The proposed research is aimed at making significant advances in
both computational cosmology and parallel computing. New parallel
computing techniques developed as a part of this research,
including automatic load balancing and communication optimizations,
will be released in the form of a software framework for general-purpose
parallel computing. These will be used to develop a computational
framework (called N-Chilada) for simulation of particles in general.
Several computational cosmology algorithms will be implemented in the
particle framework. Such a particle framework will be useful in
applications beyond computational cosmology. Using the framework and its
algorithms, specific scientific studies will be performed to constrain
theories of cosmology by simulating large scale structure, understand
the sub-structure of galaxy clusters, understand galaxy formation within
a cosmological context, and understand the complex
interactions between dust, gas and planetesimals during the process
of planet formation.
Project Publications:
L.V.Kale, "The Virtualization Approach to Parallel Programming: Runtime
Optimizations and the State of the Art", LACSI Symposium, Santa Fe,
October 2002.
L.V.Kale, S.Kumar and K.Vardarajan, "A Framework for Collective
Personalized Communication", Proceedings of IPDPS'03, Nice, France, 2003.
L.V.Kale and S.Kumar, "Scaling Collective Multicast on High Performance
Clusters", Technical Report 03-04, Parallel Programmin Laboratory,
University of Illinois, 2003.
O.Lawlor and L.V.Kale, "Supporting Dynamic Parallel Object Arrays",
Concurrency and Computation: Practice and Experience, John Wiley and Sons,
2003, 15:371-393.
C.Huang, O.Lawlor and L.V.Kale, "Adaptive MPI", Proceedings of the
16th International Workshop on Languages and Compilers for Parallel
Computing (LCPC 03), 2003.
S.Kumar and L.V.Kale, "Collective Multicast on Fat-Tree Networks",
Technical Report 03-11, Parallel Programing Laboratory, University
of Illinois, Urbana, 2003.
S.Kumar and L.V.Kale, "Opportunities and Challenges of Modern
Communication Architectures: Case Study with QsNet", CAC Workshop
at IPDPS'04, Santa Fe, 2004.
L.V.Kale, "Performance and Productivity in Parallel Programming via
Processor Virtualization", Proceedings of the First Intl. Workshop
on Productivity and Performance in High-End Computing (at HPCA-10),
February 2004.
L.V.Kale, M.Hills and C.Huang, "An Orchestration Language for Parallel
Objects", Proceeding of Seventh Workshop on Languages, Compilers,
and Run-time Support for Scalable Systems (LCR 04), 2004.
X.Ma, M.Winslett, J.Norris, J.Jiao, and R.Fiedler, "GODIVA:
Lightweight Data Management for Scientific Visualization", Proceedings
of the 20th International Conference on Data Engineering, 2004.
Software Products:
Liveviz:
Real-time visualization of data streamed from an executing Charm++ program