tutorial

GPGPU Composition with OCaml

Authors:

Mathias Bourgoin,

Emmanuel ChaillouxAuthors Info & Claims

ARRAY'14: Proceedings of ACM SIGPLAN International Workshop on Libraries, Languages, and Compilers for Array Programming

Pages 32 - 37

https://doi.org/10.1145/2627373.2627379

Published: 09 June 2014 Publication History

Abstract

GPGPU programming promises high performance. However, to achieve it, developers must overcome several challenges. The main ones are: write and use hyper-parallel kernels on GPU, manage memory transfers between CPU and GPU, and compose kernels, keeping individual performance of components while optimizing the global performance. In this article, we propose to study the composition by distinguishing the location where it is done: kernel composition on the GPU, kernel generation by the CPU, and overall composition. To achieve it, we use the SPOC library, developed in OCaml. SPOC offers abstractions over the Cuda and OpenCL frameworks. It proposes a specific language, called Sarek, to express kernels and different parallel skeletons to compose them. We show that by increasing the level of abstraction to handle kernels, programs are easier to write and that some optimizations (via kernel generation and transfers scheduling) become possible. Thus, we win on both sides: expressiveness and efficiency.

References

[1]

Nvidia, Cuda C Programming guide, 2012.

[2]

K. O. W. Group, "OpenCL 1.2 specifications," 2012.

[3]

X. Leroy, D. Doligez, A. Firsch, J. Garrigue, D. R. Remy, and J. Vouillon, "The OCaml System Release 4.01: Documentation and User's Manual," tech. rep., Inria, 2013. http://caml.inria.fr.

[4]

M. Bourgoin, E. Chailloux, and J.-L. Lamotte, "Spoc: GPGPU Programming Through Stream Processing with OCAML," Parallel Processing Letters, vol. 22, pp. 1--12, May 2012. http://www.algo-prog.info/spoc.

[5]

M. Bourgoin, E. Chailloux, and J.-L. Lamotte, "Efficient Abstractions for GPGPU Programming," International Journal of Parallel Programming, pp. 1--18, 2013.

Digital Library

[6]

NVIDIA, "Cublas Library," 2012. http://developer.nvidia.com/cublas.

[7]

S. Tomov, R. Nath, P. Du, and J. Dongarra, "MAGMA Users Guide," 2011.

[8]

J. Donham and N. Pouillard, "Camlp4 and Template Haskell," in ACM SIGPLAN Commercial Users of Functional Programming, pp. 6:1--6:1, ACM, 2010.

Digital Library

[9]

T. B. Jablin, P. Prabhu, J. A. Jablin, N. P. Johnson, S. R. Beard, and D. I. August, "Automatic CPU-GPU communication management and optimization," ACM SIGPLAN Notices, pp. 142--151, 2011.

Digital Library

[10]

N. S. Scott, M. P. Scott, P. G. Burke, T. Stitt, V. Faro-Maza, C. Denis, and A. Maniopoulou, "2DRMP: A Suite of Two-Dimensional R-matrix Propagation Codes.," Computer Physics Communications, pp. 2424--2449, 2009.

[11]

M. Bourgoin, E. Chailloux, and J.-L. Lamotte, "Retour d'expérience: portage d'une application haute-performance vers un langage de haut niveau," in la Conférence d'informatique en Parallélisme, Architecture et Système (ComPas), 2013.

[12]

P. Kipfer and R. Westermann, "Improved gpu sorting," in GPU gems, vol. 2, pp. 733--746, 2005.

[13]

M. Cole, "Bringing Skeletons out of the Closet: A Pragmatic Manifesto for Skeletal Parallel Programming," Parallel computing, pp. 389--406, 2004.

Digital Library

[14]

R. Di Cosmo, Z. Li, M. Danelutto, S. Pelagatti, X. Leroy, P. Weis, and F. Clément, CamlP3l 1.0: User Manual, 2010. http://camlp3l.inria.fr/.

[15]

M. Chakravarty, G. Keller, S. Lee, T. McDonell, and V. Grover, "Accelerating Haskell array codes with multicore GPUs," in Proceedings of the sixth workshop on Declarative aspects of multicore programming (DAMP), pp. 3--14, ACM, 2011.

Digital Library

[16]

R. Clifton-Everest, T. McDonell, M. Chakravarty, and G. Keller, "Embedding foreign code," in Practical Aspects of Declarative Languages (M. Flatt and H.-F. Guo, eds.), vol. 8324 of Lecture Notes in Computer Science, pp. 136--151, Springer International Publishing, 2014.

[17]

G. Cocco, "FSCL F# to OpenCL Compiler and Runtime," 2013. http://www.gabrielecocco.it/fscl/.

[18]

A. Rubinsteyn, E. Hielscher, N. Weinman, and D. Shasha, "Parakeet: A just-in-time parallel accelerator for python," in The 4th USENIX Workshop on Hot Topics in Parallelism, USENIX, 2012.

Digital Library

[19]

B. Catanzaro, M. Garland, and K. Keutzer, "Copperhead: compiling an embedded data parallel language," in ACM SIGPLAN Notices, vol. 46, pp. 47--56, ACM, 2011.

Digital Library

[20]

T. Oliphant, "Numba python bytecode to LLVM translator," in Proceedings of the Python for Scientific Computing Conference (SciPy), 2012.

[21]

A. INC, "Aparapi," 2013. http://code.google.com/p/aparapi/.

[22]

E. Holk, M. Pathirage, A. Chauhan, A. Lumsdaine, and N. D. Matsakis, "GPU Programming in Rust: Implementing High-Level Abstractions in a Systems-Level Language," in Proceedings of the 2013 IEEE 27th International Symposium on Parallel and Distributed Processing Workshops and PhD Forum (IPDPSW), pp. 315--324, 2013.

Digital Library

[23]

E. Holk, W. E. Byrd, N. Mahajan, J. Willcock, A. Chauhan, and A. Lumsdaine, "Declarative parallel programming for gpus.," in Internation Conference on Parallel Computing (PARCO), pp. 297--304, 2011.

[24]

J. Hoberock and N. Bell, "Thrust: C++ Template Library for CUDA," 2009. http://code.google.com/p/thrust/.

[25]

C. Augonnet, S. Thibault, R. Namyst, and P.-A. Wacrenier, "StarPU: A Unified Platform for Task Scheduling on Heterogeneous Multicore Architectures," Concurrency and Computation: Practice and Experience, Special Issue: Euro-Par 2009, vol. 23, pp. 187--198, 2009.

Digital Library

[26]

T. Gautier, J. V. Lima, N. Maillard, and B. Raffin, "Xkaapi: A runtime system for data-flow task programming on heterogeneous architectures," in Proceedings of the 2013 IEEE 27th International Symposium on Parallel & Distributed Processing (IPDPS), pp. 1299--1308, 2013.

Digital Library

[27]

M. Amini, B. Creusillet, S. Even, R. Keryell, O. Goubier, S. Guelton, J. O. McMahon, F.-X. Pasquier, G. Péan, P. Villalon, et al., "Par4all: From convex array regions to heterogeneous computing," in IMPACT 2012: Second International Workshop on Polyhedral Compilation Techniques HiPEAC 2012, 2012.

[28]

T. Aarnio and M. Bourges-Sevenier, "WebCL 1.0 specification," Khronos WebCL Working Group, 2014.

[29]

J. Vouillon and V. Balat, "From Bytecode to JavaScript: The js of ocaml Compiler," Software: Practice and Experience, 2013.

Cited By

Ashcraft MLemon APenry DSnell Q(2019)Compiler Optimization of Accelerator Data TransfersInternational Journal of Parallel Programming10.1007/s10766-017-0549-347:1(39-58)Online publication date: 1-Feb-2019
https://dl.acm.org/doi/10.1007/s10766-017-0549-3
Bourgoin MChailloux ELamotte J(2017)High Level Data Structures for GPGPU Programming in a Statically Typed LanguageInternational Journal of Parallel Programming10.1007/s10766-016-0424-745:2(242-261)Online publication date: 1-Apr-2017
https://dl.acm.org/doi/10.1007/s10766-016-0424-7
Bourgoin MChailloux EHendren LMasuhara HSheeran MVitek J(2015)High-level accelerated array programming in the web browserProceedings of the 2nd ACM SIGPLAN International Workshop on Libraries, Languages, and Compilers for Array Programming10.1145/2774959.2774964(31-36)Online publication date: 13-Jun-2015
https://dl.acm.org/doi/10.1145/2774959.2774964

Index Terms

GPGPU Composition with OCaml

Recommendations

Efficient Abstractions for GPGPU Programming

General purpose (GP)GPU programming demands to couple highly parallel computing units with classic CPUs to obtain a high performance. Heterogenous systems lead to complex designs combining multiple paradigms and programming languages to manage each ...
High Level Data Structures for GPGPU Programming in a Statically Typed Language

To increase software performance, it is now common to use hardware accelerators. Currently, GPUs are the most widespread accelerators that can handle general computations. This requires to use GPGPU frameworks such as Cuda or OpenCL. Both are very low-...
High-level accelerated array programming in the web browser
ARRAY 2015: Proceedings of the 2nd ACM SIGPLAN International Workshop on Libraries, Languages, and Compilers for Array Programming

Client-side web programming currently means using technologies embedded in web browsers to run computations on the client computer. Most solutions imply using JavaScript which allows to describe computations, and modifications of the DOM displayed by ...

Comments

Information & Contributors

Information

Published In

cover image ACM Conferences

ARRAY'14: Proceedings of ACM SIGPLAN International Workshop on Libraries, Languages, and Compilers for Array Programming

June 2014

112 pages

ISBN:9781450329378

DOI:10.1145/2627373

Copyright © 2014 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].

Sponsors

SIGPLAN: ACM Special Interest Group on Programming Languages

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 09 June 2014

Permissions

Request permissions for this article.

Request Permissions

Check for updates

Author Tags

Qualifiers

Tutorial
Research
Refereed limited

Conference

PLDI '14

Sponsor:

SIGPLAN

PLDI '14: ACM SIGPLAN Conference on Programming Language Design and Implementation

June 9 - 11, 2014

Edinburgh, United Kingdom

Acceptance Rates

ARRAY'14 Paper Acceptance Rate 17 of 25 submissions, 68%;

Overall Acceptance Rate 17 of 25 submissions, 68%

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

3
Total Citations
View Citations
180
Total Downloads

Downloads (Last 12 months)4
Downloads (Last 6 weeks)0

Reflects downloads up to 23 Feb 2025

Other Metrics

View Author Metrics

Citations

Cited By

Ashcraft MLemon APenry DSnell Q(2019)Compiler Optimization of Accelerator Data TransfersInternational Journal of Parallel Programming10.1007/s10766-017-0549-347:1(39-58)Online publication date: 1-Feb-2019
https://dl.acm.org/doi/10.1007/s10766-017-0549-3
Bourgoin MChailloux ELamotte J(2017)High Level Data Structures for GPGPU Programming in a Statically Typed LanguageInternational Journal of Parallel Programming10.1007/s10766-016-0424-745:2(242-261)Online publication date: 1-Apr-2017
https://dl.acm.org/doi/10.1007/s10766-016-0424-7
Bourgoin MChailloux EHendren LMasuhara HSheeran MVitek J(2015)High-level accelerated array programming in the web browserProceedings of the 2nd ACM SIGPLAN International Workshop on Libraries, Languages, and Compilers for Array Programming10.1145/2774959.2774964(31-36)Online publication date: 13-Jun-2015
https://dl.acm.org/doi/10.1145/2774959.2774964

View Options

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Full Access

Get this Publication

View options

PDF

View or Download as a PDF file.

eReader

View online with eReader.

Figures

Tables

Media

View Table of Conten