Optimal additional data layers amount determining for interconnect latency hiding scheme
https://doi.org/10.20914/2310-1202-2017-1-95-98
Abstract
The key component of parallel computing efficiency is the structure of data exchange between computing nodes. It is necessary to reduce the latency of data exchange to improve the efficiency of parallel computing. A B+2R algorithm for overlapping delays in the data exchange was offered for this purpose. Existing works do not offer a method for selecting the additionally transmitted data layer count R. We introduce the models of all systems affecting the parallel calculation time. It makes possible to apply the analytical optimization. We introduce a data transmission network and a parallel computing application models. We consider the cell calculation time is a constant value. The cell calculation time depends on the specific calculation parameters. We introduce an estimate of the computation time. Computation time depends on the additionally transmitted data layers count. Further we introduce the derivative of computation time equation. We use lowest positive real roots of the cubic equation. It’s possible that the final cubic equation hasn’t real positive roots. It’s mean that local structured grid calculation time is much bigger than network latency. In that case, it’s not recommended to use latency hiding schemes. Otherwise we recommend to use R equal to 1. Purpose of out research is to find optimal R. Optimal R value should lead to a calculation time equation minimum. The method proposed in the paper correspond to experimental result. Designed analytical model for B+2R algorithm makes possible to select optimal R value, which leads to the best calculation speedup.
References
1. Brandon G.A. Kalyan S.P. Sudip K.S. Efficient Simulation of Agent-Based Modoels on Multi-GPU and Multi-Core Clusters. Proceedings of SIMUTools. 2010 March 15–19
2. Kalmicov, V.V., Ibraev, R.A. Latency hiding algorithm for shallow water equations solve on parallel computers. Vestnik UGATU [Ufa State Aviation Technical University] 2013, no 5, pp. 252-259. (in Russian)
3. Jaehyuk H. Hardware Techniques to Reduce Communication Costs in Multiprocessors. Doctoral dissertation, 2006.
4. Cicotti P. Tarragon: a programming model for latency-hiding scientific computations. Doctoral dissertation, 2011.
5. Alameldeen Alaa R. Using Compression to improve chip multiprocessor performance. Doctoral dissertation, 2006.
6. Afsahi A. Design and Evaluation of Communication Latency Hiding/Reduction Techniques for Message-Passing Environments. Doctoral dissertation, 2000.
7. Chen Li-li, Huang Jian-xin, Zhang Jing A Latency-Hiding Algorithm for ABMS on Paral-lel/Distributed Computing Environment. ACM/IEEE/SCS 26th Workshop on Principles of Ad-vanced and Distributed Simulation, 2012.
8. Yong Chen, Surendra Byna, Xian-He Sun, Rajeev Thakur et al. Hiding I/O latency with pre-execution prefetching for parallel applications. In Proceedings of the 2008 ACM/IEEE conference on Supercomputing (SC '08). IEEE Press, Piscataway, NJ, USA, 2008, Article 40, pp. 10.
9. Hakan Grahn, Comparative Evaluation of Latency-Tolerating and -Reducing Techniques for Hardware-Only and Software-Only Directory Protocols. Journal of Parallel and Distributed Computing 60, 2000, pp. 807-834.
Review
For citations:
Novikov A.B., Evtushenko G.I. Optimal additional data layers amount determining for interconnect latency hiding scheme. Proceedings of the Voronezh State University of Engineering Technologies. 2017;79(1):95-98. (In Russ.) https://doi.org/10.20914/2310-1202-2017-1-95-98
Views:
629
Email this article 
Post a Comment 
