on resource, can also also receive the processing of the new data processing task of processing in second time period Node.

In the present embodiment, it is above-mentioned according to the processing operation datas of nodes multiple in the first time period, task at Data are managed, the processing node of multiple processing requirements for meeting data processing task are determined, specifically, can be dispatch server root Data are handled according to handling the operation data of node in first time period, handling node in first time period for task, determine the The operating status (such as normal operation or operation exception etc.) of node and the data of each processing node are handled in one period Processing task disposition (such as unfinished data processing task quantity or data processing task processing failure amount Deng)；And then can accordingly data, first filter out operating status from processing node and normally handle node；Again from Above-mentioned operating status normally handles the quantity that unfinished data processing task is filtered out in node less than or equal to remaining number of tasks The processing node of threshold value (or the processing failure amount of data processing task is less than or equal to failure amount threshold) is measured as above-mentioned symbol Close the processing node of the processing requirement of data processing task.

In the present embodiment, the above-mentioned use number according to processing the relied on resources of node multiple in the first time period Data are handled according to, task, determine the quantity of data processing task to be allocated in second time period, when it is implemented, scheduling clothes Being engaged in device can be according to times using processing node in data, first time period for handling the relied on resource of node in first time period Business processing data are determined for the data processing task total amount and reality that system processing node is distributed in first time period The data processing task amount of completion, (such as system spare makes the particular condition in use of the processing node system resource that is relied on With resource etc.), and can determine that out based on systematic entirety the use threshold value of system resource；Further according to being in above-mentioned first time period The system processing node data processing task total amount distributed and the data processing task amount that actually accomplishes, system resource it is specific Service condition, determining in conjunction with the use threshold value of system resource and the specific duration of second time period can be in guarantee system The maximum value for the data processing task amount that the premise of energy stable operation can be handled is as data to be allocated in second time period The quantity of processing task.

In the present embodiment, the above-mentioned processing node according to the multiple processing requirement for meeting data processing task, The quantity of data processing task to be allocated in the second time period, the average task amount of calculation processing node, as described The task bearing capacity of target processing node.When it is implemented, processing number of nodes is larger in the system of considering, different disposal node The operating condition and task disposition of individual may have differences, if for the operation feelings of each processing node individual If condition and task disposition calculate separately the task bearing capacity of each processing node, it certainly will need to occupy a large amount of operation Resource influences to determine efficiency.Therefore, dispatch server can be by data processing task to be allocated in above-mentioned second time period Quantity obtains quotient divided by the quantity of the processing node for the processing requirement for meeting data processing task, as the flat of processing node Equal task amount；And by above-mentioned average task amount collectively as the task bearing capacity of target processing node, i.e., each processing node the The maximum task amount of completion is capable of handling in two periods.The task carrying of the processing node of target determined by can guaranteeing in this way It is reasonable, accurate for measuring to most of processing node in cluster, while can be effectively reduced determining target processing node The resource of task bearing capacity to be expended and time improve treatment effeciency.

In one embodiment, processing node in systems quantity is relatively fewer or the calculating of dispatch server The stronger situation of ability can be with for accuracy, the reliability of the task bearing capacity of the processing node of target determined by improving According to the concrete condition of each processing node, corresponding task is calculated separately for each concrete condition for handling node and is held Carrying capacity.

In the present embodiment, it is above-mentioned according to the processing operation datas of nodes multiple in the first time period, task at The use data for managing data and relied on resource determine the task bearing capacity of target processing node, when it is implemented, can also wrap Include the following contents:

S2: the quantity of the processing node of the multiple processing requirement for meeting data processing task of statistical system；

S3: it is default whether the quantity of the processing node of the multiple processing requirement for meeting data processing task of detection is less than Number of nodes threshold value；

S4: it is less than preset number of nodes in the quantity of the processing node of the processing requirement for meeting data processing task In the case where threshold value, operation data, the task of multiple processing nodes handle data and rely on money from the first time period The use number using task the processing data and relied on resource for obtaining target processing node in first time period in data in source According to；

S5: according to the use number of task the processing data and relied on resource of target processing node in the first time period According to determining the task bearing capacity of target processing node.

In the present embodiment, the specific value of above-mentioned preset number of nodes threshold value can be according to dispatch service in system The quantity of node is handled in the data-handling capacity of device, system and precise requirements flexibly determine.

In the present embodiment, it is less than in the quantity of the processing node of the processing requirement for meeting data processing task pre- If number of nodes threshold value in the case where, can from the acquired first time period it is multiple processing nodes operation numbers According to, task processing data and relied on resource using successively extraction meets the place of the processing requirement of data processing task in data Manage the operation data of node (i.e. target processing node), the use data of task processing data and relied on resource；And according to mesh Operation data, the use data of task processing data and relied on resource of mark processing node, determine corresponding mesh one by one The task bearing capacity of mark processing node.The task bearing capacity for the different disposal node determined in this way is according to the difference for handling node Situation, can be different, be not necessarily unified numerical value.And then subsequent dispatch server can be respectively according to each processing The task bearing capacity of node, for different processing nodes distribute the data processing task of different number as second time period to The data processing task of processing.The processing capacity that each processing node can preferably be played in this way, further improves The accurate and reasonability of task distribution, so that more efficient based on above-mentioned distribution data processing, stable.

In one embodiment, above-mentioned according to the first time in order to enable determining task bearing capacity efficiently, accurately The operation data of multiple processing nodes, the use data of task processing data and relied on resource, determine target processing section in section The task bearing capacity of point, when it is implemented, can also include the following contents: by the prediction model of preset task bearing capacity, According to the use number of the operation data of processing nodes multiple in the first time period, task processing data and relied on resource According to, determine target processing node task bearing capacity.

In the present embodiment, the prediction model of above-mentioned preset task bearing capacity specifically can be one kind by history Data (including handle in history the relied on resource of node using data, handle the operation data of node, and task in history Handle data) learning training is carried out, that is established can handle completion to processing node in some following period The model that data processing task amount is predicted.

In at this implementation place, when it is implemented, the relied on resource of node can will be handled in above-mentioned first time period Handle data using handling the operation data of node in data, first time period, handling node in first time period for task As mode input data, it is input to the prediction model of trained preset task bearing capacity, obtains model output Result data, i.e., the task bearing capacity of each processing node in second time period.

In one embodiment, the prediction model of the preset task bearing capacity can specifically be built in the following way It is vertical: the operation data of multiple processing nodes in preset time period, the use data of task processing data and relied on resource are obtained, As sample data；It is trained using the sample data, establishes the prediction model of the preset task bearing capacity.

In the present embodiment, above-mentioned preset time period can specifically refer to historical some or multiple periods. For example, it may be data processing task handled by system and second time period are by data processing task phase to be processed in history Close multiple periods.

In the present embodiment, when it is implemented, can use above-mentioned sample data neural network model is trained, Study obtains the prediction model of above-mentioned preset task bearing capacity to establish.

In the present embodiment, you need to add is that, above-mentioned cited a variety of realizations are according in the first time period Operation data, the use data of task processing data and relied on resource of multiple processing nodes, determine target processing node The embodiment of task bearing capacity all schematically illustrates.When it is implemented, can also be according to specific application scenarios and implementation It is required that selecting other suitable embodiments to determine the task bearing capacity of target processing node.In this regard, this specification does not limit It is fixed.

In one embodiment, when it is implemented, dispatch server was distributing for the second time for each processing node It, can also be by handling number in processing node second time period each in sensing module acquisition system after the data processing task of section Task processing data, the operation data of processing node when according to processing task, and the use number of processing the relied on resource of node According to；Operation and task disposition that node is handled in second time period, and processing node institute are determined according to above-mentioned data The service condition of the system resource of dependence；And accordingly data, to the number for the second time period that processing node is distributed It is adjusted correspondingly according to processing task.For example, finding the operating status for handling node Y in second time period according to above-mentioned data Poor, the still untreated data processing task of second time period is more, can will handle the second time distributed before node Y A part in the data processing task of section distributes to that operating status is preferable, and the data processing task of second time period has been handled It finishes, the processing nodes X being in idle condition is handled.It can be effectively reduced the processing pressure of processing node Y in this way Power further improves resource utilization simultaneously effective using the processing nodes X being in idle condition, and avoids processing section Point Y leads to processing failure occur since processing pressure is excessively high, so that data processing is more efficient, stablizes.

Therefore the method for allocating tasks that this specification provides, due to by obtaining and using in first time period That manages the relied on resource of node handles data using data, the operation data of processing node and task, subtly determines source The specific tasks bearing capacity for managing node is appointed further according to task bearing capacity to the data processing of processing node distribution second time period Business, so as to according to processing node, and the concrete condition of resource that processing node is relied on, rationally, be accurately each place It manages node and distributes task, improve the utilization rate of resource, so that collection group energy is efficient, steadily carries out data processing；Also pass through utilization The prediction model of preparatory trained preset task bearing capacity is according to handling making for the relied on resource of node in first time period With data, handle the operation data of node and the task carrying of the determining processing node for task distribution of task processing data Amount improves the accuracy and reasonability of identified task bearing capacity, so that task distribution is more accurate, reasonable.

This specification embodiment additionally provides a kind of server, including processor and refers to for storage processor to be executable The memory of order, the processor can be according to instruction execution following steps when being embodied: obtaining multiple in first time period Handle the operation data of node, the use data of task processing data and relied on resource；According to more in the first time period Operation data, the use data of task processing data and relied on resource of a processing node, determine appointing for target processing node Business bearing capacity；To appointing with the matched data processing of task bearing capacity for target processing node distribution second time period Business.

In order to more accurately complete above-metioned instruction, as shown in fig.5, this specification additionally provides another kind specifically Server, wherein the server includes network communications port 501, processor 502 and memory 503, and above structure is logical It crosses Internal cable to be connected, so that each structure can carry out specific data interaction.

Wherein, the network communications port 501 specifically can be used for obtaining the fortune of multiple processing nodes in first time period The use data of row data, task processing data and relied on resource.

The processor 502, specifically can be used for according in the first time period it is multiple processing nodes operation datas, Task handles the use data of data and relied on resource, determines the task bearing capacity of target processing node；To at the target Manage node distribution second time period with the matched data processing task of task bearing capacity.

The memory 503 specifically can be used for storing a variety of data obtained through network communications port 501, Yi Jixiang The instruction repertorie answered.

In the present embodiment, the network communications port 501 can be is bound from different communication protocol, thus The virtual port of different data can be sent or received.Lead to for example, the network communications port can be responsible for progress web data No. 80 ports of letter are also possible to No. 21 ports for being responsible for carrying out FTP data communication, can also be that responsible progress mail data is logical No. 25 ports of letter.In addition, the network communications port can also be the communication interface or communication chip of entity.For example, its It can be mobile radio network communication chip, such as GSM, CDMA；It can also be Wifi chip；It can also be bluetooth core Piece.

In the present embodiment, the processor 502 can be implemented in any suitable manner.For example, processor can be with Take such as microprocessor or processor and storage can by (micro-) processor execute computer readable program code (such as Software or firmware) computer-readable medium, logic gate, switch, specific integrated circuit (Application Specific IntegratedCircuit, ASIC), programmable logic controller (PLC) and the form etc. for being embedded in microcontroller.This specification is not It limits.

In the present embodiment, the memory 503 may include many levels, in digital display circuit, as long as can save Binary data can be memory；In integrated circuits, the circuit with store function of a not no physical form Also memory, such as RAM, FIFO are；In systems, the storage equipment with physical form is also memory, such as memory bar, TF Card etc..

This specification embodiment additionally provides a kind of computer storage medium based on above-mentioned method for allocating tasks, the meter Calculation machine storage medium is stored with computer program instructions, is performed realization in the computer program instructions: when obtaining first Between the operation datas of multiple processing nodes, the use data of task processing data and relied on resource in section；According to described first The operation data of multiple processing nodes, the use data of task processing data and relied on resource, determine at target in period Manage the task bearing capacity of node；To the target processing node distribution second time period with the matched number of task bearing capacity According to processing task.

In the present embodiment, above-mentioned storage medium includes but is not limited to random access memory (Random Access Memory, RAM), read-only memory (Read-Only Memory, ROM), caching (Cache), hard disk (Hard Disk Drive, HDD) or storage card (Memory Card).The memory can be used for storing computer program instructions.Network is logical Letter unit can be according to standard setting as defined in communication protocol, for carrying out the interface of network connection communication.

In the present embodiment, the function and effect of the program instruction specific implementation of computer storage medium storage, can To compare explanation with other embodiment, details are not described herein.

As shown in fig.6, this specification embodiment additionally provides a kind of task allocation apparatus on software view, the dress It sets and can specifically include construction module below:

Module 601 is obtained, specifically can be used for obtaining in first time period at the operation data, task of multiple processing nodes Manage the use data of data and relied on resource；

Determining module 602 specifically can be used for the operation data according to processing nodes multiple in the first time period, appoint The use data of business processing data and relied on resource, determine the task bearing capacity of target processing node；

Distribution module 603, specifically can be used for the target handle node distribution second time period with the task The matched data processing task of bearing capacity.

In one embodiment, the use data that the relied on resource of node is handled in the first time period specifically can be with Including at least one of: the amount of access of system database in the utilization rate, first time period of system CPU in first time period, The occupancy etc. of Installed System Memory in first time period.

In one embodiment, in the first time period handle node operation data can specifically include with down toward It is one of few: to handle the resource data of node in first time period, handle operating status of node etc. in first time period.

In one embodiment, in the first time period handle node task processing data can specifically include with It is at least one lower: to handle node in the quantity allotted, first time period of the data processing task of processing node in first time period The quantity of the data processing task of processing node processing failure in the quantity of unfinished data processing task, first time period, The handling duration etc. of node processing data processing task is handled in first time period.

In one embodiment, the module 601 that obtains specifically can be especially by default perception mould in systems Block acquires in first time period when the multiple processing node handles distributed data processing task in first time period Processing node operation data, task processing data and relied on resource use data.

In one embodiment, the determining module 602, can specifically include following structural unit:

First determination unit, specifically can be used for according in the first time period it is multiple processing nodes operation datas, Task handles data, determines the processing node of multiple processing requirements for meeting data processing task；

Second determination unit specifically can be used for according to processing the relied on resources of node multiple in the first time period Data are handled using data, task, determine the quantity of data processing task to be allocated in second time period；

Computing unit specifically can be used for the processing section according to the multiple processing requirement for meeting data processing task The quantity of data processing task to be allocated in point, the second time period, the average task amount of calculation processing node, as institute State the task bearing capacity of target processing node.

In one embodiment, the determining module 602 can specifically include following structural unit:

Statistic unit specifically can be used for the processing of the multiple processing requirement for meeting data processing task of statistical system The quantity of node；

Detection unit specifically can be used for detecting the processing node of the multiple processing requirement for meeting data processing task Quantity whether be less than preset number of nodes threshold value；

Third determination unit specifically can be used for the processing node in the processing requirement for meeting data processing task Quantity be less than preset number of nodes threshold value in the case where, from the first time period it is multiple processing nodes operation datas, Task handles data and relied on resource using the task processing number for obtaining target processing node in first time period in data According to the use data with relied on resource；

4th determination unit specifically can be used for handling number according to the task of target processing node in the first time period According to the use data with relied on resource, the task bearing capacity of the target processing node is determined.

It should be noted that unit, device or module etc. that above-described embodiment illustrates, specifically can by computer chip or Entity is realized, or is realized by the product with certain function.For convenience of description, it describes to divide when apparatus above with function It is described respectively for various modules.It certainly, can be the function of each module in same or multiple softwares when implementing this specification And/or realized in hardware, the module for realizing same function can also be realized by the combination of multiple submodule or subelement etc..With Upper described Installation practice is only schematical, for example, the division of the unit, only a kind of logic function is drawn Point, there may be another division manner in actual implementation, such as multiple units or components may be combined or can be integrated into separately One system, or some features can be ignored or not executed.Another point, shown or discussed mutual coupling or straight Connecing coupling or communication connection can be through some interfaces, and the indirect coupling or communication connection of device or unit can be electrical property, Mechanical or other forms.

Therefore the task allocation apparatus that this specification embodiment provides, first time period is obtained by obtaining module Middle processing the relied on resource of node handles data using data, the operation data of processing node and task, passes through and determines mould Root tuber subtly determines the specific tasks bearing capacity of processing node according to above-mentioned data, then is carried by distribution module according to task The data processing task to processing node distribution second time period is measured, so as to according to the specific feelings of system and processing node Condition rationally, accurately distributes task for processing node, improves the utilization rate of resource, efficiently, steadily carry out data processing.

Although being based on routine or nothing present description provides the method operating procedure as described in embodiment or flow chart Creative means may include more or less operating procedure.The step of enumerating in embodiment sequence is only numerous steps One of rapid execution sequence mode does not represent and unique executes sequence.When device or client production in practice executes, Can be executed according to embodiment or the execution of method shown in the drawings sequence or parallel (such as parallel processor or multithreading The environment of processing, even distributed data processing environment).The terms "include", "comprise" or its any other variant are intended to Cover non-exclusive inclusion, so that the process, method, product or the equipment that include a series of elements not only include those Element, but also including other elements that are not explicitly listed, or further include for this process, method, product or setting Standby intrinsic element.In the absence of more restrictions, being not precluded is including process, method, the product of the element Or there is also other identical or equivalent elements in equipment.The first, the second equal words are used to indicate names, and are not offered as appointing What specific sequence.

It is also known in the art that other than realizing controller in a manner of pure computer readable program code, it is complete Entirely can by by method and step carry out programming in logic come so that controller with logic gate, switch, specific integrated circuit, programmable Logic controller realizes identical function with the form for being embedded in microcontroller etc..Therefore this controller is considered one kind Hardware component, and the structure that the device for realizing various functions that its inside includes can also be considered as in hardware component.Or Person even, can will be considered as realizing the device of various functions either the software module of implementation method can be hardware again Structure in component.

This specification can describe in the general context of computer-executable instructions executed by a computer, such as journey Sequence module.Generally, program module include routines performing specific tasks or implementing specific abstract data types, programs, objects, Component, data structure, class etc..This specification can also be practiced in a distributed computing environment, in these distributed computing rings In border, by executing task by the connected remote processing devices of communication network.In a distributed computing environment, program mould Block can be located in the local and remote computer storage media including storage equipment.

As seen through the above description of the embodiments, those skilled in the art can be understood that this specification It can realize by means of software and necessary general hardware platform.Based on this understanding, the technical solution of this specification Substantially the part that contributes to existing technology can be embodied in the form of software products in other words, the computer software Product can store in storage medium, such as ROM/RAM, magnetic disk, CD, including some instructions are used so that a computer Equipment (can be personal computer, mobile terminal, server or the network equipment etc.) execute each embodiment of this specification or Method described in certain parts of person's embodiment.

Each embodiment in this specification is described in a progressive manner, the same or similar portion between each embodiment Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.This specification can be used for In numerous general or special purpose computing system environments or configuration.Such as: personal computer, server computer, handheld device Or portable device, laptop device, multicomputer system, microprocessor-based system, set top box, programmable electronics set Standby, network PC, minicomputer, mainframe computer, distributed computing environment including any of the above system or equipment etc..

Although depicting this specification by embodiment, it will be appreciated by the skilled addressee that there are many become for this specification Shape and the spirit changed without departing from this specification, it is desirable to which the attached claims include these deformations and change without departing from this The spirit of specification.

Claims

1. a kind of method for allocating tasks, comprising:

Obtain the use number of the operation data of multiple processing nodes, task processing data and relied on resource in first time period According to；

According to the use of the operation data of processing nodes multiple in the first time period, task processing data and relied on resource Data determine the task bearing capacity of target processing node；

To the target processing node distribution second time period with the matched data processing task of task bearing capacity.

2. according to the method described in claim 1, handling the use data packet of the relied on resource of node in the first time period Include at least one of: the amount of access of system database, in the utilization rate, first time period of system CPU in first time period The occupancy of Installed System Memory in one period.

3. according to the method described in claim 1, in the first time period handle node operation data include it is following at least One of: the operating status for handling the resource data of node in first time period, handling node in first time period.

4. according to the method described in claim 1, the task processing data for handling node in the first time period include following At least one: node is handled not in the quantity allotted, first time period of the data processing task of processing node in first time period The quantity of the data processing task of processing node processing failure in the quantity of the data processing task of completion, first time period, the The handling duration of node processing data processing task is handled in one period.

5. according to the method described in claim 1, obtaining the operation data of multiple processing nodes, task processing in first time period The use data of data and relied on resource, comprising:

The multiple processing node is acquired by default sensing module in systems handle in first time period distributed The operation data of processing node in first time period when data processing task, task processing data and relied on resource make Use data.

6. according to the method described in claim 1, according to operation data, the task of processing nodes multiple in the first time period The use data for handling data and relied on resource determine the task bearing capacity of target processing node, comprising:

Data are handled according to the processing operation datas of nodes multiple in the first time period, task, determines and multiple meets data The processing node of the processing requirement of processing task；

According to the use data of processing the relied on resources of node multiple in the first time period, task processing data, the is determined The quantity of data processing task to be allocated in two periods；

According to be allocated in the processing node of the multiple processing requirement for meeting data processing task, the second time period The quantity of data processing task, the average task amount of calculation processing node, the task bearing capacity as target processing node.

7. according to the method described in claim 1, according to operation data, the task of processing nodes multiple in the first time period The use data for handling data and relied on resource determine the task bearing capacity of target processing node, comprising:

Count the quantity of the processing node of the multiple processing requirement for meeting data processing task；

Whether the quantity for detecting the processing node of the multiple processing requirement for meeting data processing task is less than preset node Amount threshold；

It is less than preset number of nodes threshold value in the quantity of the processing node of the processing requirement for meeting data processing task In the case of, the operation data, task processing data of multiple processing nodes and making for relied on resource from the first time period With the use data for task the processing data and relied on resource for obtaining target processing node in first time period in data；

According to the use data of task the processing data and relied on resource of target processing node in the first time period, determine The task bearing capacity of the target processing node.

8. according to the method described in claim 1, according to operation data, the task of processing nodes multiple in the first time period The use data for handling data and relied on resource determine the task bearing capacity of target processing node, comprising:

By the prediction model of preset task bearing capacity, according to the operation number of processing nodes multiple in the first time period According to the use data of, task processing data and relied on resource, determine that target handles the task bearing capacity of node.

9. according to the method described in claim 8, the prediction model of the preset task bearing capacity is established in the following way:

Obtain the use number of the operation data of multiple processing nodes, task processing data and relied on resource in preset time period According to as sample data；

It is trained using the sample data, establishes the prediction model of the preset task bearing capacity.

10. a kind of task allocation apparatus, comprising:

Module is obtained, for obtaining the operation data, task processing data of multiple processing nodes in first time period and being relied on The use data of resource；

Determining module, for the operation data, task processing data and institute according to processing nodes multiple in the first time period The use data for relying on resource determine the task bearing capacity of target processing node；

Distribution module, for the target processing node distribution second time period with the matched data of task bearing capacity Processing task.

11. device according to claim 10, the use data of the relied on resource of node are handled in the first time period Including at least one of: the amount of access of system database in the utilization rate, first time period of system CPU in first time period, The occupancy of Installed System Memory in first time period.

12. device according to claim 10, handled in the first time period node operation data include with down toward It is one of few: the operating status for handling the resource data of node in first time period, handling node in first time period.

13. device according to claim 10, handled in the first time period node task processing data include with It is at least one lower: to handle node in the quantity allotted, first time period of the data processing task of processing node in first time period The quantity of the data processing task of processing node processing failure in the quantity of unfinished data processing task, first time period, The handling duration of node processing data processing task is handled in first time period.

14. device according to claim 10, the module that obtains is adopted especially by default sensing module in systems Collect the place in first time period when the multiple processing node handles distributed data processing task in first time period Manage the operation data of node, the use data of task processing data and relied on resource.

15. device according to claim 10, the determining module, comprising:

First determination unit, for handling data according to operation data, the task of processing nodes multiple in the first time period, Determine the processing node of multiple processing requirements for meeting data processing task；

Second determination unit, for according in the first time period it is multiple processing the relied on resources of node use data, times Business processing data, determine the quantity of data processing task to be allocated in second time period；

Computing unit, for according to the processing node of the multiple processing requirement for meeting data processing task, it is described second when Between data processing task to be allocated in section quantity, the average task amount of calculation processing node saves as target processing The task bearing capacity of point.

16. device according to claim 10, the determining module include:

Statistic unit, the quantity of the processing node for counting the multiple processing requirement for meeting data processing task；

Whether detection unit is small for detecting the quantity of processing node of the multiple processing requirement for meeting data processing task In preset number of nodes threshold value；

Third determination unit, the quantity for the processing node in the processing requirement for meeting data processing task are less than default Number of nodes threshold value in the case where, multiple processing operation datas of nodes, task handle data from the first time period With relied on resource using obtaining the task processing data of target processing node in first time period in data and rely on money The use data in source；

4th determination unit, for handling data according to the task of target processing node in the first time period and relying on money The use data in source determine the task bearing capacity of the target processing node.

17. a kind of server, including processor and for the memory of storage processor executable instruction, the processor is held The step of any one of claims 1 to 9 the method is realized when row described instruction.

18. a kind of computer readable storage medium is stored thereon with computer instruction, described instruction, which is performed, realizes that right is wanted The step of seeking any one of 1 to 9 the method.