Abstract
Scalability is an important performance metric of parallel computing, but the traditional scalability metrics only try to reflect the scalability for parallel computing from one side, which makes it difficult to fully measure its overall performance. This paper studies scalability metrics intensively and completely. From lots of performance parameters of parallel computing, a group of key ones is chosen and normalized. Further the area of Kiviat graph is used to characterize the overall performance of parallel computing. Thereby a novel scalability metric about iso-area of performance for parallel computing is proposed and the relationship between the new metric and the traditional ones is analyzed. Finally the novel metric is applied to address the scalability of the matrix multiplication Cannon’s algorithm under LogP model. The proposed metric is significant to improve parallel computing architecture and to tune parallel algorithm design.
Similar content being viewed by others
References
Chen GL (2002) Parallel computer architecture. Higher Education Press, Beijing
Amdahl GM (1967) Validity of the single-processor approach to achieving large scale computing capabilities. In: Proceedings of the AFIPS Spring Joint Computer Conference, pp 483–485
Gustafson JL (1988) Reevaluating Amdahl’s law. Commun ACM 31(5):532–533
Zorbas JR, Reble DJ, Vankooten RE (1989) Measuring the scalability of parallel computer systems. In: Proceedings of Supercomputing, pp 832–841
Nussbaum D, Agarwal A (1991) Scalability of parallel machines. Commun ACM 34(3):57–61
Sun XH, Gustafson JL (1991) Toward a better parallel performance metric. Parallel Comput 17(10):1093–1109
Sun XH, Ni LM (1993) Scalable problems and memory-bounded speedup. J Parallel Distrib Comput 19(1):27–37
Grama A, Gupta A, Kumar V (1993) Iso-efficiency: measuring the scalability of parallel algorithms and architectures. IEEE Parallel Distrib Technol 1(3):12–21
Sun XH, Rover DT (1994) Scalability of parallel algorithm-machine combinations. IEEE Trans Parallel Distrib Syst 5(6):599–613
Zhang XD, Yan Y, He K (1994) Latency metric: an experimental method for measuring and evaluating parallel program and architecture scalability. J Parallel Distrib Comput 22(3):392–410
Wu XF, Li W (1998) Performance models for scalable cluster computing. J Syst Archit 44(3):189–205
Chen J, Li XM (2001) Near-optimal scalability: a practical scalability metric. Chin J Comput 24(2):179–182
Wang YL, Yang XD (2001) A more effective scalability model for parallel system. Chin J Comput 24(1):84–90
Chi LH, Liu J, Hu QF (2005) Evaluation and test for scalability of numerical parallel computation. J Comput Res Dev 42(6):1073–1078
Yang XJ (2012) Sixty years of parallel computing. Comput Eng Sci 34(8):1–10
Hao SX, Zeng GS, Tan YM (2010) Scalability analysis of heterogeneous computing based on computation task and architecture to match. Chin J Electron 38(11):2585–2589
Chi LH, Liu J, Li XM et al (1999) The scalability of parallel algorithms and architectures. J Comput Res Develop 36(1):47–51
Esmaeilzadeh H, Blem E, Amant R et al (2012) Power limitations and dark silicon challenge the future of multicore. ACM Trans Comput Syst 30(3):11–38
Hambrusch SE, Khokhar A (1996) C3: A parallel model for coarse-grained machines. J Parallel Distrib Comput 32(2):139–154
Wang L (2013) Research on task parallel programming model. Chin J Softw 24(1):77–90
Li XM et al (2007) Numerical parallel algorithms and software. Beijing Science Press, Beijing
Acknowledgments
This work was supported by the National High-Tech Research and Development Plan of China under grant No. 2009AA012201; the National Natural Science Foundation of China under grant No. 61363041, No. 61272107, No. 61202173 and No. 61103068; the Ph.D. Programs Foundation of Ministry of Education (grant No. 20110072120017), the Open Project Program of the State Key Lab of CAD&CG (Zhejiang University, grant No. A1311); the State Key Laboratory for Novel Software Technology (Nanjing University, grant No. KFKT2012B24), the Open Project Program of the National Laboratory of Pattern Recognition (grant No. 201103187); the Program of Shanghai Subject Chief Scientist under grant No. 10XD1404400; the special Fund for Fast Sharing of Science Paper in Net Era by CSTD under grant No. 20110740001.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Xiong, H., Zeng, G., Zeng, Y. et al. A novel scalability metric about iso-area of performance for parallel computing. J Supercomput 68, 652–671 (2014). https://doi.org/10.1007/s11227-013-1057-x
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11227-013-1057-x