Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A Correlation Structure for Hiding Multithreading in Heterogeneous Programming and a Mapping Method Based on It

A multi-threaded, heterogeneous technology, applied in the field of association structure and mapping based on it, can solve the problems of heterogeneous programming, scalability and portability, immature research and application, etc., to improve portability and scalability performance, simplified heterogeneous parallel programming logic, and clear code organization

Active Publication Date: 2020-07-28
XI AN JIAOTONG UNIV
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

These methods alleviate the difficulty of heterogeneous programming and the problems of scalability and portability to a certain extent, but the research and application are still immature

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
  • A Correlation Structure for Hiding Multithreading in Heterogeneous Programming and a Mapping Method Based on It
  • A Correlation Structure for Hiding Multithreading in Heterogeneous Programming and a Mapping Method Based on It
  • A Correlation Structure for Hiding Multithreading in Heterogeneous Programming and a Mapping Method Based on It

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0046] The present invention will be further described in detail below in conjunction with specific embodiments, which are explanations of the present invention rather than limitations.

[0047] On the basis of "program=data structure+algorithm", the present invention designs the associated grammatical structure based on the attribute system and specifies its semantic rules. By assigning additional attributes to the data to describe the corresponding relationship between the upper layer data and the calculation process of parallel processing, the parallel information of the upper layer application is preserved. The specific technical scheme adopted by the attribute system used to represent the association structure is as follows:

[0048] 1. Data input and output attributes

[0049] In a heterogeneous system, the heterogeneous processor has its own independent storage space and forms a separate storage architecture from the main memory of the system. Therefore, data needs to ...

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 discloses a multi-thread association structure in hidden heterogeneous parallel programming and a mapping method based on the structure and belongs to the technical field of computer programming languages. According to the method, input and output attributes of data, division attributes of single data to data interfaces corresponding to the calculation process and association properties of the attributes are designated through an attribute system, a mapping relation for calculation data to the data interfaces in the calculation process is described, and a compiling system can automatically complete environment mapping according to parallel information reserved in the association structure when an upper-layer application runs towards a bottom layer; therefore, the situation isavoided that the bound calculation process and data are displayed by a user through a multi-thread mode in the heterogeneous parallel programming process, and the hidden expression of parallelism isachieved. According to the multi-thread association structure, the heterogeneous parallel programming logic can be effectively simplified, the burden of programming personnel is reduced, and the portability and extensibility of the application are improved. The multi-thread association structure has high practical value and wide application prospect in the field of heterogeneous parallel programming and high-performance calculation.

Description

technical field [0001] The invention belongs to the technical field of computer programming languages, and in particular relates to a correlation structure for hiding heterogeneous programming multi-threads and a mapping method based thereon. Background technique [0002] With the increasing demand for computing power in various industries and the failure of Moore's Law, processors are developing in the direction of multi-core and many-core. The use of heterogeneous many-core accelerators and coprocessors has become an important means to improve the performance of computer systems. Existing mainstream heterogeneous parallel programming methods such as CUDA and OpenCL provide C-like high-level languages ​​and programming interfaces for programming on GPUs and many-core coprocessors, lowering the threshold for using heterogeneous processors, and at the same time using The idea of ​​multi-threaded programming, under the framework of program=data structure+algorithm, uses thread...

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 Patents(China)
IPC IPC(8): G06F8/41G06F8/76G06F8/73G06F9/50
CPCG06F8/45G06F8/73G06F8/76G06F9/5027
Inventor 董小社吴树森王龙翔朱正东刘强周志阳
Owner XI AN JIAOTONG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products