Universal switch modules for FPGA design
ACM Transactions on Design Automation of Electronic Systems (TODAES)
Computers and Intractability: A Guide to the Theory of NP-Completeness
Computers and Intractability: A Guide to the Theory of NP-Completeness
A Template for Scatter Search and Path Relinking
AE '97 Selected Papers from the Third European Conference on Artificial Evolution
Scatter Search: Methodology and Implementations in C
Scatter Search: Methodology and Implementations in C
Parallel Metaheuristics: A New Class of Algorithms
Parallel Metaheuristics: A New Class of Algorithms
A hardware Memetic accelerator for VLSI circuit partitioning
Computers and Electrical Engineering
Automatic Identification of Parallelism in Handel-C
DSD '08 Proceedings of the 2008 11th EUROMICRO Conference on Digital System Design Architectures, Methods and Tools
Implementation of a genetic algorithm on a virtex-ii pro FPGA
Proceedings of the ACM/SIGDA international symposium on Field programmable gate arrays
A Methodology for Rapid Optimization of HandelC Specifications
RSP '09 Proceedings of the 2009 IEEE/IFIP International Symposium on Rapid System Prototyping
Hi-index | 0.00 |
Scatter Search [1] is an effective and established population-based meta-heuristic that has been used to solve a variety of hard optimization problems. However, like most population-based meta-heuristics, the time required to find high-quality solutions can become prohibitive as problem sizes grow. In this paper, we present a hardware implementation of scatter search on a Field-Programmable Gate-Array (FPGA). Our objective is to improve the runtime of scatter search by exploiting the potentially massive performance benefits that are available through the native parallelism in hardware. When implementing scatter search we employ Handel-C [2] - a programming language specifically designed to enable software developers to easily synthesize C-like programs into synchronous hardware. As far as we know, this is the first time that scatter search has been implemented in hardware (of any form). Our empirical results show that by effectively exploiting data parallelism and pipelining a 28x speedup over software can be achieved.