Abstract
Computing as a utility has reached the mainstream. Scientists can now easily rent time on large commercial clusters that can be expanded and reduced on-demand in real-time. However, current commercial cloud computing performance falls short of systems specifically designed for scientific applications. Scientific computing needs are quite different from those of the web applications that have been the focus of cloud computing vendors. In this chapter we demonstrate through empirical evaluation the computational efficiency of high-performance numerical applications in a commercial cloud environment when resources are shared under high contention. Using the Linpack benchmark as a case study, we show that cache utilization becomes highly unpredictable and similarly affects computation time. For some problems, not only is it more efficient to underutilize resources, but the solution can be reached sooner in realtime (wall-time). We also show that the smallest, cheapest (64-bit) instance on the studied environment is the best for price to performance ration. In light of the high-contention we witness, we believe that alternative definitions of efficiency for commercial cloud environments should be introduced where strong performance guarantees do not exist. Concepts like average, expected performance and execution time, expected cost to completion, and variance measures–-traditionally ignored in the high-performance computing context–-now should complement or even substitute the standard definitions of efficiency.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Barham, P.T., Dragovic, B., Fraser, K., Hand, S., Harris, T.L., Ho, A., et al. (2003). Xen and the art of virtualization. In Symposium on operating systems principles, (pp. 164–177). New York, NY: Bolton Landing, USA.
Dongarra, J., van de Geijn, R., & Walker, D. (1994). Scalability issues affecting the design of a dense linear algebra library. Journal of Parallel and Distributed Computing, 22(3), 523–537.
Gemignani, C., & Skomoroch, P. (2010). Elasticwulf: Beowulf cluster run on Amazon EC2. Available via the http://WWW. Retrieved 1 January 2010, from http://code.google.com/p/elasticwulf/.
Goto, K. (2010). GotoBLAS. Available from http://WWW. Retrieved 1 January 2010, from http://www.tacc.utexas.edu/.
Hosting, S. D. (2010). GoGrid cloud hosting. Available from http://WWW. Retrieved 1 January 2010, from http://gogrid.com.
Iyer, R., Zhao, L., Guo, F., Illikkal, R., Makineni, S., Newell, D., et al. (2007). Qos policies and architecture for cache/memory in cmp platforms. SIGMETRICS Performance Evaluation Review, 35(1), 25–36. DOI http://doi.acm.org/10.1145/1269899.1254886.
Keahey, K., Freeman, T., Lauret, J., & Olson, D. (2007). Virtual workspaces for scientific applications. SciDAC 2007 Conference, Boston, MA.
Laboratory, A. N. (2010). MPICH2: High-performance and widely portable MPI. Available via the http://WWW. Retrieved 1 January 2010, from http://www.mcs.anl.gov/research/projects/mpich2/.
Napper, J., & Bientinesi, P. (2009). Can cloud computing reach the Top500? Unconventional High-Performance Computing (UCHPC), Italy.
Nimbus Science Clouds (2010). Available from WWW. Retrieved 1 January 2010, from http://www.nimbusproject.org/.
Nurmi, D., Wolski, R., Grzegorczyk, C., Obertelli, G., Soman, S., Youseff, L., et al. (2008). The Eucalyptus open-source cloud-computing system. Proceedings of Cloud Computing and Its Applications [online].
Petitet, A., Whaley, R. C., Dongarra, J., & Cleary, A. (2010). HPL - a portable implementation of the high-performance LINPACK benchmark for distributed-memory computers. Available from http://WWW. Retrieved 1 January 2010, from http://www.netlib.org/benchmark/hpl/.
Petrini, F., Kerbyson, D. J., & Pakin, S. (2003). The case of the missing supercomputer performance: Achieving optimal performance on the 8,192 processors of ASCI Q. SC ’03: Proceedings of the 2003 ACM/IEEE conference on Supercomputing, IEEE Computer Society, Washington, DC, USA, p. 55.
Services, A. W. (2010). Amazon elastic compute cloud (EC2). Available from http://WWW. Retrieved 1 January 2010, from http://aws.amazon.com/ec2.
Strebel, J., Stage, A. “An Economic Decision Model for Business Software Application Deployment on Hybrid Cloud Environments”, In: M. Schumann, L.M. Kolbe, M.H. Breitner, A. Frerichs (eds.), Multikonferenz Wirtschaftsinformatik 2010, Universitätsverlag Göttingen, Ottingen, 2010, pp. 195–206.
Tikotekar, A., Vallée, G., Naughton, T., Ong, H., Engelmann, C., & Scott, S. L. (2008). An analysis of HPC benchmarks in virtual machine environments. Euro-Par 2008 Workshops - Parallel Processing: VHPC 2008, UNICORE 2008, HPPC 2008, SGS 2008, PROPER 2008, ROIA 2008, and DPA 2008, Las Palmas de Gran Canaria, Spain, August 25–26. Revised Selected Papers, 63–71. Springer, Berlin, Heidelberg (2009). DOI http://dx.doi.org/10.1007/978-3-642-00955-6_8
TOP500.Org (2010). Top 500 supercomputer sites. Available from http://WWW. Retrieved 1 January 2010, from http://www.top500.org/.
Walker, E. (2008). Benchmarking Amazon EC2. LOGIN, 18–23.
Wang, G., & Ng, E. (2010). The impact of virtualization on network performance of Amazon EC2 data center. INFOCOM ’10: Proceedings of the 2010 IEEE Conference on Computer Communications. IEEE Communication Society. San Diego, CA.
Xcalibre Communications Ltd (2010). FlexiScale cloud computing. Available from http://WWW. Retrieved 1 January 2010, from http://www.flexiscale.com.
Youseff, L., Seymour, K., You, H., Dongarra, J., & Wolski, R. (2008). The impact of paravirtualized memory hierarchy on linear algebra computational kernels and software. HPDC ’08: Proceedings of the 17th International Symposium on High Performance Distributed Computing, ACM, New York, NY, USA, 141–152. DOI http://doi.acm.org/10.1145/1383422.1383440.
Acknowledgments
The authors wish to acknowledge the Aachen Institute for Advanced Study in Computational Engineering Science (AICES) as sponsor of the experimental component of this research. Financial support from the Deutsche Forschungsgemeinschaft (German Research Association) through grant GSC 111 is gratefully acknowledged. Also, support from the XtreemOS project, which is partially funded by the European Commission under contract #FP6-033576 is gratefully acknowledged.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2010 Springer Science+Business Media, LLC
About this chapter
Cite this chapter
Bientinesi, P., Iakymchuk, R., Napper, J. (2010). HPC on Competitive Cloud Resources. In: Furht, B., Escalante, A. (eds) Handbook of Cloud Computing. Springer, Boston, MA. https://doi.org/10.1007/978-1-4419-6524-0_21
Download citation
DOI: https://doi.org/10.1007/978-1-4419-6524-0_21
Published:
Publisher Name: Springer, Boston, MA
Print ISBN: 978-1-4419-6523-3
Online ISBN: 978-1-4419-6524-0
eBook Packages: Computer ScienceComputer Science (R0)