ABSTRACT
We speculate on a novel role virtualization could play in creating a rounded, balanced physical science and engineering software ecosystem to support petascale computational science. The motivation for this analysis is a quest for ways to engage a broader spectrum of expertise in state-of-the-art petascale modeling activities. Current generation petascale efforts are based around massively parallel systems with greater than 50,000 cores. This presents a considerable challenge to typical computational science and engineering application analysis and development practices. To illustrate the challenge we examine an ocean model deployed on a recently commissioned ≈60,000 core parallel system. This case study demonstrates interesting science and engineering challenges that many advanced simulations on petascale systems will face. We then describe a methodology, amenable to virtualization technology at various levels, that mitigates key aspects of the challenges outlined for a potentially broad class of applications.
- P. Apparao, S. Makineni, and D. Newell. Characterization of network processing overheads in xen. In Proceedings of the First International Workshop on Virtualization Technology in Distributed Computing, Tampa, Florida, November 17, 2006, 2006. Google ScholarDigital Library
- R. Bleck and D. B. Boudra. Initial testing of a numerical ocean circulation model using a hybrid quasi-isopycnal vertical coordinate. Journal of Phys. Oceanography, 11: 755--770, 1981.Google ScholarCross Ref
- K. Bryan. A numerical method for the study of the circulation of the world oceans. Journal of Computational Physics, 4:347--376, 1969.Google ScholarCross Ref
- L. Grit, D. Irwin, A. Yumerefendi, and J. Chase. Virtual machine hosting for networked clusters: Building the foundations for autonomic orchestration. In Proceedings of the First International Workshop on Virtualization Technology in Distributed Computing, Tampa, Florida, November 17, 2006, 2006. Google ScholarDigital Library
- C. Hill and J. Marshall. Application of a parallel navier-stokes model to ocean circulation in parallel computational fluid dynamics. In N. Satofuka A. Ecer, J. Periaux and S. Taylor, editors, Implementations and Results Using Parallel Computers, pages 545--552. Elsevier Science B.V.: New York, 1995.Google Scholar
- C. Hill, D. Menemenlis, B. Ciotti, and C. Henze. Investigating solution convergence in a global ocean model using a 2048 processor cluster of distributed shared memory machines. Scientific Programming, 15(2):107--116, 2007. Google ScholarDigital Library
- W. Huang, J. Liu, B. Abali, and D. K. Panda. A case for high performance computing with virtual machines. In Proceedings of ICS 06, Cairns, Queensland, Australia, June 2006, 2006. Google ScholarDigital Library
- W. Huang, M. Koop, Q. Gao, and D. K. Panda. Virtual machine aware communication libraries for high performance computing. In Proceedings of Supercomputing 2007, Reno, Nevada, November, 2007, 2007. Google ScholarDigital Library
- Robert Miller. Numerical Modeling of Ocean Circulation. Cambridge University Press, Cambridge, UK, 2007. ISBN 978-0-521-78182-4.Google ScholarCross Ref
- H. Raj and K. Shwan. High-performance and scalable i/o virtualization via self-virtualized devices. In Proceedings of HPDC 2007, Monterey, California, July 2007, 2007. Google ScholarDigital Library
- Marc Snir and Steve Otto. MPI-The Complete Reference: The MPI Core. MIT Press, Cambridge, MA, USA, 1998. ISBN 0262692155. Google ScholarDigital Library
- The MITgcm Team Web Site. http://mitgcm.org, 2008. The Massachusetts Institute of Technology General Circulation Model.Google Scholar
- The Texas Advanced Computer Center Web Site. http://www.tacc.utexas.edu, 2007. The TACC Sun Constellation System - Ranger.Google Scholar
- L. Youseff, R. Wolski, B. Gorda, and C. Krintz. Evaluating the performance impact of xen on mpi and process execution for hpc systems. In Proceedings of the First International Workshop on Virtualization Technology in Distributed Computing, Tampa, Florida, November 17, 2006, 2006. Google ScholarDigital Library
Index Terms
- "Zen" and the art of petascale ocean modeling: a conceptual analysis of how virtualization could be key to bringing individual science back to petascale ocean modeling
Recommendations
Virtio network paravirtualization driver
One of the techniques used to improve I/O performance of virtual machines is paravirtualization. Paravirtualized devices are intended to reduce the performance overhead on full virtualization where all hardware devices are emulated. The interface of a ...
HyperFresh: Live Refresh of Hypervisors Using Nested Virtualization
APSys '17: Proceedings of the 8th Asia-Pacific Workshop on SystemsBugs in hypervisors are becoming common as hypervisors grow in size and complexity. Latent bugs, such as memory leaks, can lead to hypervisor failures resulting in complete loss of all its virtual machines (or guests). However, reliable operation of ...
Autonomous learning for efficient resource utilization of dynamic VM migration
ICS '08: Proceedings of the 22nd annual international conference on SupercomputingDynamic migration of virtual machines on a cluster of physical machines is designed to maximize resource utilization by balancing loads across the cluster. When the utilization of a physical machine is beyond a fixed threshold, the machine is deemed ...
Comments