2012
DOI: 10.1145/2345156.1993516
|View full text |Cite
|
Sign up to set email alerts
|

Automatic CPU-GPU communication management and optimization

Abstract: The performance benefits of GPU parallelism can be enormous, but unlocking this performance potential is challenging. The applicability and performance of GPU parallelizations is limited by the complexities of CPU-GPU communication. To address these communications problems, this paper presents the first fully automatic system for managing and optimizing CPU-GPU communcation. This system, called the CPU-GPU Communication Manager (CGCM), consists of a run-time library and a set of compiler transformations that w… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
48
0

Year Published

2012
2012
2015
2015

Publication Types

Select...
5
1
1

Relationship

0
7

Authors

Journals

citations
Cited by 33 publications
(48 citation statements)
references
References 15 publications
0
48
0
Order By: Relevance
“…It relies on pragmas, but it does not ease the programmer task as much as the Cray, CAPS or PGI compilers. Other experimental compilation tools like CGCM [16] and PAR4ALL [1] aim at automating the process of CPU-GPU communication and the detection of the pieces of code that can run in parallel. The work by Lee and Eigenmann [20] proposes OpenMPC, an API to facilitate translation of OpenMP programs to CUDA, and a compilation system to support it.…”
Section: Related Workmentioning
confidence: 99%
“…It relies on pragmas, but it does not ease the programmer task as much as the Cray, CAPS or PGI compilers. Other experimental compilation tools like CGCM [16] and PAR4ALL [1] aim at automating the process of CPU-GPU communication and the detection of the pieces of code that can run in parallel. The work by Lee and Eigenmann [20] proposes OpenMPC, an API to facilitate translation of OpenMP programs to CUDA, and a compilation system to support it.…”
Section: Related Workmentioning
confidence: 99%
“…In a recent paper [18], Jablin et al introduce cgcm, a system targeting exactly the same issue. cgcm, just like our scheme is focused on transferring full allocation units.…”
Section: Comparison With Respect To a Fully Dynamic Approachmentioning
confidence: 99%
“…Earlier compiler-based approaches [2,8,11,12] insert data transfers statically and use increasingly sophisticated analyses and transformations to avoid redundant transfers. However, the conservativeness inherent in a compiler-only approach cannot avoid all redundant transfers.…”
Section: Coherence Checks Versus Data Transfersmentioning
confidence: 99%
“…The CPU-GPU Communications Manager (CGCM) [8] is a system that automates memory allocations and transfers of data between the CPU and GPU. In that work, CGCM is coupled to an autoparallelization system that produces GPU code from the programs written in C/C++.…”
Section: Cpu-gpu Communications Manager (Cgcm) For X10cudamentioning
confidence: 99%
See 1 more Smart Citation