Abstract
Hardware accelerators (or custom hardware circuit) incorporate design practices that involve multiple convoluted orthogonal optimization requisites at various abstraction levels. The convoluted optimization requisites often demand intelligent decision-making strategies during high-level synthesis (HLS) to determine the architectural solution based on conflicting metrics such as power and performance as well as exploration speed and quality of results. Traditional heuristic-driven approaches using genetic algorithm, simulated annealing, etc., fall short considerably on the above orthogonal aspects especially in their ability to reach real optimal solution at an accelerated tempo. This chapter introduces a new particle swarm optimization-driven multi-objective design space exploration methodology based on power-performance trade-off tailored for targeting application-specific processors (hardware accelerators). Furthermore, as the performance of particle swarm optimization is known for being highly dependent on its parametric variables, in the proposed methodology, sensitivity analysis has been executed to tune the baseline parametric setting before performing the actual exploration process.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Ferrandi F, Lanzi PL, Loiacono D, Pilato C, Sciuto D (2008) A multi-objective genetic algorithm for design space exploration in high-level synthesis. ISVLSI:417–422
Gajski D, Dutt ND, Wu A, Lin S (1992) High level synthesis: introduction to chip and system design. Kluwer Academic Publishers, Norwell
Gallagher JC, Vigraham S, Kramer G (2004) A family of compact genetic algorithms for intrinsic evolvable hardware. IEEE Trans Evolut Comput 8(2):1–126
Kennedy J, Eberhart RC (1995) Particle swarm optimization. In: Proceedings of the IEEE international conference on neural networks, Anchorage, pp 1942–1948
Krishnan V, Katkoori S (2006) A genetic algorithm for the design space exploration of datapaths during high-level synthesis. IEEE Trans Evolut Comput 10(3):213–229
Mishra VK, Sengupta A (2014) MO-PSE: adaptive multi objective particle swarm optimization based design space exploration in architectural synthesis for application specific processor design. Elsevier J Adv Eng Softw 67:111–124
Sengupta A, Sedaghat R, Sarkar P (2012) A multi structure genetic algorithm for integrated design space exploration of scheduling and allocation in high level synthesis for DSP kernels. Elsevier J Swarm Evolut Comput 7:35–46
Sengupta A, Mishra VK, Sarkar P (2013) Rapid search of pareto fronts using D-logic exploration during multi-objective tradeoff of computation intensive applications. In: Proceedings of IEEE 5th Asian Symposium on Quality Electronic Design (ASQED), Malaysia, pp 113–122
Torbey E, Knight J (1998a) High-level synthesis of digital circuits using genetic algorithms. In: Proceedings of the international conference on evolutionary computation, Anchorage, pp 224–229
Torbey E, Knight J (1998b) Performing scheduling and storage optimization simultaneously using genetic algorithms. In: Proceedings of the IEEE midwest symposium on circuits and systems, Anchorage, pp 284–287
Trelea CI (2003) The particle swarm optimization algorithm: convergence analysis and parameter selection. Elsevier Inf Process Lett 85(6):317–325
Acknowledgment
I thank Elsevier for allowing the reuse of the materials from ‘Elsevier Journal on Advances in Engineering Software’, Vol 67, Vipul Kumar Mishra, Anirban Sengupta, MO-PSE: Adaptive Multi Objective Particle Swarm Optimization Based Design Space Exploration in Architectural Synthesis for Application Specific Processor Design, Pages. 111–124, for my publication of book chapter. I would also like to thank Indian Institute of Technology, Indore, India, for providing me with all the resources for carrying out this work.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer India
About this chapter
Cite this chapter
Sengupta, A. (2015). Design Space Exploration of Datapath (Architecture) in High-Level Synthesis for Computation Intensive Applications. In: Bhuvaneswari, M. (eds) Application of Evolutionary Algorithms for Multi-objective Optimization in VLSI and Embedded Systems. Springer, New Delhi. https://doi.org/10.1007/978-81-322-1958-3_6
Download citation
DOI: https://doi.org/10.1007/978-81-322-1958-3_6
Published:
Publisher Name: Springer, New Delhi
Print ISBN: 978-81-322-1957-6
Online ISBN: 978-81-322-1958-3
eBook Packages: EngineeringEngineering (R0)