Directive-based Programming for GPUs: A Comparative Study
Authors:
- Ruym'N ReyesRRRuym'N Reyes
- Ivan LopezILIvan Lopez
- Juan J. FumeroJFJuan J. Fumero
- Francisco de SandeFSFrancisco de Sande
- 1 June 2012
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 410-417
Abstract
GPUs and other accelerators are available on many different devices, while GPGPU has been massively adopted by the HPC research community. Although a plethora of libraries and applications providing GPU support are available, the need of implementing new algorithms from scratch, or adapting sequential programs to accelerators, will always exist. Writing CUDA or OpenCL codes, although an easier task than using their predecessors, is not trivial. Obtaining performance is even harder, as it requires deep understanding of the underlying architecture. Some efforts have been directed toward the automatic code generation for GPU devices, with different results. In particular, several directive-oriented programming models, taking advantage of the OpenMP success, have been created. Although future OpenMP releases will integrate accelerators into the standard, tools are needed in the meantime. In this work, we present a comparison between three directive-based programming models: hiCUDA, PGI Accelerator and OpenACC, using for the last our novel accULL implementation. With this comparison, we aim to showcase the evolution of the directive-based programming models and how users can guide tools toward better performance results.Keywords
This publication has 10 references indexed in Scilit:
- Optimization strategies in different CUDA architectures using llCoMPMicroprocessors and Microsystems, 2012
- OmpSs: A PROPOSAL FOR PROGRAMMING HETEROGENEOUS MULTI-CORE ARCHITECTURESParallel Processing Letters, 2011
- Automatic code generation for GPUs in llcThe Journal of Supercomputing, 2011
- A characterization of the Rodinia benchmark suite with comparison to contemporary CMP workloads2010
- hiCUDA: High-Level GPGPU ProgrammingIEEE Transactions on Parallel and Distributed Systems, 2010
- Implementing the PGI Accelerator model2010
- An asymmetric distributed shared memory model for heterogeneous parallel systemsACM SIGARCH Computer Architecture News, 2010
- Unrolling Loops Containing Task ParallelismLecture Notes in Computer Science, 2010
- Cetus: A Source-to-Source Compiler Infrastructure for MulticoresComputer, 2009
- hi CUDA2009