Parallel Computing - A Case Study on Fine-Grained Parallelism Nurhayati Abd Manap #1, Mohamed Faidz Mohamed Said #2 # Faculty of Computer & Mathematical Science, Universiti Teknologi MARA 70300 Seremban, Negeri Sembilan, MALAYSIA 1 nurhayati5538@gmail.com 2 mohdfaidz@uitm.edu.my Abstract— This paper represents about a case study on fine-grained parallelism on different approach. Fine–grained parallelism is description about the small component and individual task in small code of size and execution time. Also in fine-grained parallelism, these tasks are assigned individually too many processors and the program is stopped to a large number of small tasks and in order to reduce the execution time of the critical path is used a handful of processor cores. Fine-grained parallelism is usually more difficult and represents a much more complex use of parallelism than is found in the use of threads to do than coarse-grained parallelism, while both type are depends on the similar principles. In fine grained parallelism, what generally supposed of as a single, undividable calculation is divided between processor. This model allows the codes to be written with higher maintainability and portability since all parallel segments perform as simple loops. The result is fine-grained implementation is able to speed up the resolution of the system equation and shows substantial performance improvements. Keywords: Fine-grained, Type of parallelism, Parallelism REFERENCES [1] T.-H. Weng and C.-P. Chung, "Exploiting fine-grain parallelism in the H.264 deblocking filter by operation reordering," pp. 76–87, 2014. [2] L. Luo, J. R. Edwards, H. Luo, and F. Mueller, "A fine-grained block ILU scheme on regular structures for GPGPUs," Computer & Fluids, vol. vol 119, pp. 149-161, 2015. [3] F. Pratas, P. Trancoso, L. Sousa, A. Stamatakis, G. Shi, and V. Kindratenko, "Fine-grain parallelism using multi-core, Cell/BE, and GPU Systems," pp. 365–390, 2012. [4] J. Martínez-Frutos, P. J. Martínez-Castejón, and D. Herrero-Pérez, "Fine-grained GPU implementation of assembly-free iterative solver for finite element problems," Computer and Structures, vol. vol 157, pp. 9-18, 2015. [5] H. Rakotoarivelo, F. Ledoux, and F. Pommereau, "Fine-grained locality-aware parallel scheme for anisotropic mesh adaptation," Procedia Engineering, vol. vol 163, pp. 123-135, 2016. [6] J. Wu, W. Li, and G. Jeon, "From coarse- to fine-grained implementation of edge-directed interpolation using a GPU," Information Sciences, vol. vol 385-386, pp. 457-478, 2017. [7] J. Ren, G. Chen, and X. Li, "A fine grained digital textile printing system based on image registration," pp. 52–160, 2007. [8] M. Manjunathaiah, "Fine-grained multi-phase array designs," pp. 1076–1082, 2013. [9] P. Song, Y. Liu, C. Liu, and D. Qian, "ParaFlow: Fine-grained parallel SDN controller for large-scale networks," Journal of Network and Computer Applications, pp. 46–59, 2017. [10] S. Steinfadt, "Fine-grained parallel implementations for SWAMP+ Smith–Waterman alignment," Parallel Computing, vol. vol 39, pp. 819-833, 2013. [11] L. Karlsson, B. Kågström, and E. Wadbro, "Fine-grained bulge-chasing kernels for strongly scalable parallel QR algorithms," Parallel Computing, vol. vol 40, pp. 271-288, 2014. [12] W. Jiang et al., "FIPIP: A novel fine-grained parallel partition based intra-frame prediction on heterogeneous many-core systems," Future Generation Computer Systems, vol. vol 78, pp. 316-329, 2016. [13] M. Francis-Landau, B. Xuey, J. Eisner, and V. Sarkary, "Fine-grained parallelism in probabilistic parsing with Habanero Java," pp. 1-4, 2016. [14] Y. Guoa, V. Vlassov, R. Ashok, R. Weiss, and C. A. Moritze, "Synchronization coherence:Atransparent hardware mechanism for cache coherence and fine-grained synchronization," vol. vol 68, 2008. [15] N. Abd Manap. (2017). Parallel Computing - A Case Study on Fine-grained Parallelism. Available: https://www.youtube.com/watch?v=ztzL02Ob1FI [Accessed: 28-Nov-2017]