Abstract
Problem statement: Memetic Algorithm (MA) is a form of population-based hybrid Genetic Algorithm (GA) coupled with an individual learning procedure capable of performing local refinements. Here we used genetic algorithm to explore the search space and simulated annealing as a local search method to exploit the information in the search region for the optimization of VLSI netlist bi-Partitioning problem. However, they may execute for a long time, because several fitness evaluations must be performed. A promising approach to overcome this limitation is to parallelize this algorithms. General Purpose computing over Graphical Processing Units (GPGPUs) is a huge shift of paradigm in parallel computing that promises a dramatic increase in performance. Approach: In this study, we propose to implement a parallel MA using graphics cards. Graphics Processor Units (GPUs) have emerged as powerful parallel processors in recent years. Using of Graphics Processing Units (GPUs) equipped computers; it is possible to accelerate the evaluation of individuals in Genetic Programming. Program compilation, fitness case data and fitness execution are spread over the cores of GPU, allowing for the efficient processing of very large datasets. Results: We perform experiments to compare our parallel MA with a Sequential MA and demonstrate that the former is much more effective than the latter. Our results, implemented on a NVIDIA GeForce GTX 9400 GPU card. Conclusion: Its indicates that our approach is on average 5×faster when compared to a CPU based implementation. With the Tesla C1060 GPU server, our approach would be potentially 10×faster. The correctness of the GPU based MA has been verified by comparing its result with a CPU based MA.
Highlights
Physical design of VLSI circuits is the process of mapping structural representations of circuits into layout representation
This study addresses the problem of VLSI netlist partitioning with the objective of optimizing cutset where, βi is the number of cells in partition i, φ is the total number of cells in the circuit, α is the tolerance which is equal zero in case of a perfect balance
Memetic Algorithms (MA) are effective in solving many practical problems in science, engineering and business domains, they may execute for a long time to find solutions for some huge problems, because several fitness evaluations must be performed
Summary
Physical design of VLSI circuits is the process of mapping structural representations of circuits into layout representation. Due to the complexity of the physical design phase it is usually broken down to sub problems like, partitioning, placement and routing which are solved one after the other. This study is concerned with the circuit partitioning problem. Circuit net list partitioning is an important step in VLSI physical design. This involves the breakup of a circuit into smaller parts for ease of design, layout and testability. The main objective of circuit partitioning is minimization of number of interconnections between the partitions
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
