Parallel computation method and system

A parallel computing and computing node technology, applied in the computer field, can solve problems affecting the execution efficiency of parallel computing, average task execution time and task throughput, and unreasonable task allocation.

Inactive Publication Date: 2014-03-05
NEUSOFT CORP
View PDF5 Cites 51 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0022] from figure 2 It can be seen from the above that the parallel computing method in the prior art can only allocate the tasks of the business system according to the load quota of the pre-set computing nodes. Once the task allocation is completed, there may be an unreasonable problem of task allocation
Unreasonable task allo

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
  • Parallel computation method and system
  • Parallel computation method and system
  • Parallel computation method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0084] This embodiment provides a image 3 Parallel computing methods shown, including:

[0085] Step S100, detecting unassigned tasks.

[0086] The unassigned tasks may be generated based on the following process:

[0087] Initialize the system or run the system;

[0088] Submit jobs on the system or system client;

[0089] The submitted job is divided to form tasks.

[0090] The tasks formed in the above process are newly generated tasks by the system and have not been assigned. In this embodiment, the newly generated tasks by the system are referred to as the unassigned tasks.

[0091] Step S101 , when an unassigned task is detected, monitor each computing node to obtain node monitoring data.

[0092] The computing nodes in this step can exist in the virtual environment of the virtual machine. When it is necessary to monitor each computing node to obtain node monitoring data, it can first connect to the virtualization environment through a virtualization connector, and...

Embodiment 2

[0115] This embodiment provides a Figure 5 Parallel computing methods shown, including:

[0116] Step S300, detecting tasks that have been allocated but not processed.

[0117] The tasks that have been assigned but not processed are based on the unassigned tasks described in Embodiment 1, and may be generated by the following process:

[0118] According to the parallel computing method as described in Embodiment 1, the unassigned tasks are assigned to computing nodes;

[0119] The unallocated tasks are left unprocessed for a long time to form the allocated but unprocessed tasks.

[0120] The task formed in the above process is not a new task generated by the system, but has been assigned to the computing node, but because the computing node has not processed the task, the task has been waiting for a long time, and the task has not been processed for a long time. It means that the task has been assigned to the computing node and has not been processed for more than 5 minute...

Embodiment 3

[0128] This embodiment provides a Image 6 The parallel computing method shown is based on Embodiment 1 and Embodiment 2. The method includes the process of allocating unallocated node tasks and reassigning allocated but unprocessed node tasks, including:

[0129] Step S400, initialize the system.

[0130] Step S401, detecting whether there is a new task assignment.

[0131] When there is a new task to be assigned, perform steps S402 to S405:

[0132] Step S402, an updating step.

[0133] The updating step is: updating the comprehensive load queue of computing nodes in real time.

[0134] Step S403, check the steps.

[0135] The viewing step is: viewing the node task comprehensive load queue.

[0136] Step S404, the allocation step.

[0137] The respective steps are: selecting the node with the smallest load in the integrated load queue of node tasks, and assigning the new task to the node with the smallest load.

[0138] Step S405, execute the step.

[0139] The execu...

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 relates to a parallel computation method and system. The method includes: monitoring each computing node to acquire node monitoring data; calculating load capacity of each computing node according to the node monitoring data; assigning tasks to be assigned according to the load capacities of the computing nodes. Correspondingly, the system comprises a node monitoring unit, a task load management unit and a task scheduling unit. Through the application of the technical scheme, task can be reasonably assigned during parallel computation and parallel computation can be executed more efficiently.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a parallel computing method and system. Background technique [0002] The current business system is faced with the supersaturated load of computing resources, resulting in the problem of low performance of the business system. If this problem is solved by upgrading the hardware, it will face a series of negative effects such as ineffective performance improvement and huge capital expenditure. The existing technology uses distributed computing technology to solve the above problems. [0003] Distributed computing technologies include mobile computing and cloud computing. Large-scale application services in a distributed environment are often decomposed into multiple tasks for scheduling and execution. Distributed computing technology can optimize the load of computing nodes in the business system, solve the problem of oversaturation of computing node resources faced by the bus...

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): G06F9/50
Inventor 李金宝
Owner NEUSOFT CORP
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
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap