CN113010273B - Human resource data distributed task processing method and system - Google Patents

Human resource data distributed task processing method and system Download PDF

Info

Publication number
CN113010273B
CN113010273B CN202110308427.8A CN202110308427A CN113010273B CN 113010273 B CN113010273 B CN 113010273B CN 202110308427 A CN202110308427 A CN 202110308427A CN 113010273 B CN113010273 B CN 113010273B
Authority
CN
China
Prior art keywords
task
processor
time slice
power
computing
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.)
Active
Application number
CN202110308427.8A
Other languages
Chinese (zh)
Other versions
CN113010273A (en
Inventor
吴方同
吴晓军
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hebei Jilian Human Resources Service Group Co ltd
Original Assignee
Hebei Jilian Human Resources Service Group Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hebei Jilian Human Resources Service Group Co ltd filed Critical Hebei Jilian Human Resources Service Group Co ltd
Priority to CN202110308427.8A priority Critical patent/CN113010273B/en
Publication of CN113010273A publication Critical patent/CN113010273A/en
Application granted granted Critical
Publication of CN113010273B publication Critical patent/CN113010273B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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
    • G06F9/465—Distributed object oriented systems
    • 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
    • G06F9/5011—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
    • G06F9/5016—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals the resource being the memory
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/10—Office automation; Time management
    • G06Q10/105—Human resources

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Strategic Management (AREA)
  • Operations Research (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Quality & Reliability (AREA)
  • Marketing (AREA)
  • Economics (AREA)
  • Data Mining & Analysis (AREA)
  • Multi Processors (AREA)

Abstract

The invention provides a method and a system for processing distributed tasks of human resource data, which divide human resource events into time slices, calculate and sequence the priorities of the time slices, obtain and adjust the power of a task processor according to the calculation result of the priorities, solve the problem of power adjustment of the task processor, improve the utilization rate of hardware resources, realize the optimal configuration of distributed calculation on different tasks, solve the problem of insufficient data calculation capacity of human resource enterprises, and improve the speed of big data calculation and the efficiency of human resource configuration.

Description

Human resource data distributed task processing method and system
Technical Field
The invention relates to the technical field of big data, in particular to a human resource data distributed task processing method and system.
Background
With the rapid development of communication technology and computer technology, emerging services such as cloud computing, internet of things and social networks promote the unprecedented speed increase of data types and scales of human society, the high integration of the human, machine and object ternary world causes the explosive increase of data scales and the high complexity of data modes, and the world enters the networked big data era. For example, intelligent analysis of large data within the human resources industry is increasingly exposed to the problem of limited computing power in the face of increasing data size. At present, the traditional method for solving the calculation task is realized by mainly executing big data calculation and increasing hardware resource capacity, but the method usually brings new problems, such as high cost and high maintenance difficulty. Therefore, a data calculation mode is to be provided, so that the speed and the efficiency of big data calculation are improved, and the problem of insufficient calculation capacity of human resource data is solved.
Disclosure of Invention
Based on the problems, the invention solves the problem of power adjustment of the task processor through the ADMM algorithm, realizes the optimal configuration of distributed calculation on different tasks, thereby solving the problem of insufficient data calculation capability of human resource enterprises and improving the speed of big data calculation and the efficiency of human resource configuration.
In order to achieve the above object, the present invention provides a human resource data distributed task processing method, including:
step 101, acquiring a human resource event to be calculated;
step 102, dividing an event into at least one time slice;
103, dynamically calculating task priorities in different time slice processes;
step 104, adjusting the power of each task handler.
Further, the event can be resume push, resume filtering, and post configuration.
Further, when the polling reaches the specified time slice, the task priority calculation and the task processor power dynamic adjustment task are triggered.
Further, the task priority calculation is specifically,
Figure BDA0002988824010000021
wherein R isikIs the kth task priority value in a certain time slice i;
wherein eta isiFor the ratio of computing resources to computing task volume in the ith time slice,
Figure BDA0002988824010000022
wherein eta isminThe ratio of the computing resource to the computing task amount is the smallest value in all time slices.
Wherein eta ismaxThe ratio of the computing resource to the computing task amount is the largest value in all time slices.
Wherein G isSThe computation time limit required for the kth task in a certain time slice i.
Wherein the content of the first and second substances,
Figure BDA0002988824010000023
the total number of computing resources in a certain time slice i.
Further, the dynamic adjustment of the task processor power is realized by allocating different processors to different processors in the i time slicesThe power of each task processor is dynamically adjusted, the power of each task processor accounts for the number of computing resources of each processor, and the power Z of the kth processor in the ith time sliceikAdjusting according to the calculation result of the ADMM algorithm, wherein the ADMM algorithm is as follows:
Figure BDA0002988824010000024
wherein Z isi-1,kThe power of the kth processor in the last time slice i-1.
Wherein, the first and the second end of the pipe are connected with each other,
Figure BDA0002988824010000025
the total power of the computing resources in the last time slice i-1, namely the total number of the computing resources.
Wherein the content of the first and second substances,
Figure BDA0002988824010000026
λ and λ are empirical coefficients.
Wherein phi and
Figure BDA0002988824010000031
in the case of a dual variable, the number of variables,
Figure BDA0002988824010000032
Zikthe computing storage resources divided for the kth task processor in the ith time slice, and the sum of the powers of all the processors in the ith time slice is
Figure BDA0002988824010000033
Total number of computing resources of
Figure BDA0002988824010000034
The computational resource obtained for the kth processor is
Figure BDA0002988824010000035
In addition, the invention also provides a human resource data distributed task processing system, which comprises:
the system comprises a timer, a task priority calculation module, a dynamic task processor power adjustment calculation module, a task processor warehouse, a data warehouse and a task warehouse;
the timer is used for dividing the event into at least one time slice, and triggering the task priority computing module and the task processor power dynamic adjustment computing module to perform the task when the specified time slice is polled;
the task priority calculating module is used for calculating task priority;
the task processor power dynamic adjustment calculation module is used for adjusting the task processor power;
the task processor warehouse is used for storing the task processors;
the task warehouse is used for receiving a computing task input by a user;
the data warehouse is used for storing system data.
Further, the task priority calculating module calculates the priority specifically as,
Figure BDA0002988824010000036
wherein R isikIs the kth task priority value in a certain time slice i;
wherein eta isiFor the ratio of computing resources to computing task volume in the ith time slice,
Figure BDA0002988824010000037
wherein etaminThe ratio of the computing resource to the computing task amount is the smallest value in all time slices.
Wherein eta ismaxThe ratio of the computing resource to the computing task amount is the largest value in all time slices.
Wherein, GSRequired for the kth task in a certain time slice iA time limit is calculated.
Wherein the content of the first and second substances,
Figure BDA0002988824010000041
the total number of computing resources in a certain time slice i.
Further, the dynamic adjustment of task processor power is specifically that, in the i time slices, the power of each task processor is dynamically adjusted by allocating different computing resources to different processors, the power of each task processor accounts for the number of computing resources of each processor, and in the i time slice, the power Z of the kth processorikAdjusting according to the calculation result of the ADMM algorithm, wherein the ADMM algorithm is as follows:
Figure BDA0002988824010000042
wherein, Zi-1,kThe power of the kth processor in the last time slice i-1.
Wherein the content of the first and second substances,
Figure BDA0002988824010000043
the total power of the computing resources in the last time slice i-1, namely the total number of the computing resources.
Wherein the content of the first and second substances,
Figure BDA0002988824010000044
and λ is an empirical coefficient.
Wherein phi and
Figure BDA0002988824010000045
in the case of a dual variable, the number of variables,
Figure BDA0002988824010000046
Zikthe computing storage resources divided for the kth task processor in the ith time slice, and the sum of the powers of all the processors in the ith time slice is
Figure BDA0002988824010000047
Total number of computing resources of
Figure BDA0002988824010000048
The computational resource obtained for the kth processor is
Figure BDA0002988824010000049
The invention provides a method and a system for processing human resource data distributed tasks, which divide human resource events into time slices, calculate and sequence the priorities of the time slices, obtain the power of a task processor through a priority calculation result and adjust the power of the task processor, solve the problem of power adjustment of the task processor, improve the utilization rate of hardware resources, realize the optimal configuration of distributed calculation on different tasks, solve the problem of insufficient data calculation capacity of human resource enterprises, and improve the speed of big data calculation and the efficiency of human resource configuration.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the description below are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flowchart of a human resources data distributed task processing method of the present invention;
FIG. 2 is a block diagram of a human resources data distributed task processing system according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments obtained by workers skilled in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.
In order to solve the problems of being suitable for processing of massive human resource big data and achieving rapid calculation of massive data of human resources, the invention provides a human resource data distributed task processing method and system, which can be applied to processing of massive human resource big data, achieve parallelized data processing, solve the problem of insufficient data calculation capacity of human resource enterprises, and improve the speed of big data calculation and the working efficiency of each subsystem module of a human resource system.
A human resource data distributed task processing method, as shown in fig. 1, the method includes:
step 101, acquiring a human resource event to be calculated;
capturing an event to be calculated of the human resource platform through the human resource platform, wherein the event can be a common event in human resources such as resume pushing, resume screening and post configuration.
Step 102, dividing an event into at least one time slice;
the timer module divides the event into i time slices, and Ti represents the ith time slice. When the polling reaches the specified time slice, the task priority calculation and the task processor power dynamic adjustment task are triggered.
103, dynamically calculating task priorities in different time slice processes;
the task processor warehouse initializes n task processors, and each task processor allocates computing resources such as a CPU, a memory, a video memory and the like. The j-th task handler gets Zji the computation memory resource in the i-th time slice.
And the task priority computing module is used for dynamically evaluating the priority of each task. Suppose that the kth task priority value in a time slice i is RikThe task priority is calculated by the following algorithm:
Figure BDA0002988824010000061
wherein etaiComputing resources and computing tasks for the ith time sliceThe ratio of the amount of traffic,
Figure BDA0002988824010000062
wherein eta isminThe ratio of the computing resource to the computing task amount is the smallest value in all time slices.
Wherein eta ismaxThe ratio of the computing resource to the computing task amount is the largest value in all time slices.
Wherein G isSThe computation time limit required for the kth task in a certain time slice i.
Wherein, the first and the second end of the pipe are connected with each other,
Figure BDA0002988824010000063
the total number of computing resources in a certain time slice i.
Step 104, adjusting the power of each task processor;
a dynamic adjustment calculation module for task processor power, which dynamically adjusts the power of each task processor by allocating different calculation resources to different processors in the i time slices, wherein the power of each task processor is the number of calculation resources occupied by each processor, and the power Z of the kth processor in the ith time sliceikAdjusting according to the calculation result of the ADMM algorithm to realize the optimized configuration, wherein the ADMM algorithm is as follows:
Figure BDA0002988824010000064
wherein Z isi-1,kThe power of the kth processor in the last time slice i-1.
Wherein the content of the first and second substances,
Figure BDA0002988824010000071
the total power of the computing resources in the last time slice i-1, namely the total number of the computing resources.
Wherein the content of the first and second substances,
Figure BDA0002988824010000072
and λ is an empirical coefficient.
Wherein phi and
Figure BDA0002988824010000073
are dual variables.
Figure BDA0002988824010000074
ZikThe computing storage resources divided for the kth task processor in the ith time slice, and the sum of the powers of all the processors in the ith time slice is
Figure BDA0002988824010000075
Total number of computing resources of
Figure BDA0002988824010000076
The computational resource obtained for the kth processor is
Figure BDA0002988824010000077
ZikThe computing storage resources divided for the kth task processor in the ith time slice, and the sum of the powers of all the processors in the ith time slice is
Figure BDA0002988824010000078
Total number of computing resources of
Figure BDA0002988824010000079
The computational resource obtained by the kth processor is
Figure BDA00029888240100000710
Specifically, taking actual human resources data processing as an example:
the set time period is 10 minutes, the total calculation task is that 2458900 ten thousand human resource data analysis calculation tasks are executed, the time slice is 1 minute, the total time slice number is 10, the number of processors is 10, the total number of calculation resources is 8 20 core intel to strong processor 2.2ghz, 10TGB memory model number is 2133m, 20 blocks of 1.92T solid state disk RAID0), for example, the 5 th processor in the first time slice initially allocated power is 10%, and the 5 th processor in the second time slice has power Z2512.8%, in the third time slice, the fifth processor power adopts ADMM algorithm
Figure BDA0002988824010000081
Wherein the content of the first and second substances,
Figure BDA0002988824010000082
the total power of the computing resources in the last time slice i-1 is 100%,
Figure BDA0002988824010000083
in the last time slice, the actual power used is 95%,
wherein the content of the first and second substances,
Figure BDA0002988824010000084
and λ are empirical coefficients 3 and 5 (which are currently the best coefficients), and λ is an empirical coefficient 10. Wherein phi and
Figure BDA0002988824010000085
is a dual variable, and is characterized in that,
Figure BDA0002988824010000086
Figure BDA0002988824010000087
wherein etaiCalculating the ratio of the computing resource to the computing task amount in the ith time slice;
wherein etaminThe ratio of the computing resource to the computing task amount in all time slices is the minimum value,
Figure BDA0002988824010000088
wherein eta ismaxIs the value with the maximum ratio of computing resources to computing task amount in all time slices
Figure BDA0002988824010000089
Wherein G isSA time limit 189.547 seconds is calculated for the 5 th task in a certain time slice 3. The purpose of dynamic adjustment of the power of the task processor is to balance load pressure and uniformly distribute calculation tasks.
The task repository is used for receiving user input computing tasks. The data warehouse is used for storing system data.
In addition, the present invention further provides a human resource data distributed task processing system, as shown in fig. 2, the system includes: a timer D1, a task priority calculation module D2, a task processor power dynamic adjustment calculation module D3, a task processor repository D, a data repository E1, and a task repository E2.
The system firstly obtains a human resource event to be calculated, and captures the human resource event to be calculated through a human resource platform, wherein the event can be common events in human resources such as resume pushing, resume screening and post configuration.
The timer D1 divides the event into i time slices, and Ti represents the ith time slice. When polling to the specified time slice, the task priority computation module D2 and the task processor power dynamic adjustment computation module D3 task are triggered.
The task processor warehouse D initializes n task processors, and each task processor allocates computing resources such as a CPU, a memory, a video memory and the like. The j-th task handler has a computation memory resource of Zji in the i-th time slice.
The task priority calculating module D2 is used for dynamically evaluating the priority of each task. Suppose that the kth task priority value in a time slice i is RikThe task priority is calculated by the following algorithm:
Figure BDA0002988824010000091
wherein etaiFor the ratio of the computing resources to the computing task volume in the ith time slice,
Figure BDA0002988824010000092
wherein eta isminThe ratio of the computing resource to the computing task amount is the smallest value in all time slices.
Wherein eta ismaxThe ratio of the computing resource to the computing task amount is the largest value in all time slices.
Wherein G isSThe computation time limit required for the kth task in a certain time slice i.
Wherein, the first and the second end of the pipe are connected with each other,
Figure BDA0002988824010000093
the total number of computing resources in a certain time slice i.
The dynamic adjustment calculation module D3 for task processor power dynamically adjusts the power of each task processor by allocating different calculation resources to different processors in the i time slices, the power of each task processor is the number of calculation resources occupied by each processor, and the power Z of the kth processor in the i time sliceikAdjusting according to the calculation result of the ADMM algorithm, wherein the ADMM algorithm is as follows:
Figure BDA0002988824010000101
wherein Z isi-1,kThe power of the kth processor in the last time slice i-1.
Wherein the content of the first and second substances,
Figure BDA0002988824010000102
the total power of the computing resources in the last time slice i-1, namely the total number of the computing resources.
Wherein the content of the first and second substances,
Figure BDA0002988824010000103
and λ is an empirical coefficient.
Wherein phi and
Figure BDA0002988824010000104
are dual variables.
Figure BDA0002988824010000105
ZikThe computing storage resources divided for the kth task processor in the ith time slice, and the sum of the powers of all the processors in the ith time slice is
Figure BDA0002988824010000106
Total number of computing resources of
Figure BDA0002988824010000107
The computational resource obtained for the kth processor is
Figure BDA0002988824010000108
Specifically, taking actual human resource data processing as an example:
setting the time period to 10 minutes, the total computing task is to execute 2458900 ten thousand human resource data analysis computing tasks, the time slice is 1 minute, the total number of time slices is 10, the number of processors is 10, and the total number of computing resources is (8 20 core intel to strong processor 2.2ghz, 10TGB memory model 2133m, 20 blocks of 1.92T solid state disk RAID0), for example, in the first time slice, the power initially allocated to the 5 th processor is 10%, in the second time slice, the power of the 5 th processor is Z2512.8%, in the third time slice, the fifth processor power adopts ADMM algorithm
Figure BDA0002988824010000109
Wherein the content of the first and second substances,
Figure BDA0002988824010000111
is the last hourIn the inter-slice i-1, the total power of the computing resources is 100%,
Figure BDA0002988824010000112
in the last time slice, the actual power used is 95%,
wherein the content of the first and second substances,
Figure BDA0002988824010000113
and λ are empirical coefficients 3 and 5 (which are currently the best coefficients), and λ is an empirical coefficient 10. Wherein phi and
Figure BDA0002988824010000114
in the case of a dual variable, the number of variables,
Figure BDA0002988824010000115
Figure BDA0002988824010000116
wherein eta isiCalculating the ratio of the computing resource to the computing task amount in the ith time slice;
wherein etaminThe ratio of the computing resource to the computing task amount in all time slices is the minimum value,
Figure BDA0002988824010000117
wherein etamaxIs the value with the maximum ratio of computing resources to computing task amount in all time slices
Figure BDA0002988824010000118
Wherein G isSA time limit 189.547 seconds is calculated for the 5 th task in a certain time slice 3. The purpose of dynamic adjustment of the power of the task processor is to balance load pressure and uniformly distribute calculation tasks.
Task repository E1 is used to receive user input computing tasks. Data warehouse E2 is used to store system data.
The invention provides a method and a system for processing human resource data distributed tasks, which divide human resource events into time slices, calculate and sequence the priorities of the time slices, obtain the power of a task processor through a priority calculation result and adjust the power of the task processor, solve the problem of power adjustment of the task processor, improve the utilization rate of hardware resources, realize the optimal configuration of distributed calculation on different tasks, solve the problem of insufficient data calculation capacity of human resource enterprises, and improve the speed of big data calculation and the efficiency of human resource configuration.
The principles and embodiments of the present invention are explained herein using specific examples, which are presented only to assist in understanding the method and its core concepts. It should be noted that, for the person skilled in the art, without departing from the principle of the invention, several improvements and modifications can be made to the invention, and these improvements and modifications also fall within the scope of protection of the claims of the invention.

Claims (7)

1. A human resources data distributed task processing method, the method comprising:
step 101, acquiring a human resource event to be calculated;
step 102, dividing an event into at least one time slice; when the polling reaches the specified time slice, triggering task priority calculation and task processor power dynamic adjustment task; the task processor power dynamic adjustment specifically includes that in i time slices, different computing resources are allocated to different processors, the power of each task processor is dynamically adjusted, the power of each task processor occupies the number of computing resources for each processor, and in the ith time slice, the power Zik of the kth processor is adjusted according to the result of the ADMM algorithm, wherein the ADMM algorithm is as follows:
Figure FDA0003685980400000011
wherein Zi-1, k is the power of the kth processor in the last time slice i-1;
rik is the kth task priority value in a certain time slice i;
wherein the content of the first and second substances,
Figure FDA0003685980400000012
calculating the total power of resources in the last time slice i-1, namely the total number of the calculated resources;
wherein the content of the first and second substances,
Figure FDA0003685980400000013
and λ is an empirical coefficient;
wherein phi and
Figure FDA0003685980400000014
in the case of a dual variable, the number of variables,
Figure FDA0003685980400000015
Figure FDA0003685980400000016
zik is the calculation storage resource divided by the kth task processor in the ith time slice, the sum of the power of all the processors is
Figure FDA0003685980400000017
Total number of computing resources of
Figure FDA0003685980400000021
The computational resource obtained by the kth processor is
Figure FDA0003685980400000022
103, dynamically calculating task priorities in different time slice processes;
step 104, adjusting the power of each task handler.
2. The method of claim 1, wherein the event can be resume push, resume filtering, and post configuration.
3. The method according to claim 1, wherein the task priority calculation is in particular,
Figure FDA0003685980400000023
wherein η i is the ratio of the computing resources to the computing task amount in the ith time slice,
Figure FDA0003685980400000024
wherein, η min is a value with the minimum ratio of the computing resources to the computing task amount in all time slices;
wherein η max is a value at which the ratio of the computing resources to the computing task amount is the maximum in all the time slices;
GS is a calculation time limit required by a kth task in a certain time slice i;
wherein the content of the first and second substances,
Figure FDA0003685980400000025
the total number of all computing resources in a certain time slice i.
4. A human resources data distributed task processing system, the system comprising:
the system comprises a timer, a task priority calculation module, a dynamic task processor power adjustment calculation module, a task processor warehouse, a data warehouse and a task warehouse;
the timer is used for dividing the event into at least one time slice, and triggering the task priority computing module and the task processor power dynamic adjustment computing module to perform the task when the specified time slice is polled; the task processor power dynamic adjustment specifically includes that in i time slices, different computing resources are allocated to different processors, the power of each task processor is dynamically adjusted, the power of each task processor occupies the number of computing resources for each processor, and in the ith time slice, the power Zik of the kth processor is adjusted according to the result of the ADMM algorithm, wherein the ADMM algorithm is as follows:
Figure FDA0003685980400000031
wherein Zi-1, k is the power of the kth processor in the last time slice i-1;
wherein, Rik is the kth task priority value in a certain time slice i;
wherein the content of the first and second substances,
Figure FDA0003685980400000032
calculating the total power of resources in the last time slice i-1, namely the total number of the calculated resources;
wherein the content of the first and second substances,
Figure FDA0003685980400000033
and λ is an empirical coefficient;
wherein phi and
Figure FDA0003685980400000034
is a dual variable, and is characterized in that,
Figure FDA0003685980400000035
Figure FDA0003685980400000036
zik is the calculation storage resource divided by the kth task processor in the ith time slice, the sum of the power of all the processors is
Figure FDA0003685980400000037
Total number of computing resources of
Figure FDA0003685980400000041
The computational resource obtained by the kth processor is
Figure FDA0003685980400000042
The task priority calculating module is used for calculating task priority;
the task processor power dynamic adjustment calculation module is used for adjusting the task processor power;
the task processor warehouse is used for storing the task processors;
the task warehouse is used for receiving a computing task input by a user;
the data warehouse is used for storing system data.
5. The system according to claim 4, wherein the task priority calculating module calculates the priority by, in particular,
Figure FDA0003685980400000043
wherein η i is the ratio of the computing resources to the computing task amount in the ith time slice,
Figure FDA0003685980400000044
wherein, η min is a value with the minimum ratio of the computing resources to the computing task amount in all time slices;
wherein η max is a value at which the ratio of the computing resources to the computing task amount is the maximum in all the time slices;
GS is the calculation time limit required by the k-th task in a certain time slice i;
wherein, the first and the second end of the pipe are connected with each other,
Figure FDA0003685980400000045
the total number of all computing resources in a certain time slice i.
6. An electronic device, comprising: a processor, a memory storing machine readable instructions executable by the processor, the processor executing the machine readable instructions to perform the steps of the method according to any one of claims 1 to 3 when the electronic device is run.
7. A computer-readable storage medium, having stored thereon a computer program which, when being executed by a processor, is adapted to carry out the steps of the method according to any one of claims 1 to 3.
CN202110308427.8A 2021-03-23 2021-03-23 Human resource data distributed task processing method and system Active CN113010273B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110308427.8A CN113010273B (en) 2021-03-23 2021-03-23 Human resource data distributed task processing method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110308427.8A CN113010273B (en) 2021-03-23 2021-03-23 Human resource data distributed task processing method and system

Publications (2)

Publication Number Publication Date
CN113010273A CN113010273A (en) 2021-06-22
CN113010273B true CN113010273B (en) 2022-07-19

Family

ID=76405385

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110308427.8A Active CN113010273B (en) 2021-03-23 2021-03-23 Human resource data distributed task processing method and system

Country Status (1)

Country Link
CN (1) CN113010273B (en)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104346220A (en) * 2013-07-31 2015-02-11 中国科学院计算技术研究所 Task scheduling method and system
CN107589993A (en) * 2017-08-15 2018-01-16 郑州云海信息技术有限公司 A kind of dynamic priority scheduling algorithm based on linux real time operating systems
CN109298940A (en) * 2018-09-28 2019-02-01 考拉征信服务有限公司 Calculation task allocating method, device, electronic equipment and computer storage medium
CN109639595A (en) * 2018-09-26 2019-04-16 北京云端智度科技有限公司 A kind of CDN dynamic priority scheduling algorithm based on time delay
CN110347504A (en) * 2019-06-28 2019-10-18 中国科学院空间应用工程与技术中心 Many-core computing resource dispatching method and device
CN110362388A (en) * 2018-04-11 2019-10-22 中移(苏州)软件技术有限公司 A kind of resource regulating method and device
CN112286637A (en) * 2020-10-30 2021-01-29 西安万像电子科技有限公司 Method and device for adjusting computing resources

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8862914B2 (en) * 2010-02-26 2014-10-14 Microsoft Corporation Virtual machine power consumption measurement and management
CN111277437B (en) * 2020-01-17 2022-11-22 全球能源互联网研究院有限公司 Network slice resource allocation method for smart power grid

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104346220A (en) * 2013-07-31 2015-02-11 中国科学院计算技术研究所 Task scheduling method and system
CN107589993A (en) * 2017-08-15 2018-01-16 郑州云海信息技术有限公司 A kind of dynamic priority scheduling algorithm based on linux real time operating systems
CN110362388A (en) * 2018-04-11 2019-10-22 中移(苏州)软件技术有限公司 A kind of resource regulating method and device
CN109639595A (en) * 2018-09-26 2019-04-16 北京云端智度科技有限公司 A kind of CDN dynamic priority scheduling algorithm based on time delay
CN109298940A (en) * 2018-09-28 2019-02-01 考拉征信服务有限公司 Calculation task allocating method, device, electronic equipment and computer storage medium
CN110347504A (en) * 2019-06-28 2019-10-18 中国科学院空间应用工程与技术中心 Many-core computing resource dispatching method and device
CN112286637A (en) * 2020-10-30 2021-01-29 西安万像电子科技有限公司 Method and device for adjusting computing resources

Also Published As

Publication number Publication date
CN113010273A (en) 2021-06-22

Similar Documents

Publication Publication Date Title
US8954971B2 (en) Data collecting method, data collecting apparatus and network management device
CN107247651B (en) Cloud computing platform monitoring and early warning method and system
US10678596B2 (en) User behavior-based dynamic resource capacity adjustment
CN108874535B (en) Task adjusting method, computer readable storage medium and terminal device
CN109981744B (en) Data distribution method and device, storage medium and electronic equipment
CN104731595A (en) Big-data-analysis-oriented mixing computing system
CN108270805B (en) Resource allocation method and device for data processing
WO2016184048A1 (en) Method and device for frequency management for multi-core processor cpu
Alyouzbaki et al. Novel load balancing approach based on ant colony optimization technique in cloud computing
CN114064261A (en) Multi-dimensional heterogeneous resource quantification method and device based on industrial edge computing system
CN117135131A (en) Task resource demand perception method for cloud edge cooperative scene
CN113010273B (en) Human resource data distributed task processing method and system
Jeyakumar et al. A study on MX/G (a, b)/1 queue with server breakdown without interruption and controllable arrivals during multiple adaptive vacations
CN115562841B (en) Cloud video service self-adaptive resource scheduling system and method
CN111860810A (en) Neural network operation method, device and equipment based on FPGA
Poltavtseva et al. Planning of aggregation and normalization of data from the Internet of Things for processing on a multiprocessor cluster
CN116166427A (en) Automatic capacity expansion and contraction method, device, equipment and storage medium
CN114253688A (en) Method and application for rescheduling application load in cloud environment
Chen et al. Cloud-edge collaborative data processing architecture for state assessment of transmission equipments
CN112187894A (en) Container dynamic scheduling method based on load correlation prediction
CN117971509B (en) Heterogeneous computing power cluster operation performance optimization method, heterogeneous computing power cluster operation performance optimization device, heterogeneous computing power cluster operation performance optimization equipment and medium
CN111160283A (en) Data access method, device, equipment and medium
US10235414B1 (en) Iterative kurtosis calculation for streamed data using components
CN117611425B (en) Method, apparatus, computer device and storage medium for configuring computing power of graphic processor
Wang et al. DDNN based data allocation method for IIoT

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