Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

SKA task scheduling system and method based on genetic algorithm and computing topology model

A task scheduling and topology model technology, applied in the field of big data, can solve problems such as characteristics that are not considered, and achieve the effects of high throughput, short task completion time, and high throughput

Active Publication Date: 2021-01-08
SHANGHAI JIAO TONG UNIV
View PDF9 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, because of its fixed scheduling order, this solution does not take into account the characteristics of different tasks in different scenarios and adapt measures to local conditions, so it is not an optimal solution.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • SKA task scheduling system and method based on genetic algorithm and computing topology model
  • SKA task scheduling system and method based on genetic algorithm and computing topology model
  • SKA task scheduling system and method based on genetic algorithm and computing topology model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0053] According to a kind of SKA task scheduling system based on genetic algorithm and computing topology model provided by the present invention, comprising:

[0054] Module M1: The server forms a vector X according to the processing time required by each subtask of the parallel task, obtains the data processing capability of each node according to the number of CPU cores, and forms the data processing capability of each node into a vector Y, according to the vector X and the vector Y And parallel task allocation scheme A, constructing a computing topology model;

[0055] Module M2: The server uses the genetic algorithm to obtain the suboptimal scheduling scheme based on the dependencies between the subtasks of the parallel tasks and the completion time required for each task;

[0056] Module M3: The server obtains the suboptimal scheduling scheme by calculating the topology model to obtain the task scheduling scheme, obtains the completion time required for each task accord...

Embodiment 2

[0082] Embodiment 2 is a modification of embodiment 1

[0083] The following is a detailed introduction in combination with specific examples. This example is carried out on the premise of the technical solution of the present invention, and a detailed implementation mode and specific operation process are given.

[0084] First, we use a parallel task job as our example. The computing topology of the job is as follows figure 1 shown.

[0085] We convert the dependencies between tasks of the job into an input file whose format is as follows:

[0086]

[0087]

[0088] There are three numbers in the first row, respectively representing the total number of tasks (41), the number of dependencies between tasks (120), that is, the number of edges in the graph and the number of computing nodes (7). The second row is the running time of each task. The running time of task No. 0 is at the first, the running time of task No. 1 is at the second, and so on, separated by spaces. ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides an SKA task scheduling system and method based on a genetic algorithm and a computing topology model, and the system comprises: a module M1 which builds the computing topology model according to a vector X formed by the processing time needed by each subtask of a parallel task, a vector Y formed by the data processing capability of each node, and a parallel task distributionscheme A; a module M2 which is used for obtaining a suboptimal scheduling scheme through a genetic algorithm according to the dependency relationship among the sub-tasks of the parallel tasks and thecompletion time required by each task; and a module M3 which is used for obtaining a task scheduling scheme by calculating a topological model according to the obtained suboptimal scheduling scheme,obtaining the completion time required by each task according to the task scheduling scheme, and repeatedly triggering the modules M2 to M3 until the number of iterations reaches a preset number of times, so as to obtain an optimal SKA task scheduling scheme. The system and the method are simple in input, and a user can conveniently and quickly construct a task-dependent topological graph file andthen take the task-dependent topological graph file as the input of an algorithm.

Description

technical field [0001] The present invention relates to the field of big data, in particular, to a SKA task scheduling system and method based on a genetic algorithm and a computational topology model, and more specifically, to a method for a square kilometer array (SKA) radio telescope through a genetic algorithm and a computational topology model. Compute the topology model to find an optimal task scheduling scheme. Background technique [0002] With the development of data collection and storage technology, humans have accumulated a large amount of astronomical observation data. In astronomy, images make up an increasing proportion of this data. Therefore, it is very important to find the most suitable framework for big data processing of astronomical imagery, and the purpose of the astronomical image processing system is to help astronomers use the programming model easily by providing scalable and efficient means of data storage and analysis. [0003] The Square Kilom...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06F9/48G06N3/12
CPCG06F9/4881G06N3/126Y02D10/00
Inventor 骆源伏开宇
Owner SHANGHAI JIAO TONG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Eureka Blog
Learn More
PatSnap group products