Federated Computing for the Masses--Aggregating Resources to Tackle Large-Scale Engineering Problems
Issue No. 04 - July-Aug. (2014 vol. 16)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/MCSE.2013.134
Javier Diaz-Montes , Rutgers University
Yu Xie , Iowa State University
Ivan Rodero , Rutgers University
Jaroslaw Zola , Rutgers University
Baskar Ganapathysubramanian , Iowa State University
Manish Parashar , Rutgers University
The complexity of many problems in science and engineering requires computational capacity exceeding what the average user can expect from a single computational center. While many of these problems can be viewed as a set of independent tasks, their collective complexity easily requires millions of core-hours on any high-power computing (HPC) resource, and throughput that can't be sustained by a single, multiuser queuing system. An exploration of the use of aggregated HPC resources to solve large-scale engineering problems shows that it's possible to build a computational federation that's easy for end users to implement, and is elastic, resilient, and scalable. Here, the authors argue that the fusion of federated computing and real-life engineering problems can be brought to the average user if relevant middleware is provided. They report on the use of federation of 10 distributed heterogeneous HPC resources to perform a large-scale interrogation of the parameter space in the microscale fluid flow problem.
Computational modeling, Programming, Complexity theory, Throughput, Big data, Scientific computing, Metasearch
J. Diaz-Montes, Y. Xie, I. Rodero, J. Zola, B. Ganapathysubramanian and M. Parashar, "Federated Computing for the Masses--Aggregating Resources to Tackle Large-Scale Engineering Problems," in Computing in Science & Engineering, vol. 16, no. 4, pp. 62-72, 2014.