Memory-access-aware Safety and Profitability Analysis for Transformation of Accelerator-bound OpenMP Loops
Authors: 
Artem Chikin, Taylor Lloyd, 
José Nelson Amaral, Ettore Tiotto, Muhammad UsmanAuthors Info & Claims
Artem Chikin, Taylor Lloyd, José Nelson Amaral, Ettore Tiotto, Muhammad UsmanAuthors Info & ClaimsArticle No.: 30, Pages 1 - 26
Abstract
Iteration Point Difference Analysis is a new static analysis framework that can be used to determine the memory coalescing characteristics of parallel loops that target GPU offloading and to ascertain safety and profitability of loop transformations with the goal of improving their memory access characteristics. This analysis can propagate definitions through control flow, works for non-affine expressions, and is capable of analyzing expressions that reference conditionally defined values. This analysis framework enables safe and profitable loop transformations. Experimental results demonstrate potential for dramatic performance improvements. GPU kernel execution time across the Polybench suite is improved by up to 25.5× on an Nvidia P100 with benchmark overall improvement of up to 3.2×. An opportunity detected in a SPEC ACCEL benchmark yields kernel speedup of 86.5× with a benchmark improvement of 3.3×. This work also demonstrates how architecture-aware compilers improve code portability and reduce programmer effort.
References
[1]
ACF-Coalescing-LLVM 2019. ACF static analysis framework source-code. Retrieved from: https://github.com/uasys/ACF-Coalescing-LLVM.
[2]
A. Bakhoda, G. L. Yuan, W. W. L. Fung, H. Wong, and T. M. Aamodt. 2009. Analyzing CUDA workloads using a detailed GPU simulator. In Proceedings of the International Symposium on Performance Analysis of Systems and Software (ISPASS’09). 163--174.
[3]
Utpal K. Banerjee. 1988. Dependence Analysis for Supercomputing. Kluwer Academic Publishers, Norwell, MA.
[4]
W. Blume and R. Eigenmann. 1994. The range test: A dependence test for symbolic, non-linear expressions. In Proceedings of the Conference on Supercomputing. 528--537.
[5]
Uday Bondhugula, Albert Hartono, J. Ramanujam, and P. Sadayappan. 2008. A practical automatic polyhedral parallelizer and locality optimizer. In Proceedings of the ACM-SIGPLAN Symposium on Programming Language Design and Implementation (PLDI’08). 101--113.
[6]
T. E. Cheatham and J. A. Townley. 1976. Symbolic evaluation of programs: A look at loop analysis. In Proceedings of the Symposium on Symbolic and Algebraic Computation.
[7]
Patrick Cousot and Radhia Cousot. 1977. Abstract interpretation: A unified lattice model for static analysis of programs by construction or approximation of fixpoints. In Proceedings of the 4th ACM SIGACT-SIGPLAN Symposium on Principles of Programming Languages (POPL’77). ACM, 238--252.
[8]
Thomas Fahringer, Michael Gerndt, Graham Riley, and Jesper Larsson Träff. 2000. Formalizing OpenMP performance properties with ASL. In Proceedings of the International Symposium on High Performance Computing, Mateo Valero, Kazuki Joe, Masaru Kitsuregawa, and Hidehiko Tanaka (Eds.). 428--439.
[9]
Michael P. Gerlek, Eric Stoltz, and Michael Wolfe. 1995. Beyond induction variables: Detecting and classifying sequences using a demand-driven SSA form. ACM Trans. Prog. Lang. Syst. 17, 1 (Jan. 1995), 85--122.
[10]
Gina Goff, Ken Kennedy, and Chau-Wen Tseng. 1991. Practical dependence testing. In Proceedings of the ACM-SIGPLAN Symposium on Programming Language Design and Implementation (PLDI’91). 15--29.
[11]
M. Haghighat and C. Polychronopoulos. 1993. Symbolic program analysis and optimization for parallelizing compilers. In Proceedings of the Workshop on Languages and Compilers and Parallel Computing (LCPC’93). 538--562.
[12]
Intel. 2011. Avoiding and identifying false sharing among threads. Retrieved on March 13, 2018 from: https://software.intel.com/en-us/articles/avoiding-and-identifying-false-sharing-among-threads.
[13]
Guido Juckeland, William Brantley, Sunita Chandrasekaran, Barbara Chapman, Shuai Che, Mathew Colgrove, Huiyu Feng, Alexander Grund, Robert Henschel, Wen-Mei W. Hwu, Huian Li, Matthias S. Müller, Wolfgang E. Nagel, Maxim Perminov, Pavel Shelepugin, Kevin Skadron, John Stratton, Alexey Titov, Ke Wang, Matthijs van Waveren, Brian Whitney, Sandra Wienke, Rengan Xu, and Kalyan Kumaran. 2015. SPEC ACCEL: A standard application suite for measuring hardware accelerator performance. In High Performance Computing Systems. Performance Modeling, Benchmarking, and Simulation (LNCS), Vol. 8966. Stephen A. Jarvis, Steven A. Wright, and Simon D. Hammond (Eds.). 46--67.
[14]
Leslie Lamport. 1974. The parallel execution of DO loops. Commun. ACM 17, 2 (Feb. 1974), 83--93.
[15]
Seyong Lee, Seung-Jai Min, and Rudolf Eigenmann. 2009. OpenMP to GPGPU: A compiler framework for automatic translation and optimization. In Proceedings of the Symposium on Principles 8 Practice of Parallel Programming (PPoPP’09). 101--110.
[16]
T. Lloyd, K. Ali, and J. N. Amaral. 2019. GPUCheck: Detecting CUDA Performance Problems with Static Analysis. Technical Report. University of Alberta, Edmonton, AB, Canada.
[17]
T. Lloyd, A. Chikin, J. N. Amaral, and E. Tiotto. 2018. Automated GPU grid geometry selection for OpenMP kernels. In Proceedings of the Workshop on Applications for Multi-Core Architectures (WAMCA’18). Retrieved from: https://webdocs.cs.ualberta.ca/amaral/papers/LloydWAMCA18.pdf.
[18]
Scott A. Mahlke, David C. Lin, William Y. Chen, Richard E. Hank, and Roger A. Bringmann. 1992. Effective compiler support for predicated execution using the hyperblock. In Proceedings of the International Symposium on Microarchitecture (MICRO’92). 45--54.
[19]
Jiayuan Meng and Kevin Skadron. 2009. Performance modeling and automatic ghost zone optimization for iterative stencil loops on GPUs. In Proceedings of the International Conference on Supercomputing (ICS’09). 256--265.
[20]
D. Miles, D. Norton, and M. Wolfe. 2014. Performance portability and OpenACC. In Proceedings of the Conference of Cray User Group (CUG’14).
[21]
Sungdo Moon, Mary W. Hall, and Brian R. Murphy. 1998. Predicated array data-flow analysis for run-time parallelization. In Proceedings of the 12th International Conference on Supercomputing (ICS’98). 204--211.
[22]
Nvidia. {n.d.}. Nvidia Tesla V100 GPU architecture—The world’s most advanced data center GPU. Retrieved from: http://images.nvidia.com/content/volta-architecture/pdf/volta-architecture-whitepaper.pdf.
[23]
Cosmin E. Oancea and Lawrence Rauchwerger. 2012. Logical inference techniques for loop parallelization. In Proceedings of the ACM-SIGPLAN Symposium on Programming Language Design and Implementation (PLDI’12). 509--520.
[24]
Cosmin E. Oancea and Lawrence Rauchwerger. 2015. Scalable conditional induction variables (CIV) analysis. In Proceedings of the International Symposium on Code Generation and Optimization (CGO’15). 213--224.
[25]
Michael F. P. O’Boyle, Zheng Wang, and Dominik Grewe. 2013. Portable mapping of data parallel programs to OpenCL for heterogeneous systems. In Proceedings of the International Symposium on Code Generation and Optimization (CGO’13). 1--10.
[26]
OpenMP Language Committee. 2013. OpenMP application program interface version 4.0. Retrieved on March 13, 2018 from: http://www.openmp.org/wp-content/uploads/OpenMP4.0.0.pdf.
[27]
Yunheung Paek, Jay Hoeflinger, and David Padua. 2002. Efficient and precise array access analysis. ACM Trans. Prog. Lang. Syst. 24, 1 (Jan. 2002), 65--109.
[28]
Daniel Rolls, Carl Joslin, and Sven-Bodo Scholz. 2010. Unibench: A tool for automated and collaborative benchmarking. In Proceedings of the International Conference on Program Comprehension (ICPC’10). 50--51.
[29]
Silvius Rus, Dongmin Zhang, and Lawrence Rauchwerger. 2004. The value evolution graph and its use in memory reference analysis. In Proceedings of the International Conference on Parallel Architectures and Compilation Techniques (PACT’04). 243--254.
[30]
Arthur Stoutchinin and Francois de Ferriere. 2001. Efficient static single assignment form for predication. In Proceedings of the International Symposium on Microarchitecture (MICRO’01). 172--181.
[31]
Robert A. van Engelen, J. Birch, Y. Shou, B. Walsh, and Kyle A. Gallivan. 2004. A unified framework for nonlinear dependence testing and symbolic analysis. In Proceedings of the International Conference on Supercomputing (ICS’04). 106--115.
[32]
Zheng Wang and Michael F. P. O’Boyle. 2009. Mapping parallelism to multi-cores: A machine learning based approach. In Proceedings of the ACM SIGPLAN Symposium on Principles 8 Practice of Parallel Programming (PPoPP’09).
[33]
Michael Joseph Wolfe. 1982. Optimizing Supercompilers for Supercomputers. Ph.D. Dissertation. University of Illinois at Urbana-Champaign, Champaign, IL.
[34]
Zhong Zheng, Xuhao Chen, Zhiying Wang, Li Shen, and Jiawen Li. 2011. Performance model for OpenMP parallelized loops. In Proceedings of the Conference on Transportation, Mechanical, and Electrical Engineering (TMEE’11).
Index Terms
- Memory-access-aware Safety and Profitability Analysis for Transformation of Accelerator-bound OpenMP Loops
Recommendations
OpenMP to GPGPU: a compiler framework for automatic translation and optimization
PPoPP '09: Proceedings of the 14th ACM SIGPLAN symposium on Principles and practice of parallel programmingGPGPUs have recently emerged as powerful vehicles for general-purpose high-performance computing. Although a new Compute Unified Device Architecture (CUDA) programming model from NVIDIA offers improved programmability for general computing, programming ...
Improving Memory Hierarchy Performance through Combined Loop Interchange and Multi-Level Fusion
Because of the increasing gap between the speeds of processors and main memories, compilers must enhance the locality of applications to achieve high performance. Loop fusion enhances locality by fusing loops that access similar sets of data. Typically, ...
Assessing One-to-One Parallelism Levels Mapping for OpenMP Offloading to GPUs
PMAM'17: Proceedings of the 8th International Workshop on Programming Models and Applications for Multicores and ManycoresThe proliferation of accelerators in modern clusters makes efficient coprocessor programming a key requirement if application codes are to achieve high levels of performance with acceptable energy consumption on such platforms. This has led to ...
Comments
Information & Contributors
Information
Published In
September 2019
347 pages
Copyright © 2019 ACM.
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected].
Publisher
Association for Computing Machinery
New York, NY, United States
Publication History
Published: 18 July 2019
Accepted: 01 May 2019
Revised: 01 April 2019
Received: 01 August 2018
Published in TACO Volume 16, Issue 3
Check for updates
Author Tags
Qualifiers
- Research-article
- Research
- Refereed
Funding Sources
- Natural Sciences and Engineering Research Council of Canada
- IBM Centre for Advanced Studies
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 707Total Downloads
- Downloads (Last 12 months)128
- Downloads (Last 6 weeks)18
Reflects downloads up to 10 Nov 2024
Other Metrics
Citations
View Options
View options
View or Download as a PDF file.
PDFeReader
View online with eReader.
eReaderHTML Format
View this article in HTML Format.
HTML FormatGet Access
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in