Display options
Cite
Teodoro G, Kurc T, Andrade G, et al. Application Performance Analysis and Efficient Execution on Systems with multi-core CPUs, GPUs and MICs: A Case Study with Microscopy Image Analysis. Int J High Perform Comput Appl. 2015;31(1):32-51doi: 10.1177/1094342015594519.
Teodoro, G., Kurc, T., Andrade, G., Kong, J., Ferreira, R., & Saltz, J. (2017). Application Performance Analysis and Efficient Execution on Systems with multi-core CPUs, GPUs and MICs: A Case Study with Microscopy Image Analysis. The international journal of high performance computing applications, 31(1), 32-51. https://doi.org/10.1177/1094342015594519
Teodoro, George, et al. "Application Performance Analysis and Efficient Execution on Systems with multi-core CPUs, GPUs and MICs: A Case Study with Microscopy Image Analysis." The international journal of high performance computing applications vol. 31,1 (2017): 32-51. doi: https://doi.org/10.1177/1094342015594519
Teodoro G, Kurc T, Andrade G, Kong J, Ferreira R, Saltz J. Application Performance Analysis and Efficient Execution on Systems with multi-core CPUs, GPUs and MICs: A Case Study with Microscopy Image Analysis. Int J High Perform Comput Appl. 2017 Jan;31(1):32-51. doi: 10.1177/1094342015594519. Epub 2015 Jul 27. PMID: 28239253; PMCID: PMC5319667.
Copy Download .nbib Format: NLM
Share it on

Int J High Perform Comput Appl. 2017 Jan;31(1):32-51. doi: 10.1177/1094342015594519. Epub 2015 Jul 27.

Application Performance Analysis and Efficient Execution on Systems with multi-core CPUs, GPUs and MICs: A Case Study with Microscopy Image Analysis.

The international journal of high performance computing applications

George Teodoro, Tahsin Kurc, Guilherme Andrade, Jun Kong, Renato Ferreira, Joel Saltz

Affiliations

  1. Department of Computer Science, University of Brasília, Brasília, DF, Brazil.
  2. Department of Biomedical Informatics, Stony Brook University, Stony Brook, NY, USA; Scientific Data Group, Oak Ridge National Laboratory, Oak Ridge, TN, USA.
  3. Department of Computer Science, Federal University of Minas Gerais, Belo Horizonte, MG, Brazil.
  4. Department of Biomedical Informatics, Emory University, Atlanta, GA, USA.

PMID: 28239253 PMCID: PMC5319667 DOI: 10.1177/1094342015594519

Abstract

We carry out a comparative performance study of multi-core CPUs, GPUs and Intel Xeon Phi (Many Integrated Core-MIC) with a microscopy image analysis application. We experimentally evaluate the performance of computing devices on core operations of the application. We correlate the observed performance with the characteristics of computing devices and data access patterns, computation complexities, and parallelization forms of the operations. The results show a significant variability in the performance of operations with respect to the device used. The performances of operations with regular data access are comparable or sometimes better on a MIC than that on a GPU. GPUs are more efficient than MICs for operations that access data irregularly, because of the lower bandwidth of the MIC for random data accesses. We propose new performance-aware scheduling strategies that consider variabilities in operation speedups. Our scheduling strategies significantly improve application performance compared to classic strategies in hybrid configurations.

Keywords: Cooperative Execution; GPGPU; Hybrid Systems; Image Analysis; Intel Xeon Phi

References

  1. Anal Quant Cytol Histol. 2001 Aug;23(4):291-9 - PubMed
  2. IEEE Trans Image Process. 1993;2(2):176-201 - PubMed
  3. IEEE Trans Biomed Eng. 2010 Oct;57(10):2617-21 - PubMed
  4. J Am Med Inform Assoc. 2012 Mar-Apr;19(2):317-23 - PubMed
  5. Parallel Comput. 2013 Apr 1;39(4-5):189-211 - PubMed
  6. PLoS One. 2013 Nov 13;8(11):e81049 - PubMed
  7. Proceedings (IEEE Int Conf Bioinformatics Biomed). 2013 Dec;:229-236 - PubMed
  8. IEEE Trans Parallel Distrib Syst. 2014 May;2014:1063-1072 - PubMed
  9. IPDPS. 2012 May;2012:1093-1104 - PubMed
  10. IPDPS. 2013 May;2013:103-114 - PubMed
  11. Int J High Perform Comput Appl. 2013 Aug;27(3):263-272 - PubMed

Publication Types

Grant support