A Performance Model to Estimate Execution Time of Scientific Workflows on the Cloud

Scientific workflows, which capture large computational problems, may be executed on large-scale distributed systems such as Clouds. Determining the amount of resources to be provisioned for the execution of scientific workflows is a key component to achieve cost-efficient resource management and good performance. In this paper, a performance prediction model is presented to estimate execution time of scientific workflows for a different number of resources, taking into account their structure as well as their system-dependent characteristics. In the evaluation, three real-world scientific workflows are used to compare the estimated makespan calculated by the model with the actual makespan achieved on different system configurations of Amazon EC2. The results show that the proposed model can predict execution time with an error of less than 20% for over 96.8% of the experiments..

[1]  Paul Watson,et al.  A framework for dynamically generating predictive models of workflow execution , 2013, WORKS@SC.

[2]  Xingfu Wu,et al.  Using kernel couplings to predict parallel application performance , 2002, Proceedings 11th IEEE International Symposium on High Performance Distributed Computing.

[3]  Robert D. van der Mei,et al.  Effective Prediction of Job Processing Times in a Large-Scale Grid Environment , 2006, 2006 15th IEEE International Conference on High Performance Distributed Computing.

[4]  Rizos Sakellariou,et al.  Imbalance optimization in scientific workflows , 2013, ICS '13.

[5]  Daniel S. Katz,et al.  A comparison of two methods for building astronomical image mosaics on a grid , 2005, 2005 International Conference on Parallel Processing Workshops (ICPPW'05).

[6]  Dan Walsh,et al.  Design and implementation of the Sun network filesystem , 1985, USENIX Conference Proceedings.

[7]  G. Bruce Berriman,et al.  Data Sharing Options for Scientific Workflows on Amazon EC2 , 2010, 2010 ACM/IEEE International Conference for High Performance Computing, Networking, Storage and Analysis.

[8]  Mary K. Vernon,et al.  Poems: end-to-end performance design of large parallel adaptive computational systems , 1998, WOSP '98.

[9]  Dennis Gannon,et al.  Workflows for e-Science, Scientific Workflows for Grids , 2014 .

[10]  José A. B. Fortes,et al.  On the Use of Machine Learning to Predict the Time and Resources Consumed by Applications , 2010, 2010 10th IEEE/ACM International Conference on Cluster, Cloud and Grid Computing.

[11]  Johan Montagnat,et al.  A Probabilistic Model to Analyse Workflow Performance on Production Grids , 2008, 2008 Eighth IEEE International Symposium on Cluster Computing and the Grid (CCGRID).

[12]  Thomas Fahringer,et al.  Using Templates to Predict Execution Time of Scientific Workflow Applications in the Grid , 2009, 2009 9th IEEE/ACM International Symposium on Cluster Computing and the Grid.

[13]  Robert B. Ross,et al.  PVFS: A Parallel File System for Linux Clusters , 2000, Annual Linux Showcase & Conference.

[14]  Katarzyna Keahey,et al.  Contextualization: Providing One-Click Virtual Clusters , 2008, 2008 IEEE Fourth International Conference on eScience.

[15]  Ewa Deelman,et al.  Fault Tolerant Clustering in Scientific Workflows , 2012, 2012 IEEE Eighth World Congress on Services.

[16]  Ewa Deelman,et al.  The application of cloud computing to scientific workflows: a study of cost and performance , 2013, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences.

[17]  Rizos Sakellariou,et al.  Balanced Task Clustering in Scientific Workflows , 2013, 2013 IEEE 9th International Conference on e-Science.

[18]  Bin Li,et al.  Fair Benchmarking for Cloud Computing systems , 2012, Journal of Cloud Computing: Advances, Systems and Applications.

[19]  Thomas Fahringer,et al.  Predicting the execution time of grid workflow applications through local learning , 2009, Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis.

[20]  Paolo Missier,et al.  Predicting the Execution Time of Workflow Activities Based on Their Input Features , 2012, 2012 SC Companion: High Performance Computing, Networking Storage and Analysis.

[21]  Marta Mattoso,et al.  Towards a Cost Model for Scheduling Scientific Workflows Activities in Cloud Environments , 2011, 2011 IEEE World Congress on Services.

[22]  Liping Zhang,et al.  A multi-strategy collaborative prediction model for the runtime of online tasks in computing cluster/grid , 2010, Cluster Computing.

[23]  Dan Tsafrir,et al.  Backfilling Using System-Generated Predictions Rather than User Runtime Estimates , 2007, IEEE Transactions on Parallel and Distributed Systems.

[24]  Shishir Bharathi,et al.  Data Staging Strategies and Their Impact on the Execution of Scientific Workflows , 2009, DADC '09.

[25]  Daniel S. Katz,et al.  Pegasus: A framework for mapping complex scientific workflows onto distributed systems , 2005, Sci. Program..

[26]  Warren Smith,et al.  Predicting Application Run Times Using Historical Information , 1998, JSSPP.

[27]  Marco Aurélio Amaral Henriques,et al.  Contention-sensitive static performance prediction for parallel distributed applications , 2006, Perform. Evaluation.

[28]  Daniel S. Katz,et al.  Workflow task clustering for best effort systems with Pegasus , 2008, Mardi Gras Conference.

[29]  Douglas Thain,et al.  Toward fine-grained online task characteristics estimation in scientific workflows , 2013, WORKS@SC.

[30]  Dan Tsafrir,et al.  Modeling User Runtime Estimates , 2005, JSSPP.

[31]  Ann L. Chervenak,et al.  Characterizing and profiling scientific workflows , 2013, Future Gener. Comput. Syst..

[32]  Yuan-Chun Jiang,et al.  A novel statistical time-series pattern based interval forecasting strategy for activity durations in workflow systems , 2011, J. Syst. Softw..