NEW to Berkeley Lab Computing Sciences Team
June 30, 2009
Seung-Jai Min, Future Technologies Group
 |
| Seung-Jai Min |
As the newest member of the Berkeley Lab's Future Technology Group, Seung-Jai Min will be working on Unified Parallel C (UPC) projects. UPC is an extension of the C programming language designed for high-performance computing on large-scale parallel machines.
Originally from Seoul, South Korea, Min says that he has always been fascinated by speedup graphs that illustrate how much faster his programs run. The excitement eventually inspired him to focus on high-performance computing. He earned his Bachelor of Science Degree in electrical engineering at Sogang University; Master of Science in Computer Science and Electrical Engineering at Seoul National University, in South Korea; and a Doctorate from the school of Electrical and Computer Engineering at Purdue University in Indiana.
“The compiler research and development for high performance computers that I did for my PhD thesis is very similar to the work that I will be doing for the Future Technologies Group,” says Min. “As a graduate student found the research very rewarding and look forward to making contributions in the field.”
In his spare time, Min enjoys cooking exotic foods, tango dancing and reading. He says he is thrilled to be in California's Bay Area, one of the best places in the world to tango.
Viraj Rameshrao Paropkari, NERSC High Performance Computing Consultant
 |
| Viraj Rameshrao Paropkari |
As NERSC's newest SciDAC HPC Consultant, Viraj Rameshrao Paropkari will be playing a dual role in the User Services and Software Integration teams, both directly assisting users and improving the software used at NERSC.
Before coming to Berkeley he was a visiting scholar at the University of Illinois-Urbana Champaign where he worked with professor Laxmikant Kale in the Parallel Programming Laboratory. During this year, he contributed to improving the astrophysics application ChaNGa and was a software developer for the Blue Waters project.
“These two projects helped me understand key issues like scalability, optimization and debugging,” says Paropkari. “This experience gives me a great foundation for helping NERSC researchers make their scientific applications run more efficiently.”
Although Paropkari specialized in computer science as an undergraduate student at the University of Pune, India, it was actually a computational science and engineering class with David Bader during his graduate studies at the Georgia Institute of Technology, in Atlanta, that inspired him to pursue a career in high performance computing. The class allowed him to work with Cell processors, one of the first in heterogeneous computing. A summer internship at the Oak Ridge National Laboratory then allowed him to apply this knowledge to scientific applications like climate modeling and molecular dynamics.
Originally from India where temperatures are typically scorching, Paropkari says he enjoys the Bay Area's pleasant climate and diverse selection of food. In his spare time, Paropkari likes to hike, play squash, watch movies, spend time with family and explore historical sites.
Khaled Ibrahim, Future Technologies Group
 |
| Khaled Ibrahim |
As a relatively new member of the Future Technologies Group, Khaled Ibrahim will be investigating performance modeling and code optimization for parallel architectures.
“Having something that is faster, bigger, or stronger always excited me. Computer system evolvement was driven by the need for greater power. I am fascinated by HPC because they are constantly hungry for more computational power,” says Ibrahim. “As a research scientist in the Future Technologies Group, I am thrilled to contribute to codes that will help computers run faster and stronger.”
Although he is new to the Berkeley Lab, the challenge of optimizing parallel architectures is not new to Ibrahim. Before he came to the Future Technologies Group, Ibrahim investigated methods for optimize codes to improve the performance in parallel applications at the French National Institute for Research in Computer Science and Control (INRIA). As a graduate student at North Carolina State University, he used speculation techniques to improve the performance of parallel computers. He notes that both of these experiences complemented each other, and allowed him to improve the performance and efficiently solve challenging problems.
Originally from Port Said, Egypt, Ibrahim says he most enjoys California's sunshine and cultural diversity. Ibrahim enjoys spending his spare time with his children, building model aircraft, playing basketball and piano.
