FLAT: A GPU programming framework to provide embedded MPI

Takefumi Miyoshi, Hidetsugu Irie, Keigo Shima, Hiroki Honda, Masaaki Kondo, Tsutomu Yoshinaga

Research output: Chapter in Book/Report/Conference proceedingConference contribution

11 Citations (Scopus)

Abstract

For leveraging multiple GPUs in a cluster system, it is necessary to assign application tasks to multiple GPUs and execute those tasks with appropriately using communication primitives to handle data transfer among GPUs. In current GPU programming models, communication primitives such as MPI functions cannot be used within GPU kernels. Instead, such functions should be used in the CPU code. Therefore, programmer must handle both GPU kernel and CPU code for data communications. This makes GPU programming and its optimization very difficult. In this paper, we propose a programming framework named FLAT which enables programmers to use MPI functions within GPU kernels. Our framework automatically transforms MPI functions written in a GPU kernel into runtime routines executed on the CPU. The execution model and the implementation of FLAT are described, and the applicability of FLAT in terms of scalability and programmability is discussed. We also evaluate the performance of FLAT. The result shows that FLAT achieves good scalability for intended applications.

Original languageEnglish
Title of host publication5th Workshop on General-Purpose Computation on Graphics Processing Units, GPGPU-5 - Held in Cooperation with ACM ASPLOS XVII
Pages20-29
Number of pages10
DOIs
Publication statusPublished - 2012
Externally publishedYes
Event5th Workshop on General-Purpose Computation on Graphics Processing Units, GPGPU-5 - Held in Cooperation with ACM ASPLOS XVII - London, United Kingdom
Duration: 2012 Mar 32012 Mar 3

Publication series

NameACM International Conference Proceeding Series

Conference

Conference5th Workshop on General-Purpose Computation on Graphics Processing Units, GPGPU-5 - Held in Cooperation with ACM ASPLOS XVII
Country/TerritoryUnited Kingdom
CityLondon
Period12/3/312/3/3

Keywords

  • GPGPU
  • MPI
  • cluster system
  • programming framework

ASJC Scopus subject areas

  • Software
  • Human-Computer Interaction
  • Computer Vision and Pattern Recognition
  • Computer Networks and Communications

Fingerprint

Dive into the research topics of 'FLAT: A GPU programming framework to provide embedded MPI'. Together they form a unique fingerprint.

Cite this