Check patentability & draft patents in minutes with Patsnap Eureka AI!

Parallelism detection model-based serial program parallelism detection method and system, terminal and readable storage medium

A technology for detecting models and serial programs, applied in the computer field, can solve problems such as strong professionalism, program analysis limitations, and large time overhead, and achieve the effects of money cost savings, extremely high stability requirements, and equipment cost savings.

Inactive Publication Date: 2021-10-26
HUNAN UNIV
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0010] The purpose of the present invention is to provide a parallelism detection method, system, and terminal and readable storage medium to overcome

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
  • Parallelism detection model-based serial program parallelism detection method and system, terminal and readable storage medium
  • Parallelism detection model-based serial program parallelism detection method and system, terminal and readable storage medium
  • Parallelism detection model-based serial program parallelism detection method and system, terminal and readable storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0066] like figure 1 As shown, this embodiment provides a parallelism detection method for serial programs based on the parallelism detection model, which is mainly divided into two stages. The first stage is the processing from the source program data set to the source program embedded representation. This stage receives source code as input and converts it into a context flow chart, including four steps of refactoring, classification, compilation and construction. The second stage is to preprocess the graph structure constructed from the first stage to obtain a vector representation space named inst2vec and the node classification information obtained by using the analogy test, and combine the graph structure and classification labels constructed in the first stage Collaboratively input the graph neural network, and obtain the parallelism detection model through the automatic learning of the graph neural network model. Specifically, the method provided in this embodiment in...

Embodiment 2

[0101] Based on the parallelism detection method of a serial program based on the parallelism detection model provided in Embodiment 1, this embodiment provides a system based on the method, including: a source code acquisition module, an extraction module, and a program sample acquisition module , a classification module, a graph structure generation module, and a parallelism detection model generation module.

[0102] Wherein, the source code acquisition module is used to obtain the source code; the extraction module is used to extract each loop segment in the source code, and retain the remaining segments; the program sample acquisition module converts the loop segment to the original segment for each loop segment The position is inserted into the remaining fragments to reconstruct the program to obtain the program sample. The classification module is used to determine the classification label of the program sample; the graph structure generation module is used to construct...

Embodiment 3

[0106] This embodiment provides a terminal, which includes one or more processors; a memory for storing one or more programs;

[0107] When the one or more programs are executed by the one or more processors, the one or more processors are made to implement: the steps of a parallelism detection method for serial programs based on a parallelism detection model.

[0108] The terminal also includes: a communication interface, which is used for communicating with external devices and performing interactive data transmission.

[0109] Wherein, the memory may include a high-speed RAM memory, and may also include a non-volatile defibrillator, such as at least one disk memory.

[0110] If the memory, processor, and communication interface are implemented independently, the memory, processor, and communication interface can be connected to each other through a bus to complete mutual communication. The bus may be an industry standard architecture bus, a peripheral device interconnectio...

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 parallelism detection model-based serial program parallelism detection method and a system, a terminal and a readable storage medium, the method comprises the following steps: obtaining program samples and classification tags thereof, each program sample comprising a unique loop fragment, and the classification tags representing whether the program samples can be parallelized; constructing a graph structure of each program sample, and preferably representing the program samples by using a context flow chart; training a classification model or a neural network based on the graph structure of the program sample and the classification tag to obtain a parallelism detection model; adopting the parallelism detection model for detecting whether the target program can be parallelized or not. According to the method, a deep learning method based on data driving is utilized to automatically learn potential features of codes from a data set, graph structures are selected to represent parallel program features, and learning features are extracted on the basis of the graph structures to train a parallelism detection model; the problems that in an existing parallelism detection technology, time expenditure is large, program analysis is limited, and professionality is high are solved.

Description

technical field [0001] The invention belongs to the technical field of computers, and in particular relates to a parallelism detection method, system, terminal and readable storage medium of a serial program based on a parallelism detection model. Background technique [0002] While microprocessors have facilitated reductions in the performance and cost of computer applications, the process has stalled due to issues such as energy consumption. As a result, processor developers have shifted from the microprocessor model to the multi-core, multi-core processor model. Today, parallel computers are becoming more and more popular. The purpose of introducing and using parallel programming technology in the traditional programming process is to make full use of existing computer hardware and software resources, and process serial tasks more, faster and more efficiently. That is, through parallel programming, the computer can correctly process multiple tasks at the same time or pr...

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): G06F11/36G06K9/62G06N3/08
CPCG06F11/3684G06N3/08G06F18/24
Inventor 申园园吴强王世领彭蔓蔓
Owner HUNAN UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com