Modular performance prediction for scientific workflows using Machine Learning

Singh, Alok; Purawat, Shweta; Rao, Arvind; Altintas, Ilkay

doi:10.1016/j.future.2020.04.048

Modular performance prediction for scientific workflows using Machine Learning

Journal Article · Mon May 11 00:00:00 EDT 2020 · Future Generations Computer Systems

DOI:https://doi.org/10.1016/j.future.2020.04.048· OSTI ID:1851724

Singh, Alok ^[1]; Purawat, Shweta ^[2]; Rao, Arvind ^[2]; Altintas, Ilkay ^[2]

Univ. of California, San Diego, La Jolla, CA (United States). San Diego Supercomputer Center; Univ. of California, San Diego, La Jolla, CA (United States)
Univ. of California, San Diego, La Jolla, CA (United States). San Diego Supercomputer Center

Scientific workflows provide an opportunity for declarative computational experiment design in an intuitive and efficient way. A distributed workflow is typically executed on a variety of resources, and it uses a variety of computational algorithms or tools to achieve the desired outcomes. Such a variety imposes additional complexity in scheduling these workflows on large scale computers. As computation becomes more distributed, insights into expected workload that a workflow presents become critical for effective resource allocation. In this paper, we present a modular framework that leverages Machine Learning for creating precise performance predictions of a workflow. The central idea is to partition a workflow in such a way that makes the task of forecasting each atomic unit manageable and gives us a way to combine the individual predictions efficiently. We recognize a combination of an executable and a specific physical resource as a single module. This gives us a handle to characterize workload and machine power as a single unit of prediction. Overall, our modular technique of creating atomic modules and deployment of longest-path approach to estimate workflow performance, allows the framework to adapt to highly complex nested directed acyclic workflows and scale to new scenarios, since it does not make assumptions of underlying workflow structure. We present performance estimation results of independent workflow modules executed on the XSEDE SDSC Comet cluster using various Machine Learning algorithms. The results provide insights into the behavior and effectiveness of different algorithms in the context of scientific workflow performance prediction.

View Accepted Manuscript (DOE)

Research Organization:: Univ. of California, San Diego, CA (United States)

Sponsoring Organization:: National Institutes of Health (NIH); National Science Foundation (NSF); USDOE; USDOE Office of Science (SC)

Grant/Contract Number:: SC0012630

OSTI ID:: 1851724

Alternate ID(s):: OSTI ID: 1776457

Journal Information:: Future Generations Computer Systems, Journal Name: Future Generations Computer Systems Journal Issue: C Vol. 114; ISSN 0167-739X

Publisher:: ElsevierCopyright Statement

Country of Publication:: United States

Language:: English

References (21)

Scientific workflow management and the Kepler system Ludäscher, Bertram; Altintas, Ilkay; Berkley, Chad Concurrency and Computation: Practice and Experience, Vol. 18, Issue 10 https://doi.org/10.1002/cpe.994	journal	January 2006
Support-vector networks Cortes, Corinna; Vapnik, Vladimir Machine Learning, Vol. 20, Issue 3 https://doi.org/10.1007/BF00994018	journal	September 1995
A multi-strategy collaborative prediction model for the runtime of online tasks in computing cluster/grid Tao, Ming; Dong, Shoubin; Zhang, Liping Cluster Computing, Vol. 14, Issue 2 https://doi.org/10.1007/s10586-010-0145-4	journal	October 2010
Milepost GCC: Machine Learning Enabled Self-tuning Compiler Fursin, Grigori; Kashnikov, Yuriy; Memon, Abdul Wahid International Journal of Parallel Programming, Vol. 39, Issue 3 https://doi.org/10.1007/s10766-010-0161-2	journal	January 2011
Workflows and e-Science: An overview of workflow system features and capabilities Deelman, Ewa; Gannon, Dennis; Shields, Matthew Future Generation Computer Systems, Vol. 25, Issue 5 https://doi.org/10.1016/j.future.2008.06.012	journal	May 2009
Characterizing and profiling scientific workflows Juve, Gideon; Chervenak, Ann; Deelman, Ewa Future Generation Computer Systems, Vol. 29, Issue 3 https://doi.org/10.1016/j.future.2012.08.015	journal	March 2013
Biomedical Big Data Training Collaborative (BBDTC): An effort to bridge the talent gap in biomedical science and research Purawat, Shweta; Cowart, Charles; Amaro, Rommie E. Journal of Computational Science, Vol. 20 https://doi.org/10.1016/j.jocs.2017.03.010	journal	May 2017
A novel statistical time-series pattern based interval forecasting strategy for activity durations in workflow systems Liu, Xiao; Ni, Zhiwei; Yuan, Dong Journal of Systems and Software, Vol. 84, Issue 3 https://doi.org/10.1016/j.jss.2010.11.927	journal	March 2011
Kepler + CometCloud: Dynamic Scientific Workflow Execution on Federated Cloud Resources Wang, Jianwu; AbdelBaky, Moustafa; Diaz-Montes, Javier Procedia Computer Science, Vol. 80 https://doi.org/10.1016/j.procs.2016.05.363	journal	January 2016
Random Forests Breiman, Leo Machine Learning, Vol. 45, Issue 1, p. 5-32 https://doi.org/10.1023/A:1010933404324	journal	January 2001
Gene Selection for Cancer Classification using Support Vector Machines Guyon, Isabelle; Weston, Jason; Barnhill, Stephen Machine Learning, Vol. 46, Issue 1/3, p. 389-422 https://doi.org/10.1023/A:1012487302797	journal	January 2002
A tutorial on support vector regression Smola, Alex J.; Schölkopf, Bernhard Statistics and Computing, Vol. 14, Issue 3 https://doi.org/10.1023/B:STCO.0000035301.49549.88	journal	August 2004
Predicting the Execution Time of Workflow Activities Based on Their Input Features Miu, Tudor; Missier, Paolo 2012 SC Companion: High Performance Computing, Networking, Storage and Analysis (SCC), 2012 SC Companion: High Performance Computing, Networking Storage and Analysis https://doi.org/10.1109/SC.Companion.2012.21	conference	November 2012
On Performance Modeling and Prediction in Support of Scientific Workflow Optimization Wu, Qishi; Datla, Vivek V. 2011 IEEE World Congress on Services (SERVICES) https://doi.org/10.1109/SERVICES.2011.37	conference	July 2011
A regression-based approach to scalability prediction Barnes, Bradley J.; Rountree, Barry; Lowenthal, David K. Proceedings of the 22nd annual international conference on Supercomputing - ICS '08 https://doi.org/10.1145/1375527.1375580	conference	January 2008
Challenges and approaches for distributed workflow-driven analysis of large-scale biological data: vision paper Altintas, Ilkay; Wang, Jianwu; Crawl, Daniel Proceedings of the 2012 Joint EDBT/ICDT Workshops on - EDBT-ICDT '12 https://doi.org/10.1145/2320765.2320791	conference	January 2012
Analysis of benchmark characteristics and benchmark performance prediction Saavedra, Rafael H.; Smith, Alan J. ACM Transactions on Computer Systems, Vol. 14, Issue 4 https://doi.org/10.1145/235543.235545	journal	November 1996
Large memory high performance computing enables comparison across human gut microbiome of patients with autoimmune diseases and healthy subjects Wu, Sitao; Li, Weizhong; Smarr, Larry XSEDE '13: Extreme Science and Engineering Discovery Environment: Gateway to Discovery, Proceedings of the Conference on Extreme Science and Engineering Discovery Environment: Gateway to Discovery https://doi.org/10.1145/2484762.2484828	conference	July 2013
Prophesy: an infrastructure for performance analysis and modeling of parallel and grid applications Taylor, Valerie; Wu, Xingfu; Stevens, Rick ACM SIGMETRICS Performance Evaluation Review, Vol. 30, Issue 4 https://doi.org/10.1145/773056.773060	journal	March 2003
The future of scientific workflows Deelman, Ewa; Peterka, Tom; Altintas, Ilkay The International Journal of High Performance Computing Applications, Vol. 32, Issue 1 https://doi.org/10.1177/1094342017704893	journal	April 2017
machine. Friedman, Jerome H. The Annals of Statistics, Vol. 29, Issue 5 https://doi.org/10.1214/aos/1013203451	journal	October 2001

Similar Records

iDDS: intelligent distributed dispatch and scheduling for workflow orchestration

Journal Article · Fri Jan 23 19:00:00 EST 2026 · European Physical Journal. C, Particles and Fields (Online) · OSTI ID:3017617

Asynchronous Execution of Heterogeneous Tasks in ML-Driven HPC Workflows

Conference · Fri May 19 00:00:00 EDT 2023 · OSTI ID:2333668

Integration of scanning probe microscope with high-performance computing: Fixed-policy and reward-driven workflows implementation

Journal Article · Sun Sep 15 20:00:00 EDT 2024 · Review of Scientific Instruments · OSTI ID:2571043

Related Subjects

97 MATHEMATICS AND COMPUTING
Computer Science
Distributed computing
Exascale computing
Machine learning
Parallel computing
Performance prediction
Scientific workflows

Modular performance prediction for scientific workflows using Machine Learning

Citation Formats

References (21)

Similar Records

Related Subjects