eScience seminar Tuesday, June 26th, EE 303 4-5pm

Please join us *Tuesday*, June 26th, 4-5pm in EE303. Refreshments (drinks
and pizza) will be provided.

*Seunghwa Kang *(Biocellion)
Seunghwa Kang completed his Ph.D. at Georgia Institute of Technology in
2011 and joined Pacific Northwest National Laboratory as a postdoctoral
researcher. He is also affiliated in Institute for Systems Biology. His
early research interest was on optimizing various kernels and applications
for multicore processors and accelerators and analyzing the match between
various applications and programming models and architectures. His current
research interest is at the intersection of computational biology and
high-performance computing; he solves challenging computational biology
problems by designing novel models and algorithms and exploiting
high-performance computing

*Biocellion: Accelerating Multicellular Biological System Simulation**
*
Interesting biological systems (e.g. biofilm, yeast colony, and human
tissue) consist of a large number of cells, and interactions among the
cells determine system behavior. Simulation of multicellular biological
systems helps biologists in relating cell level knowledge to multicellular
system behavior.

There are multiple approaches to model multicellular biological systems
with trade-offs in computational complexity and model capability.
High-performance computing has strong potential to enable biologists to
build more capable models, but this potential has not been realized due to
the difficulty of writing software for large parallel computers. I will
present Biocellion, a high-performance acceleration framework for
multicellular biological system simulation. Biocellion asks users to
describe model specifics in sequential C code but hides all the challenges
in distributing multiscale irregular computations to large scale
distributed computers. We developed yeast colony and soil bacteria models
using Biocellion as early examples. We have tested Biocellion on
distributed clusters and Amazon Cloud and plan to distribute Biocellion
through Amazon Cloud to minimize user efforts to use the framework.

Upcoming Seminar:

** July 31, 4 PM* (EE303)

Sinan al-Saffar (PNNL)

Scalable Semantic Graph Analyses

Advertisements