iCode wins the first place for the IEEE SCALE 2011 Challenge
Blue Gene sniffs for Black Gold in the Cloud thanks to Rutgers, IBM & UT Austin
A multi-institutional team consisting of The Cloud and Autonomic Computing Center (Rutgers University), IBM T.J. Watson Research Center and Center for Subsurface Modeling (The University of Texas at Austin) was awarded first place in the IEEE SCALE 2011 Challenge for their demonstration titled “A Scalable Ensemble-based Oil-Reservoir Simulations using Blue Gene/P-as-a-Service”. The project team was led by Professor Manish Parashar and consisted of Moustafa AbdelBaky and Hyunjoo Kim (CAC, Rutgers Univ.), Kirk Jordan, Hani Jamjoom, Vipin Sachdeva, Zon-Yin Shae and James Sexton (IBM T.J. Watson Research Center), and Gergina Pencheva, Reza Tavakoli, and Mary F. Wheeler (CSM, UT Austin).
Supercomputing: Now there’s an app for that ! ®
For questions or comments about this site, send us an email
‘There’s an app for that’ is a trademark of Apple Computer, Inc.
For more information about iCode please click here. For more information about CometCloud please click here. For more information about the demonstration please click here
Kirk E. Jordan
Mary F. Wheeler
The team would like to thank King Abdullah University of Science and Technology (KAUST) for allowing us to use their Supercomputer system.
The project, demonstrated by Moustafa AbdelBaky, addresses the requirements of ensemble applications in navigating large parameter spaces in order to optimize strategies and minimize uncertainty. This class of applications requires the effective utilization of extreme scale computing systems, but their deployments are often hindered by the low-level interfaces of current supercomputing systems and required expert knowledge to operate these systems.
The demonstration showed how the cloud abstraction, with its elastic as-a-service usage model, can be effectively used to support these ensemble applications on a geographically distributed federation of supercomputing systems. In particular, it demonstrated an Enhanced Oil Recovery (EOR) that was configured and deployed on an elastic federated HPC-cloud composed of two IBM Blue Gene/P supercomputer systems located at IBM T.J. Watson Research Center (NY, USA) and King Abdullah University of Science and Technology (KAUST) (Jeddah, Saudi Arabia), using a pervasive portal running on an iPad -- Supercomputing: Now there’s an app for that!
The demonstrated solution is built on iCode and CometCloud from the Cloud and Autonomic Computing Center at Rutgers University, DeepCloud from IBM T.J. Watson, and IPARS/EnKF from CSM, University of Texas at Austin.
SCALE 2011, the 4th IEEE International Scalable Computing Challenge was held on May 26th 2011 in conjunction with the 11th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing at Newport Beach CA, USA. The objective of the challenge was to highlight and showcase real-world problem solving using computing that scales. The contest focused on end-to-end problem solving using concepts, technologies and architectures that facilitate scaling.
CometCloud is an autonomic framework for enabling real-world applications on dynamically federated, hybrid infrastructure integrating (public & private) clouds, data-centers and Grids.
1. Infrastructures services for dynamic federation and coordination to enable on-demand scale-up, scale-down and scale-out. 2. Programming support to enable a range of programming models and services for autonomic monitoring and management of the infrastructure and applications.
It’s not just a catchy title
The overarching goal of CometCloud is to realize virtual computational cloud infrastructure that integrates local computational environments and public cloud services on-demand, and provide abstractions and mechanisms to support a range of programming paradigms and real-world applications on such an infrastructure. Specifically, CometCloud provides programming abstractions and underlying mechanisms and services. Furthermore, it enables policy-based autonomic cloud-bridging and cloud-bursting. Autonomic cloud-bridging enables on-the-fly integration of local computational environments (data-centers, grids) and public cloud services (such as Amazon EC2), and autonomic cloud-bursting enables dynamic application scale-out to address dynamic workloads, spikes in demands, and other extreme requirements.