OpenMP for Accelerators

Beyer, James C.; Stotzer, Eric; Hart, Alistair; Supinski, Bronis R. de

doi:10.1007/978-3-642-21487-5_9

Cited by 35 publications

(25 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Most recently, quite a few directive-based GPU programming models have been proposed from both the research community (hiCUDA [13], OpenMPC [76] , etc.) and industry (PGI Accelerator [77], HMPP , R-Stream [78], OpenACC, OpenMP for Accelerators [79], etc.). On the surface, these models appear to offer different levels of abstraction, and expected programming effort for code restructuring and optimization.…”

Section: Analysis Of Gpgpu Frameworkmentioning

confidence: 99%

RT-CUDA: A Software Tool for CUDA Code Restructuring

Khan

Al-Mouhamed

Al-Mulhem

et al. 2016

Int J Parallel Prog

View full text Add to dashboard Cite

Section: Analysis Of Gpgpu Frameworkmentioning

confidence: 99%

RT-CUDA: A Software Tool for CUDA Code Restructuring

Khan

Al-Mouhamed

Al-Mulhem

et al. 2016

Int J Parallel Prog

View full text Add to dashboard Cite

“…For the purpose of this work, we use the version proposed by James Beyer et al [1]. Another set of directives with similar goals are available from PGI as the PGI accelerator directives [2].…”

Section: B Accelerator Extensionsmentioning

confidence: 99%

Heterogeneous Task Scheduling for Accelerated OpenMP

Scogland

Rountree

Feng

et al. 2012

2012 IEEE 26th International Parallel and Distributed Processing Symposium

Self Cite

View full text Add to dashboard Cite

Abstract-Heterogeneous systems with CPUs and computational accelerators such as GPUs, FPGAs or the upcoming Intel MIC are becoming mainstream. In these systems, peak performance includes the performance of not just the CPUs but also all available accelerators. In spite of this fact, the majority of programming models for heterogeneous computing focus on only one of these. With the development of Accelerated OpenMP for GPUs, both from PGI and Cray, we have a clear path to extend traditional OpenMP applications incrementally to use GPUs. The extensions are geared toward switching from CPU parallelism to GPU parallelism. However they do not preserve the former while adding the latter. Thus computational potential is wasted since either the CPU cores or the GPU cores are left idle. Our goal is to create a runtime system that can intelligently divide an accelerated OpenMP region across all available resources automatically. This paper presents our proof-of-concept runtime system for dynamic task scheduling across CPUs and GPUs. Further, we motivate the addition of this system into the proposed OpenMP for Accelerators standard. Finally, we show that this option can produce as much as a two-fold performance improvement over using either the CPU or GPU alone.

show abstract

“…Implementations specifically for GPUs are beginning to proliferate. Academic versions such as the OpenMP to GPGPU translator proposed by Eigenmann [58], as well as in industry with the release of HMPP [35], OpenACC [7,59], PGI Accelerator [95] and OpenMP for Accelerators OpenMP [19] are now all available. Each one offers a method for a user to target multiple GPUs and CPUs by explicitly splitting the workload and targeting each device with a region or codelet.…”

Section: Heterogeneous Programming Modelsmentioning

confidence: 99%

“…For the purpose of this work, we use the version proposed by James Beyer et al [19]. Another set of directives with similar goals are available from PGI as the PGI accelerator directives [95].…”

Section: Introductionmentioning

confidence: 99%

“…Rather than grapple with unfamiliar programming models, users often run their CPU-only code on accelerated resources, leaving a significant portion of the computing resources idle. Accelerated OpenMP [19], our term for a class of directive-based programming models including OpenMP for Accelerators [19] and the PGI accelerator model [95] among others, can ease this transition by allowing users to target accelerators with familiar OpenMP-style syntax. However, Accelerated OpenMP is not a panacea: current iterations help one move their computation to a single accelerator with straightforward syntax.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Runtime Adaptation for Autonomic Heterogeneous Computing

Scogland

Feng

2014

2014 14th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing

View full text Add to dashboard Cite

Heterogeneity is increasing across all levels of computing, with the rise of accelerators such as GPUs, FPGAs, and other coprocessors into everything from cell phones to supercomputers. More quietly it is increasing with the rise of NUMA systems, hierarchical caching, OS noise, and a myriad of other factors. As heterogeneity becomes a fact of life, efficiently managing heterogeneous compute resources is becoming a critical, and ever more complex, task. The focus of this dissertation is to lay the foundation for an autonomic system for heterogeneous computing, employing runtime adaptation to improve performance portability and performance consistency while maintaining or increasing programmability. We investigate heterogeneity arising from a myriad of factors, grouped into the dimensions of locality and capability. This work has resulted in runtime schedulers capable of automatically detecting and mitigating heterogeneity in physically homogeneous systems through MPI and adaptive coscheduling for physically heterogeneous accelerator based systems as well as a synthesis of the two to address multiple levels of heterogeneity as a coherent whole. We also discuss our current work towards the next generation of fine-grained scheduling and synchronization across heterogeneous platforms in the design of a highly-scalable and portable concurrent queue for many-core systems. Each component addresses aspects of the urgent need for automated management of the extreme and ever expanding complexity introduced by heterogeneity. I have also had the good fortune to collaborate with both Lawrence Livermore National Laboratory and Argonne National Laboratory. Experiences as an intern at each lab has had a significant effect on the final shape of my dissertation. Special thanks go to Dr. Pavan Balaji (again), Dr. Bronis de Supinski (again) and Dr. Barry Rountree. These collaborations proved to be major turning points for me, both in my research and my life as a whole.

show abstract

OpenMP for Accelerators

Cited by 35 publications

References 5 publications

RT-CUDA: A Software Tool for CUDA Code Restructuring

RT-CUDA: A Software Tool for CUDA Code Restructuring

Heterogeneous Task Scheduling for Accelerated OpenMP

Runtime Adaptation for Autonomic Heterogeneous Computing

Contact Info

Product

Resources

About