Loading Events
  • This event has passed.

Accelerate Large Scale Geoscience Analysis Problems – A Case Study on Solar Irradiance Simulation

February 23, 2017
2:00 pm - 3:00 pm

Event Navigation

Dr. Jyh (Steve) Liu
Department of Computer Science & Engineering

Professor

Abstract

Geoscience analysis problems need to process massive amount of data to simulate complex environmental processes. In this talk, Dr. Liu will share his team’s successful experience in both acceleration and scale up of a sophisticated solar irradiance simulation model. The results showed that some portion of the complex software system can be best accelerated by the GPU coprocessor, while the overall computing task for large area study can be distributed across different nodes. The work has also been successfully ported to the large cluster environment of the supercomputing center. The experience suggests that a flexible testbed environment to incorporate both fine grained acceleration and scaling up of the computer nodes will enable researchers to drastically improve their research productivity by identification of computing bottlenecks, and mapping of such elements to parallel computing coprocessors. Furthermore, we are also interested in developing an interactive processing and visualization processing pipeline on the newly acquired super computer cluster Terra and its file system to support the next level of performance for temporal-spatial analysis, modeling, and visualization.

Biography

Dr. Liu received his B.S. and M.S. degrees in Electrical Engineering from the National Cheng Kung University in Tainan, Taiwan at 1979 and 1981, respectively. He received his PhD in Electrical Engineering and Computer Science Department of the University of Michigan, Ann Arbor, at 1989. He then joined the Computer Science and Engineering department of Texas A&M University at 1989. His current research interests include, but are not limited to, system modeling, real-time and distributed systems, Global Satellite Navigation Systems, and Cyber Physical System security. He also works on image and signal processing problems for years. His research has been supported by a wide range of sponsors from federal and state agencies, and private companies.

Contact

Dr. Michael Bishop: michael.bishop@tamu.edu

Details

Date:
February 23, 2017
Time:
2:00 pm - 3:00 pm

Venue

Eller O&M Bldg., Room 807