Method for realizing automatic pipeline parallelism

An implementation method and a streamlined technology, applied in the field of program compilation, can solve the problems that restrict the efficient use and popularization of high-performance computers, backwardness, etc., and achieve the effects of reducing communication overhead, increasing optimization capabilities, and achieving load balancing.

Inactive Publication Date: 2011-01-12
FUDAN UNIV
View PDF2 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The relative backwardness of parallel compilation technology restricts the e...

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
  • Method for realizing automatic pipeline parallelism
  • Method for realizing automatic pipeline parallelism
  • Method for realizing automatic pipeline parallelism

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0016] The specific operation steps of the present invention will be further described below.

[0017] figure 1 Shown is the entire automatic parallel processing flow of the present invention.

[0018] First, determine the dependency type of the loop structure

[0019] Automatic parallel optimization first converts the source program into an intermediate representation. The intermediate representation of the program abstracts the source program in a structured form, and records various information collected and generated during the program analysis and optimization process, which provides the required program information support for each stage of program analysis, transformation and optimization. Based on the intermediate representation of the program, unroll the parallel type identification of the loop structure, marked as fully parallel, pipeline parallel or serial.

[0020] 1. Through traditional data dependency analysis and dependency testing, for loops without data dep...

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 belongs to the technical field of program compilation and in particular relates to a method for realizing automatic pipeline parallelism. The method of the invention mainly comprises the following steps of: (1) identification of the pipeline parallelism, namely judging a loop structure which is provided with cross-loop iteration dependence and a dependence distance vector is a constant; (2) synchronization among threads, namely inserting the synchronization according to the dependence distance vector and deleting the redundant synchronization with the same distance vector; and (3) thread scheduling in a static step length, namely self-defining a thread scheduling strategy for balancing the workload of each thread and reducing the communication expense. The type identification of the loop structure is depended on the conventional array data stream analysis and dependence tests, while the pipeline parallelism only processes the regular loop structure with backward cross-loop iteration. The synchronization expense of the pipeline parallelism is high, so the pipeline parallelism is only performed on the outmost layer of a nested loop. Profit of the pipeline parallelism depends on programs, the number of the cyclic iteration is larger and the dependence distance is longer, the performance promotion is greater. The method for realizing the pipeline parallelism improvesthe capacity of automatic parallel optimization and contributes to further improving the performance of scientific calculation programs.

Description

technical field [0001] The invention belongs to the technical field of program compilation, and in particular relates to a method for realizing automatic pipeline parallelism. Background technique [0002] Modern computer architecture is developing towards multi-core and multi-thread direction, which poses new challenges to automatic program parallelization technology. The relative backwardness of parallel compilation technology restricts the efficient use and popularization of high-performance computers in certain procedures. Automatically compiling serial programs into parallel programs is of great significance for taking advantage of multi-core, improving program performance and promoting the development of parallel computer systems. [0003] Loop is the structure with the richest parallelism in the program, and it is often the most time-consuming part of program execution. The essence of automatic parallel optimization of serial programs is the parallel analysis of loo...

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): G06F9/38
Inventor 杨克峤李弋臧斌宇
Owner FUDAN 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
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