A team led by Rutgers University’s Center for Autonomic Computing (CAC) won first place in the IEEE SCALE 2011 Challenge for their scalable ensemble-based oil-reservoir simulations. The enhanced oil recovery simulation was run on what is described as ‘an elastic federated HPC-cloud’ composed of two remote IBM Blue Gene/P supercomputers.
The project addressed the requirements of ensemble applications in navigating large parameter spaces in order to optimize strategies and minimize uncertainty. The project claimed to demonstrate that, ‘the cloud abstraction can be used to run ensemble applications on a geographically distributed federation of supercomputing systems.’
Demonstrator components included Rutgers iCode (www.oilit.com/links/1107_20) and CometCloud (www.oilit.com/links/1107_21), IBM’s DeepCloud (www.oilit.com/links/1107_22) and the University of Texas at Austin’s ‘implicit parallel accurate reservoir simulator (IPARS www.oilit.com/links/1107_23). The Blue Gene supercomputers were located at IBM’s T.J. Watson Research Center (NY, USA) and King Abdullah University of Science and Technology (KAUST) (Jeddah, Saudi Arabia). More from www.oilit.com/links/1107_24.
© Oil IT Journal - all rights reserved.