The DOI matches its Crossref record; 1 of the 1 references it cites were found in a registry or an index.
Linpack evaluation on a supercomputer with heterogeneous accelerators
Abstract
We report Linpack benchmark results on the TSUBAME supercomputer, a large scale heterogeneous system equipped with NVIDIA Tesla GPUs and ClearSpeed SIMD accelerators. With all of 10,480 Opteron cores, 640 Xeon cores, 648 ClearSpeed accelerators and 624 NVIDIA Tesla GPUs, we have achieved 87.01TFlops, which is the third record as a heterogeneous system in the world. This paper describes careful tuning and load balancing method required to achieve this performance. On the other hand, since the peak speed is 163 TFlops, the efficiency is 53%, which is lower than other systems. This paper also analyses this gap from the aspect of system architecture.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
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?
Aggressively using CPU for computation affects its ability to act as a host, which harms performance.
“Also, aggressively using CPU for computation affects its ability to act as a host, which harms the performance [Endo et al. 2010;Gregg et al. 2010;Sun et al. 2012].”
From A Survey of CPU-GPU Heterogeneous Computing Techniques (2015) · cites this paper as Endo et al. 2010
Citations
Cites
Lean Thinking—Banish Waste and Create Wealth in your Corporation
Womack, J. P., & Jones, D. T. (1997) · Journal of the Operational Research Society · doi:10.1057/palgrave.jors.2600967 · Cited by 2,122
Flagged first, then by citations
Research works printed but not linked: 11
- 10.1145/1542275.1542313 · DOI
- 10.1109/SC.2005.42 · DOI
- kistler, 2009, Petascale computing with accelerators, Proceedings of ACM Symposium on Principles and Practice of Paralle Computing (PPoPP09), 241, 10.1145/1504176.1504212 · DOI
- 10.1145/1513895.1513901 · DOI
- endo, 2008, Massive supercomputing coping with heterogeneity of modern accelerators, Proceedings of IEEE International Parallel and Distributed Processing Symposium (IPDPS08), 10.1109/IPDPS.2008.4536251 · DOI
- nukada, 2008, Band-width intensive 3-D FFT kernel for GPUs using CUDA, Proc IEEE/ACM Supercomputing (SC 08)
- 10.1109/SC.2008.5214716 · DOI
- 10.1016/j.parco.2007.09.002 · DOI
- edelkamp, 2009, Parallel state space search on the GPU, Proceedings of Symposium on Combinatorial Search (SoCS 2009)
- 10.1109/SC.2008.5214713 · DOI
- 10.1109/IPDPS.2009.5161031 · DOI
Cited by
Sorted by citations
A Survey of CPU-GPU Heterogeneous Computing Techniques
Mittal, S., & Vetter, J. S. (2015) · ACM Computing Surveys · doi:10.1145/2788396 · Cited by 568
Metadata
- Authors
- Endo, T., Matsuoka, S., Nukada, A., & Maruyama, N.
- Venue
- 2010 IEEE International Symposium on Parallel & Distributed Processing (IPDPS)
- Publisher
- IEEE
- Published
- 2010
- Cited by
- 43
- References
- 1
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
