Task scheduling device, task scheduling method and data parallel processing device

A technology of task processing and task scheduling, applied in the direction of machine execution devices, concurrent instruction execution, etc., can solve the problems of reduced stability and reliability, difficulty in controlling power consumption and temperature, failure of integrated circuits, etc., to achieve shortened time and increased speed , the effect of increased speed

Inactive Publication Date: 2015-06-17
中芯睿智(北京)微电子科技有限公司
View PDF4 Cites 23 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The integration level of integrated circuits has an upper limit. When the integration level of integrated circuits is so high that the width of the lines in the integrated circuit reaches the nanometer level, the design and manufacture of integrated circuits become very complicated, and the stability and reliability are reduced. Testable It is difficult to control the power consumption temperature, etc., which often leads to the failure of the integrated circuit, which in turn leads to an upper limit on the speed at which the terminal processes data.

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
  • Task scheduling device, task scheduling method and data parallel processing device
  • Task scheduling device, task scheduling method and data parallel processing device
  • Task scheduling device, task scheduling method and data parallel processing device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0044] see Figure 1A A schematic diagram of the task scheduling device shown, the task scheduling device is used to schedule the task information in the parallel processing task queue, the device includes:

[0045] A statistics module 101, configured to determine idle processing units from a plurality of processing units, and count the number of idle processing units;

[0046] The scheduling module 102 is used to sequentially take out a corresponding number of task information from the head of the task queue according to the number of idle processing units;

[0047] The assignment module 103 is configured to assign the fetched task information to idle processing units, so that the idle processing units process the fetched task information.

[0048] Wherein, the statistical module 101 monitors the processing status of each processing unit included in the terminal or device in real time, and the processing status of the processing unit may be busy or idle. When it is monitored...

Embodiment 2

[0069] see figure 2 A schematic diagram of the device for parallel processing data shown, the device includes a memory 201, a processor 202 and a task scheduling device 203 as described in Embodiment 1;

[0070] The memory 201 is used to store data and executable instructions of the processor 202;

[0071] The processor 202 includes at least one processing unit, and the processing unit receives task information assigned by the task scheduling device 203 and performs corresponding processing.

[0072] In the device provided by the embodiment of the present invention, the idle processing units are first determined from the plurality of processing units, and the number of idle processing units is counted, and then a corresponding number of tasks are taken out from the task queue according to the number of idle processing units. For task information, the fetched task information is assigned to idle processing units, so that the idle processing units process the fetched task info...

Embodiment 3

[0074] see image 3 The flow chart of the task scheduling method is shown, the method can be executed by the task scheduling device, and is used to schedule the task information in the parallel processing task queue, and the method includes the following steps:

[0075] Step 301: determining idle processing units from a plurality of processing units, and counting the number of idle processing units;

[0076] Step 302: According to the number of idle processing units, sequentially take out the corresponding number of task information from the head of the task queue;

[0077] Step 303: Distribute the fetched task information to idle processing units, so that the idle processing units process the fetched task information.

[0078] In the method provided by the embodiment of the present invention, the idle processing units are first determined from the plurality of processing units, and the number of idle processing units is counted, and then a corresponding number of tasks are t...

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 a task scheduling device, a task scheduling method and a data parallel processing device. The task scheduling device comprises a statistic module, a scheduling module and a distribution module, wherein the statistic module is used for determining idle processing units from a plurality of processing units and carrying out statistics on the number of the idle processing units; the scheduling module is used for sequentially taking out task information with corresponding number by starting from the head of a task queue according to the number of the idle processing units; the distribution module is used for respectively distributing the task information which is taken out into the idle processing units so as to enable the task information which is taken out to be processed by the idle processing units. After the task scheduling device, the task scheduling method and the data parallel processing device are used, the data processing speed of the device can be continuously improved.

Description

technical field [0001] The present invention relates to the technical field of terminals, in particular, to a task scheduling device, method and equipment for parallel processing data. Background technique [0002] At present, with the continuous popularization of various terminals, people need to handle various daily affairs through the terminal, and people have higher and higher requirements for the speed of terminal data processing. Therefore, how to continuously improve the speed of terminal data processing has become an urgent problem. question. [0003] At present, the existing technology provides a method for increasing the speed of terminal data processing, including: the processor of the terminal is a large-scale integrated circuit, and the integrated circuit includes components such as arithmetic unit, memory and controller. By improving the process To improve the integration of integrated circuits, reduce the response time of components and the line delay between...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06F9/38
Inventor 王世好
Owner 中芯睿智(北京)微电子科技有限公司
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