Jungle computing
Jungle computing refers to the use of diverse[vague], distributed and highly non-uniform[vague] high performance computer systems to achieve peak performance.[1][2]
The increasing complexity of the high performance computing environment has provided a bewildering range of choices beside traditional supercomputers and clusters. Scientists can now use grid and cloud infrastructures, in a variety of combinations along with traditional supercomputers - all connected via fast networks. And the emergence of many-core technologies such as GPUs, as well as supercomputers on chip within these environments has added to the complexity. Thus high performance computing can now use multiple diverse platforms and systems simultaneously, giving rise to the term "computing jungle".[1]
See also
References
- ^ a b Jason Maassen, et al Towards jungle computing with Ibis/Constellation in Proceedings of the 2011 workshop on Dynamic distributed data-intensive applications, programming abstractions, and systems, ACM New York, ISBN 978-1-4503-0705-5 [1]
- ^ Jungle Computing: Distributed Supercomputing Beyond Clusters, Grids, and Clouds by Frank Seinstra et al in "Grids, Clouds and Virtualization, Computer Communications and Networks", ISBN 978-0-85729-048-9. Springer-Verlag London Limited, 2011, p. 167 [2]