The cost of doing science on the cloud: the Montage example

ABSTRACT

Utility grids such as the Amazon EC2 cloud and Amazon S3 offer computational and storage resources that can be used on-demand for a fee by compute and data-intensive applications. The cost of running an application on such a cloud depends on the compute, storage and communication resources it will provision and consume. Different execution plans of the same application may result in significantly different costs. Using the Amazon cloud fee structure and a real-life astronomy application, we study via simulation the cost performance tradeoffs of different execution and resource provisioning plans. We also study these trade-offs in the context of the storage and communication fees of Amazon S3 when used for long-term application data archival. Our results show that by provisioning the right amount of storage and compute resources, cost can be significantly reduced with no significant impact on application performance.

REFERENCES

• 1: "Open Science Grid,"www.opensciencegrid.org.
• 2: "TeraGrid,"http://www.teragrid.org/.
• 3: "Enabling Grids for E-sciencE (EGEE),"http://www.euegee. org/.
• 4: "TeraGrid Resource Reservations," www.teragrid.org/userinfo/resource_reservation.php
• 5: "Special Priority and Urgent Computing Environment," http://spruce.teragrid.org.
• 6: George Lawton, Moving the OS to the Web, Computer, v.41 n.3, p.16-19, March 2008 [doi>10.1109/MC.2008.94]
• 7: "Amazon Web Services," http://aws.amazon.com,http://aws.amazon.com.
• 8: "Google App Engine,"http://code.google.com/appengine/.
• 9: "Montage Project," http://montage.ipac.caltech.edu.
• 10: The Two Micron All Sky Survey., "http://www.ipac.caltech.edu/2mass," -+*.
• 11: Sloan Digital Sky Survey., "http://www.sdss.org/."
• 12: "Image Mosaic Service," http://hachi.ipac.caltech.edu:8080/montage/.
• 13: E. Deelman, G. Mehta, et al., "Pegasus: Mapping Large- Scale Workflows to Distributed Resources," in Workflows in e-Science, I. Taylor, E. Deelman, et al., Eds.: Springer, 2006.
• 14: "REST vs SOAP at Amazon," http://www.oreillynet.com/pub/wlg/3005?wlg=yes.
• 15: Arun Ramakrishnan , Gurmeet Singh , Henan Zhao , Ewa Deelman , Rizos Sakellariou , Karan Vahi , Kent Blackburn , David Meyers , Michael Samidi, Scheduling Data-IntensiveWorkflows onto Storage-Constrained Distributed Resources, Proceedings of the Seventh IEEE International Symposium on Cluster Computing and the Grid, p.401-409, May 14-17, 2007 [doi>10.1109/CCGRID.2007.101]
• 16: Gurmeet Singh , Karan Vahi , Arun Ramakrishnan , Gaurang Mehta , Ewa Deelman , Henan Zhao , Rizos Sakellariou , Kent Blackburn , Duncan Brown , Stephen Fairhurst , David Meyers , G. Bruce Berriman , John Good , Daniel S. Katz, Optimizing workflow data footprint, Scientific Programming, v.15 n.4, p.249-268, December 2007
• 17: R. Buyya and M. Murshed, "GridSim: A Toolkit for the Modeling and Simulation of Distributed Resource Management and Scheduling for Grid Computing," Concurrency and Computation: Practice and Experience, vol. 14, pp. 1175-1220, 2002.
• 18: R. Buyya, D. Abramson, et al., "The grid economy," Proceedings of the IEEE, vol. 93, pp. 698-714, 2005.
• 19: Carl A. Waldspurger , Tad Hogg , Bernardo A. Huberman , Jeffrey O. Kephart , W. Scott Stornetta, Spawn: A Distributed Computational Economy, IEEE Transactions on Software Engineering, v.18 n.2, p.103-117, February 1992 [doi>10.1109/32.121753]
• 20: K. Lai, B. A. Huberman, et al., "Tycoon: a Distributed Market-based Resource Allocation System," Technical Report, Hewlett-Packard Laboratories, Palo Alto, CA, September2004.
• 21: "IBM Blue Cloud," http://www- 03.ibm.com/press/us/en/pressrelease/22613.wss.
• 22: "Nirvanix," http://www.nirvanix.com.
• 23: M. Palankar, A. Onibokun, et al., "Amazon S3 for Science Grids: a Viable Solution," in 4th USENIX Symposium on Networked Systems Design & Implementation (NSDI'07), 2007.
• 24: Gurmeet Singh , Carl Kesselman , Ewa Deelman, A provisioning model and its comparison with best-effort for performance-cost optimization in grids, Proceedings of the 16th international symposium on High performance distributed computing, June 25-29, 2007, Monterey, California, USA [doi>10.1145/1272366.1272382]
• 25: H. Zhao and R. Sakellariou, "Advance Reservation Policies for Workflows," in 12th Workshop on Job Scheduling Strategies for Parallel Processing (JSSPP), Saint-Malo, France, 2006.
• 26: G. Singh, C. Kesselman, et al., "Performance Impact of Resource Provisioning on Workflows," USC http://www.cs.usc.edu/Research/TechReports/05- 850.pdf 05-850, 2005.
• 27: Jia Yu , Rajkumar Buyya , Chen Khong Tham, Cost-Based Scheduling of Scientific Workflow Application on Utility Grids, Proceedings of the First International Conference on e-Science and Grid Computing, p.140-147, December 05-08, 2005 [doi>10.1109/E-SCIENCE.2005.26]
• 28: Richard Huang , Henri Casanova , Andrew A. Chien, Automatic resource specification generation for resource selection, Proceedings of the 2007 ACM/IEEE conference on Supercomputing, November 10-16, 2007, Reno, Nevada [doi>10.1145/1362622.1362638]
• 29: Walfredo Cirne , Francine Berman, Using moldability to improve the performance of supercomputer jobs, Journal of Parallel and Distributed Computing, v.62 n.10, p.1571-1601, October 2002 [doi>10.1006/jpdc.2002.1869]
• 30: Allen B. Downey, Using Queue Time Predictions for Processor Allocation, Proceedings of the Job Scheduling Strategies for Parallel Processing, p.35-57, April 05, 1997
• 31: G. Singh, C. Kesselman, et al., "Adaptive Pricing for Resource Reservations in Shared Environments," Grid 2007.
• 32: Anthony Sulistio , Kyong Hoon Kim , Rajkumar Buyya, Using Revenue Management to Determine Pricing of Reservations, Proceedings of the Third IEEE International Conference on e-Science and Grid Computing, p.396-405, December 10-13, 2007 [doi>10.1109/E-SCIENCE.2007.83]
• 33: "Nimbus Science Cloud," http://workspace.globus.org/clouds/nimbus.html.
• 34: "Amazon SLA," http://www.amazon.com/gp/browse.html?node=379654 011.

INDEX TERMS

Primary Classification:
  C. Computer Systems Organization
  C.2 COMPUTER-COMMUNICATION NETWORKS
      C.2.4 Distributed Systems

Additional Classification:
  C. Computer Systems Organization
  C.4 PERFORMANCE OF SYSTEMS

  J. Computer Applications
  J.2 PHYSICAL SCIENCES AND ENGINEERING

General Terms:
Design, Experimentation, Measurement, Performance