Title :
A highly scalable parallel boundary element method for capacitance extraction
Author :
Hsiao, Yu-Chung ; Daniel, Luca
Author_Institution :
Massachusetts Inst. of Technol., Cambridge, MA, USA
Abstract :
Traditional parallel boundary element methods suffer from low parallel efficiency and poor scalability due to the long system solving time bottleneck. In this paper, we demonstrate how to avoid such a bottleneck by using an instantiable basis function approach. In our demonstrated examples, we achieve 90% parallel efficiency and scalability both in shared memory and distributed memory parallel systems.
Keywords :
parallel processing; partial differential equations; capacitance extraction; distributed memory parallel systems; instantiable basis function approach; parallel boundary element method; shared memory systems; Argon; Equations; Heating; Moment methods; Neodymium; Silicon; Boundary element method; Capacitance extraction; Field solver; Parallel computing;
Conference_Titel :
Design Automation Conference (DAC), 2011 48th ACM/EDAC/IEEE
Conference_Location :
New York, NY
Print_ISBN :
978-1-4503-0636-2
