CoChat
Sign Up
Cited by 640Parallel Computing and Optimization TechniquesComputer Graphics and Visualization TechniquesDistributed and Parallel Computing Systems

The DOI matches its Crossref record; 1 of the 1 references it cites were found in a registry or an index.

Debunking the 100X GPU vs. CPU myth: an evaluation of throughput computing on CPU and GPU

Lee, Victor W., Kim, Changkyu, Chhugani, Jatin, Deisher, Michael, Kim, Daehyun, Nguyen, Anthony D., Satish, Nadathur, Smelyanskiy, Mikhail, Chennupaty, Srinivas, Hammarlund, Per, Singhal, Ronak, Dubey, Pradeep·2010·Proceedings of the 37th annual international symposium on Computer architecture
DOI 10.1145/1815961.1816021
Open this paper in CoChat — free account

Abstract

Recent advances in computing have led to an explosion in the amount of data being generated. Processing the ever-growing data in a timely manner has made throughput computing an important aspect for emerging applications. Our analysis of a set of important throughput computing kernels shows that there is an ample amount of parallelism in these kernels which makes them suitable for today's multi-core CPUs and GPUs. In the past few years there have been many studies claiming GPUs deliver substantial speedups (between 10X and 1000X) over multi-core CPUs on these kernels. To understand where such large performance difference comes from, we perform a rigorous performance analysis and find that after applying optimizations appropriate for both CPUs and GPUs the performance gap between an Nvidia GTX280 processor and the Intel Core i7-960 processor narrows to only 2.5x on average. In this paper, we discuss optimization techniques for both CPU and GPU, analyze what architecture features contributed to performance differences between the two architectures, and recommend a set of architectural features which provide significant improvement in architectural efficiency for throughput kernels.

Overview

Integrity signals

Registry

Crossref · 10.1145/1815961.1816021

Cites

1 of 1 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Verification

Claims the paper makes

This paper’s full text isn’t openly available, so its claims and self-check weren’t run. The registry and citation checks still apply.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Several initial works report GPUs provide up to 100× to 1000× speedup; other researchers claim that on applying careful optimizations on both CPUs and GPUs, CPUs may equal or even outperform GPUs.

    “Further, while several initial works report that GPUs provide up to 100× to 1000× speedup, other researchers claim that on applying careful optimizations on both CPUs and GPUs; CPUs may equal or even outperform the performance of GPUs [Gummaraju et al. 2010;Lee et al. 2010].”

    From A Survey of CPU-GPU Heterogeneous Computing Techniques (2015) · cites this paper as Lee et al. 2010

  • The limited bandwidth of PCIe has conventionally remained a bottleneck in GPU performance, especially for applications which transfer large amount of data between CPU and GPU.

    “The limited bandwidth of PCIe has conventionally remained a bottleneck in GPU performance, especially for applications which transfer large amount of data between CPU and GPU [Gregg and Hazelwood 2011;Lee et al. 2010].”

    From A Survey of CPU-GPU Heterogeneous Computing Techniques (2015) · cites this paper as Lee et al. 2010

Citations

Cites

Flagged first, then by citations

Research works printed but not linked: 40

  1. General-purpose computation on graphics hardware. http://gpgpu.org/ , 2009 . General-purpose computation on graphics hardware. http://gpgpu.org/, 2009.
  2. 10.1145/1555754.1555810 · DOI
  3. D. H. Bailey . A high-performance fft algorithm for vector supercomputers-abstract . In Proceedings of the Third SIAM Conference on Parallel Processing for Scientific Computing , page 114 , Philadelphia, PA, USA , 1989 . Society for Industrial and Applied Mathematics. D. H. Bailey. A high-performance fft algorithm for vector supercomputers-abstract. In Proceedings of the Third SIAM Conference on Parallel Processing for Scientific Computing, page 114, Philadelphia, PA, USA, 1989. Society for Industrial and Applied Mathematics.
  4. 10.1145/1654059.1654078 · DOI
  5. 10.1145/1454115.1454128 · DOI · Paper page
  6. 10.1145/1555815.1555777 · DOI
  7. 10.1109/MICRO.2006.18 · DOI
  8. 10.1109/JPROC.2008.917729 · DOI
  9. Y.-K. Chen , J. Chhugani , C. J. Hughes , D. Kim , S. Kumar , V. W. Lee , A. Lin , A. D. Nguyen , E. Sifakis , and M. Smelyanskiy . High-performance physical simulations on next-generation architecture with many cores . Intel Technology Journal , 11 , 2007 . Y.-K. Chen, J. Chhugani, C. J. Hughes, D. Kim, S. Kumar, V. W. Lee, A. Lin, A. D. Nguyen, E. Sifakis, and M. Smelyanskiy. High-performance physical simulations on next-generation architecture with many cores. Intel Technology Journal, 11, 2007.
  10. 10.14778/1454159.1454171 · DOI
  11. 10.5555/1413370.1413375 · DOI
  12. F. Franchetti , M. Püschel , Y. Voronenko , S. Chellappa , and J. M. F. Moura . Discrete Fourier transform on multicore . IEEE Signal Processing Magazine, special issue on "Signal Processing on Platforms with Multiple Cores ", 26(6):90--102, 2009 . F. Franchetti, M. Püschel, Y. Voronenko, S. Chellappa, and J. M. F. Moura. Discrete Fourier transform on multicore. IEEE Signal Processing Magazine, special issue on "Signal Processing on Platforms with Multiple Cores", 26(6):90--102, 2009.
  13. 10.1109/JPROC.2004.840301 · DOI · Paper page
  14. L. Genovese . Graphic processing units: A possible answer to HPC . In 4th ABINIT Developer Workshop , 2009 . L. Genovese. Graphic processing units: A possible answer to HPC. In 4th ABINIT Developer Workshop, 2009.
  15. 10.1145/1142473.1142511 · DOI
  16. 10.5555/1413370.1413373 · DOI
  17. 10.1145/1555815.1555775 · DOI
  18. Intel. SSE4 Programming Reference . 2007 . Intel. SSE4 Programming Reference. 2007.
  19. 10.1109/PACT.2005.10 · DOI
  20. 10.1007/BF01189337 · DOI
  21. 10.1145/1807167.1807206 · DOI
  22. 10.1145/1250662.1250683 · DOI
  23. 10.1109/ISCA.2008.38 · DOI
  24. N. Leischner , V. Osipov , and P. Sanders . Fermi Architecture White Paper , 2009 . N. Leischner, V. Osipov, and P. Sanders. Fermi Architecture White Paper, 2009.
  25. 10.1111/j.1467-8659.2007.01012.x · DOI
  26. 10.1109/JPROC.2004.840306 · DOI · Paper page
  27. 10.1145/1555815.1555793 · DOI
  28. 10.1145/1179622.1179679 · DOI
  29. 10.1109/IPDPS.2009.5161005 · DOI
  30. 10.1145/1807167.1807207 · DOI
  31. 10.1145/1360612.1360617 · DOI
  32. 10.1145/1375527.1375572 · DOI
  33. 10.1109/TVCG.2009.164 · DOI
  34. 10.1080/10618560802238275 · DOI
  35. N. Univ . of Illinois. Technical reference: Base operating system and extensions , volume 2 , 2009 . N. Univ. of Illinois. Technical reference: Base operating system and extensions, volume 2, 2009.
  36. F. Vazquez , E. M. Garzon , J.A. Martinez , and J.J. Fernandez . The sparse matrix vector product on GPUs. Technical report , University of Almeria , June 2009 . F. Vazquez, E. M. Garzon, J.A.Martinez, and J.J.Fernandez. The sparse matrix vector product on GPUs. Technical report, University of Almeria, June 2009.
  37. 10.5555/1413370.1413402 · DOI
  38. 10.1145/1362622.1362674 · DOI · Paper page
  39. W. Xu and K. Mueller . A performance-driven study of regularization methods for gpu-accelerated iterative ct . In Workshop on High Performance Image Reconstruction (HPIR) , 2009 . W. Xu and K. Mueller. A performance-driven study of regularization methods for gpu-accelerated iterative ct. In Workshop on High Performance Image Reconstruction (HPIR), 2009.
  40. 10.1109/CSSE.2008.1448 · DOI

Cited by

Sorted by citations

Metadata

Authors
Lee, V. W., Kim, C., Chhugani, J., Deisher, M., Kim, D., Nguyen, A. D., Satish, N., Smelyanskiy, M., Chennupaty, S., Hammarlund, P., Singhal, R., & Dubey, P.
Affiliations
Intel (United States) (US)
ORCID
Venue
Proceedings of the 37th annual international symposium on Computer architecture
Publisher
ACM
Published
2010-06-19
Pages
451-460
ISSN
1063-6897 (ISSN-L)
DOI
10.1145/1815961.1816021
OpenAlex
W2159481344
Cited by
640
References
1
Citations by year
2026: 6 · 2025: 12 · 2024: 7 · 2023: 7 · 2022: 5 · 2021: 13 · 2020: 21 · 2019: 31 · 2018: 31 · 2017: 44 · 2016: 43 · 2015: 55 · 2014: 84 · 2013: 81 · 2012: 98
FWCI
50.15
Type
conference paper
Language
en
Open access
closed
Licence
www.acm.org/publications/policies/copyright_policy#Background (vor)
Locations
Proceedings - International Symposium on Computer Architecture/Proceedings (publishedVersion)
Topics
  • Physical Sciences › Computer Science › Hardware and Architecture › Parallel Computing and Optimization Techniques
  • Physical Sciences › Computer Science › Computer Graphics and Computer-Aided Design › Computer Graphics and Visualization Techniques
  • Physical Sciences › Computer Science › Computer Networks and Communications › Distributed and Parallel Computing Systems
Keywords
parallelism, multi-core processors, GPU acceleration, performance optimization, performance evaluation

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review