Addressing Challenges of Adaptivity and Scale in Parallel Scientific Simulations

Manish Parashar
Rutgers University

Abstract:
Simulations are playing an increasingly important role in science and
engineering and are rapidly becoming critical research
modalities. Large scale, coupled and dynamically adaptive simulations
can enable highly accurate solutions to realistic models, and provide
dramatic insight into complex phenomena. However, the phenomena being
modeled by these simulations are inherently dynamic and heterogeneous,
spanning multiple time and space scales. As a result, their large
scale parallel implementation presents significant challenges. In this
talk, I will present a computational engine that incorporates
algorithmic and infrastructure solutions to addresses these challenges
and enables efficient and scalable implementation of adaptive
formulations in a wide range of application domains. Specifically, I
will focus on addressing the space, time and computational
heterogeneity, and dynamism in simulations based on parallel adaptive
mesh refinement.

Biography: 
Manish Parashar is Professor of Electrical and Computer
Engineering at Rutgers University, where he also is co-director of the
Center for Advanced Information Processing (CAIP) and director of the
Applied Software Systems Laboratory (TASSL). He received a BE degree
in Electronics and Telecommunications from Bombay University, India
and MS and Ph.D. degrees in Computer Engineering from Syracuse
University. He has received the Rutgers Board of Trustees Award for
Excellence in Research (2004-2005), NSF CAREER Award (1999) and the
Enrico Fermi Scholarship from Argonne National Laboratory (1996). His
research interests include computational science and applied parallel
& distributed computing, with a specific focus on solving scientific
and engineering problems on very large systems. A key aspect of his
current research is the integration of physical and computational
systems. For more information please visit
http://www.caip.rutgers.edu/~parashar/.