Algorithmic Foundation for Benchmarking of Computational Platforms Running Asymmetric Cipher Systems
The 0–1 Knapsack Problem is a well known NP-complete problem. It is used as the core primitive in several asymmetric cipher systems. Designing such systems requires a reliable method of computational platform benchmarking. But the existing general-purpose benchmarks are not accurate enough, as they are mostly based on floating-point arithmetics, while the Knapsack Problem relies on large amount of calculations with very long integers. Therefore, a new specialized benchmark is required to get accurate performance estimates. In this paper we study some features of exact parallel algorithms for the Knapsack Problem, as well as load balancing techniques for them. We then choose several algorithms based on their scalability and applicability to the asymmetric cipher system analysis and suggest a new algorithmic foundation for computational platform benchmarking comprised of these algorithms.
KeywordsParallel computation Benchmarking Asymmetric cipher systems Knapsack Problem
This work was supported by the MEPhI Academic Excellence Project (agreement with the Ministry of Education and Science of the Russian Federation of August 27, 2013, project no. 02.a03.21.0005).
- 1.HPL — a portable implementation of the High-Performance Linpack Benchmark for distributed-memory computers (2016). http://www.netlib.org/benchmark/hpl/
- 2.HPCG benchmark (2016). http://hpcg-benchmark.org/
- 3.Brief introduction | Graph 500 (2016). http://www.graph500.org/
- 6.Kupriyashin, M.A., Borzunov, G.I.: Computational load balancing algorithm for parallel Knapsack packing tree traversal. Procedia Comput. Sci. 88, 330–335 (2016). http://www.sciencedirect.com/science/article/pii/S1877050916317008CrossRefGoogle Scholar
- 8.Kupriyashin, M.A., Borzunov, G.I.: Finding the exact solutions of the Knapsack Problem using dynamic programming. PNRPU Bull. Electrotechnics Inf. Technol. Control Syst. 17, 121–130 (2016)Google Scholar