International Journal of Computer
Trends and Technology

Research Article | Open Access | Download PDF

Volume 53 | Number 1 | Year 2017 | Article Id. IJCTT-V53P102 | DOI : https://doi.org/10.14445/22312803/IJCTT-V53P102

Analysis of Parallel Architectures: SIMD, tightly-coupled MIMD, and loosely-coupled MIMD

Babasegun Adeleye, Salman Mohammed Jiddah

Citation :

Babasegun Adeleye, Salman Mohammed Jiddah, "Analysis of Parallel Architectures: SIMD, tightly-coupled MIMD, and loosely-coupled MIMD," International Journal of Computer Trends and Technology (IJCTT), vol. 53, no. 1, pp. 6-8, 2017. Crossref, https://doi.org/10.14445/22312803/IJCTT-V53P102

Abstract

This paper looks into the two forms of parallel computing which are Single Instruction Multiple Data (SIMD) and Multiple Instruction Multiple Data (MIMD), and further diving into the sub categories of MIMD; loosely coupled and tightly coupled with an overview of their individual architectures and their implementation in the computing industry. For a case study of their industrial application, this paper looked into the Intel Iris processor as a practical implementation of SIMD architecture in the industry, IBM Power8 for tightly coupled MIMD systems which is a supercomputer, and lastly the Beowulf cluster system which is an implementation of loosely coupled MIMD. Each architecture implementation is looked into regarding why it is being chosen to be used in the industry.

Keywords

Each architecture implementation is looked into regarding why it is being chosen to be used in the industry.

References

[1] Flynn, Michael J. "Very high-speed computing systems." Proceedings of the IEEE 54.12 (1966): 1901- 1909.
[2] Kaur, Mandeep, and RajdeepKaur. "A comparative analysis of SIMD and MIMD architectures." International Journal of Advanced Research in Computer Science and Software Engineering 3.9 (2013).
[3] Adve, Sarita V., and KouroshGharachorloo. "Shared memory consistency models: A tutorial." computer 29.12 (1996): 66-76.
[4] Hord, R. Michael. Parallel supercomputing in MIMD architectures. CRC press, 1993.
[5] Sampath, S., Bharat BhushanSagar, and B. R. Nanjesh. "Performance evaluation and comparison of MPI and PVM using a cluster based parallel computing architecture." Circuits, Power and Computing Technologies (ICCPCT), 2013 International Conference on. IEEE, 2013.
[6] Luo, Wen-lang, An-dong Xie, and Wen Ruan. "The construction and test for a small beowulf parallel computing system." Intelligent Information Technology and Security Informatics (IITSI), 2010 Third International Symposium on. IEEE, 2010.
[7] Zhu, Yongzhi, and Baoxiang Cao. "A Scalability Metric Based on Beowulf Cluster System." Distributed Computing and Applications to Business Engineering and Science (DCABES), 2010 Ninth International Symposium on. IEEE, 2010.
[8] Becker, Donald J., et al. "BEOWULF: A parallel workstation for scientific computation." Proceedings, International Conference on Parallel Processing. Vol. 95. 1995.
[9] Bhuyan, Laxmi N. "On the performance of loosely coupled multiprocessors." ACM SIGARCH Computer Architecture News 12.3 (1984): 256-262.
[10] Liu, Yingying, Dake Liu, and Wei Wang. "IIR parallelization on multi-datapath SIMD architecture." Integrated Circuits and Microsystems (ICICM), International Conference on. IEEE, 2016.
[11] Sharangpani, Harsh. "Intel® Itanium™ processor microarchitecture overview." Microprocessor Forum. 1999.
[12] Lameter, Christoph. "Numa (non-uniform memory access): An overview." Queue 11.7 (2013): 40.
[13] Sinharoy, Balaram, et al. "IBM POWER8 processor core microarchitecture." IBM Journal of Research and Development 59.1 (2015): 2-1.
[14] Junkins, Stephen. "The Compute Architecture of Intel® Processor Graphics Gen9." Intel whitepaper v1 (2015).