Scientific Computing Seminar
Spring 2008

Ravi Vadapalli
High Performance Computing Center

Tuesday, 4/29/2008
4:00 -- 5:00 p.m.
Experimental Sciences Building, Room 120

"TIGRE Environment for Extending Research and Educational Capabilities in Hydrocarbon Reservoir Simulations"


Texas Internet Grid for Research and Education (TIGRE) [1] is a state-wide Grid computing infrastructure project. The primary goal of TIGRE project is to provide operational Grid middleware and the secondary goal is to demonstrate TIGRE capabilities using applications from driving areas of interest to the State of Texas. They are bioscience and medicine, energy exploration and environmental modeling. Within the energy exploration application area, we have identified TAMU's message-passing based ensemble Kalman filter (EnKF) [2] data assimilation methodology for reservoir characterization. The EnKF approach deals with several geological models for comprehensive prediction of the hydrocarbon reserves and invokes Schlumberger ECLIPSE [3] -- a vendor supported software -- for the simulations. The application's parallelizability therefore depends on the number ECLIPSE licenses available. By Grid-enabling [4] the EnKF implementation and by pooling the licenses across the participating schools in TIGRE, the simulation runtime can be substantially reduced without compromising for the simulation accuracy. Pertinent details of this effort will be presented. Potential extensions especially through engaging Schlumberger to extend research and educational opportunities in energy exploration through TIGRE will be discussed.

[1] http://tigreportal.hipcat.net
[2] E. Arryo-Negrete, D.Devegowda, and A. Datta-Gupta, "Streamline-assisted ensemble Kalman filter for rapid and continuous reservoir model updating," SPE 104255, Int. Oil & Gas Conf and Exhibition, Beijing, China.
[3] http://www.slb.com/content/services/software/reseng/index.asp
[4] R Vadapalli, P Luo, T Kim, A Kumar and S Siddiqui, "Demonstration of Grid-enabled ensemble Kalman filter data assimilation methodology for reservoir characterization," Proc of ACM-MG08: Workshop on Grid enabling applications, Jan 30, Baton Rouge, LA (2008).

Return to Scientific Computing Seminar home page.