ABSTRACT
Virtual machine (VM) technologies are experiencing a resurgence in both industry and research communities. VMs offer many desirable features such as security, ease of management, OS customization, performance isolation, check-pointing, and migration, which can be very beneficial to the performance and the manageability of high performance computing (HPC) applications. However, very few HPC applications are currently running in a virtualized environment due to the performance overhead of virtualization. Further, using VMs for HPC also introduces additional challenges such as management and distribution of OS images.In this paper we present a case for HPC with virtual machines by introducing a framework which addresses the performance and management overhead associated with VM-based computing. Two key ideas in our design are: Virtual Machine Monitor (VMM) bypass I/O and scalable VM image management. VMM-bypass I/O achieves high communication performance for VMs by exploiting the OS-bypass feature of modern high speed interconnects such as Infini-Band. Scalable VM image management significantly reduces the overhead of distributing and managing VMs in large scale clusters. Our current implementation is based on the Xen VM environment and InfiniBand. However, many of our ideas are readily applicable to other VM environments and high speed interconnects.We carry out detailed analysis on the performance and management overhead of our VM-based HPC framework. Our evaluation shows that HPC applications can achieve almost the same performance as those running in a native, non-virtualized environment. Therefore, our approach holds promise to bring the benefits of VMs to HPC applications with very little degradation in performance.
- A. Chien et al. Design and Evaluation of an HPVM-Based Windows NT Supercomputer. The International Journal of High Performance Computing Applications, 13(3):201--219, Fall 1999. Google ScholarDigital Library
- A. Menon et al. Diagnosing Performance Overheads in the Xen Virtual Machine Environment. In Proceedings of the First ACM/USENIX Conference on Virtual Execution Environments (VEE'05), June 2005. Google ScholarDigital Library
- Argonne National Laboratory. http://www-unix.mcs.anl.gov/mpi/mpich/.Google Scholar
- Argonne National Laboratory. Zeptoos: The small linux for big computers. http://www-unix.mcs.anl.gov/zeptoos/.Google Scholar
- A. Awadallah and M. Rosenblum. The vMatrix: A network of virtual machine monitors for dynamic content distribution. In Seventh International Workshop on Web Content Caching and Distribution, 2002.Google Scholar
- B. Dragovic et al. Xen and the Art of Virtualization. In Proceedings of the ACM Symposium on Operating Systems Principles, Oct. 2003. Google ScholarDigital Library
- R. Brightwell and L. A. Fisk. Scalable Parallel Application Launch on Cplant. In Proceedings of SC '01, 2001. Google ScholarDigital Library
- B. Chun and D. Culler. User-centric Performance Analysis of Market-based Cluster Batch Schedulers. In Proceedings of CCGrid, 2002. Google ScholarDigital Library
- R. J. Creasy. The Origin of the VM/370 Time-sharing System. IBM Journal of Research and Development, 25(5):483--490, 1981.Google ScholarDigital Library
- FastOS: Forum to Address Scalable Technology for runtime and Operating Systems. http://www.cs.unm.edu/fastos/.Google Scholar
- R. Figueiredo, P. Dinda, and J. Fortes. A Case for Grid Computing on Virtual Machines. In Proceedings of International Conference on Distributed Computing Systems (ICDCS), May 2003., 2003. Google ScholarDigital Library
- K. Fraser, S. Hand, R. Neugebauer, I. Pratt, A. W. and M. Williamson. Safe Hardware Access with the Xen Virtual Machine Monitor. In Proceedings of OASIS ASPLOS Workshop, 2004.Google Scholar
- A. Geist, A. Beguelin, J. Dongarra, W. Jiang, R. Manchek, and V. S. Sunderam. PVM: Parallel Virtual Machine: A Users' Guide and Tutorial for Networked Parallel Computing. MIT Press, Cambridge, MA, USA, 1994. Google ScholarDigital Library
- HPC-Colony Project: Services and Interfaces for Very Large Linux Clusters. http://www.hpc-colony.org/.Google Scholar
- IETF IPoIB Workgroup. http://www.ietf.org/html.charters/ipoib-charter.html.Google Scholar
- InfiniBand Trade Association. InfiniBand Architecture Specification, Release 1.2.Google Scholar
- K. Koch. How does ASCI Actually Complete Multi-month 1000-processor Milestone Simulations? In Proceedings of the Conference on High Speed Computing, 2002.Google Scholar
- J. Liu, W. Huang, B. Abali, and D. K. Panda. High Performance VMM-Bypass I/O in Virtual Machines. In Proceedings of USENIX '06, Boston, MA, 2006. Google ScholarDigital Library
- J. Liu, J. Wu, S. P. Kini, P. Wyckoff, and D. K. Panda. High Performance RDMA-Based MPI Implementation over InfiniBand. In Proceedings of ICS '03, June 2003. Google ScholarDigital Library
- B. Maccabe, P. G. Bridges, R. Brightwell, R. Riesen, and T. Hudson. Highly Configurable Operating Systems for Ultrascale Systems. In Proceedings of the First International Workshop on Operating Systems, Programming Environments and Management Tools for High-Performance Computing on Clusters, 2004.Google Scholar
- Mellanox Technologies. http://www.mellanox.com.Google Scholar
- MOLAR: Modular Linux and Adaptive Runtime Support for High-end Computing Operating and Runtime Systems. http://forge-fre.ornl.gov/molar/.Google Scholar
- MVAPICH Project Website. http://nowlab.cse.ohio-state.edu/projects/mpi-iba/index.html.Google Scholar
- Myricom, Inc. Myrinet. http://www.myri.com.Google Scholar
- NASA. NAS Parallel Benchmarks. http://www.nas.nasa.gov/Software/NPB/.Google Scholar
- Open InfiniBand Alliance. http://www.openib.org.Google Scholar
- OProfile. http://oprofile.sourceforge.net.Google Scholar
- A. Petitet, R. C. Whaley, J. Dongarra, and A. Cleary. HPL - A Portable Implementation of the High-Performance Linpack Benchmark for Distributed-Memory Computers. http://www.netlib.org/benchmark/hpl/.Google Scholar
- F. Petrini, D. J. Kerbyson, and S. Pakin. The Case of the Missing Supercomputer Performance: Achieving Optimal Performance on the 8,192 Processors of ASCI Q. In Proceedings of SC '03, Washington, DC, USA, 2003. Google ScholarDigital Library
- I. Pratt. Xen Virtualization. Linux World 2005 Virtualization BOF Presentation.Google Scholar
- Quadrics, Ltd. QsNet. http://www.quadrics.com.Google Scholar
- D. Reed, I. Pratt, P. Menage, S. Early, and N. Stratford. Xenoservers: Accountable Execution of Untrusted Programs. In Workshop on Hot Topics in Operating Systems, 1999. Google ScholarDigital Library
- M. Rosenblum and T. Garfinkel. Virtual Machine Monitors: Current Technology and Future Trends. IEEE Computer, 38(5):39--47, 2005. Google ScholarDigital Library
- J. Sugerman, G. Venkitachalam, and B. H. Lim. Virtualizing I/O Devices on VMware Workstation's Hosted Virtual Machine Monitor. In Proceedings of USENIX, 2001. Google ScholarDigital Library
- Top 500 Supercomputer Site. http://www.top500.com.Google Scholar
- ttylinux. http://www.minimalinux.org/.Google Scholar
- University of Wisconsin. Condor High Throughput Computing. http://www.cs.wisc.edu/condor/.Google Scholar
- T. von Eicken, A. Basu, V. Buch, and W. Vogels. U-Net: A User-level Network Interface for Parallel and Distributed Computing. In ACM Symposium on Operating Systems Principles, 1995. Google ScholarDigital Library
- T. von Eicken, D. E. Culler, S. C. Goldstein, and K. E. Schauser. Active Messages: A Mechanism for Integrated Communication and Computation. In International Symposium on Computer Architecture, 1992. Google ScholarDigital Library
- C. Waldspurger. Memory resource management in VMware ESX server. In Proceedings of the Fifth Symposium on Operating Systems Design and Implementation, 2002. Google ScholarDigital Library
- A. Whitaker, M. Shaw, and S. D. Gribble. Denali: Lightweight Virtual Machines for Distributed and Networked Applications. Technical report, University of Washington, February 2002.Google Scholar
- W. Yu, J. Wu, and D. K. Panda. Fast and Scalable Startup of MPI Programs In InfiniBand Clusters. In Proceedings of HiPC'04, Banglore, Inida, December 2004. Google ScholarDigital Library
Index Terms
- A case for high performance computing with virtual machines
Recommendations
A high performance inter-domain communication approach for virtual machines
Highlights Presented a novel inter-domain communication approach (IVCOM) for virtual machine. Using bypassing protocol stacks, shunning page flipping methods. IVCOM applies a direct communication channel between domain 0 and U. IVOCM can improve the ...
Performance profiling of virtual machines
VEE '11Profilers based on hardware performance counters are indispensable for performance debugging of complex software systems. All modern processors feature hardware performance counters, but current virtual machine monitors (VMMs) do not properly expose ...
Performance profiling of virtual machines
VEE '11: Proceedings of the 7th ACM SIGPLAN/SIGOPS international conference on Virtual execution environmentsProfilers based on hardware performance counters are indispensable for performance debugging of complex software systems. All modern processors feature hardware performance counters, but current virtual machine monitors (VMMs) do not properly expose ...
Comments