On the Feasibility of Simulation-Driven Portfolio Scheduling for Cyberinfrastructure Runtime Systems

Casanova, Henri; Wong, Yick Ching; Pottier, Loic; Ferreira Da Silva, Rafael

doi:10.1007/978-3-031-22698-4_1

On the Feasibility of Simulation-Driven Portfolio Scheduling for Cyberinfrastructure Runtime Systems

Conference · 01 January 2023

DOI:https://doi.org/10.1007/978-3-031-22698-4_1· OSTI ID:1922323

Casanova, Henri ^[1]; Wong, Yick Ching ^[1]; Pottier, Loic ^[2]; ^[3]

University of Hawaii at Manoa, Honolulu
University of Southern California, Information Sciences Institute
ORNL

Runtime systems that automate the execution of applications on distributed cyberinfrastructures need to make scheduling decisions. Researchers have proposed many scheduling algorithms, but most of them are designed based on analytical models and assumptions that may not hold in practice. The literature is thus rife with algorithms that have been evaluated only within the scope of their underlying assumptions but whose practical effectiveness is unclear. It is thus difficult for developers to decide which algorithm to implement in their runtime systems.To obviate the above difficulty, we propose an approach by which the runtime system executes, throughout application execution, simulations of this very execution. Each simulation is for a different algorithm in a scheduling algorithm portfolio, and the best algorithm is selected based on simulation results. The main objective of this work is to evaluate the feasibility and potential merit of this portfolio scheduling approach, even in the presence of simulation inaccuracy, when compared to the traditional one-algorithm approach. We perform this evaluation via a case study in the context of scientific workflows. Our main finding is that portfolio scheduling can outperform the best one-algorithm approach even in the presence of relatively large simulation inaccuracies.

View Conference

Research Organization:: Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)

Sponsoring Organization:: USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR)

DOE Contract Number:: AC05-00OR22725

OSTI ID:: 1922323

Resource Relation:: Journal Volume: 13592; Conference: 25th Workshop on Job Scheduling Strategies for Parallel Processing (JSSPP 2022) - Lyon, , France - 6/3/2022 4:00:00 AM-6/3/2022 4:00:00 AM

Country of Publication:: United States

Language:: English

References (24)

A Survey on Scheduling Strategies for Workflows in Cloud Environment and Emerging Trends No authors listed ACM Computing Surveys https://doi.org/10.1145/3325097		August 2019
Validity of the single processor approach to achieving large scale computing capabilities No authors listed Proceedings of the April 18-20, 1967, spring joint computer conference on - AFIPS '67 (Spring) https://doi.org/10.1145/1465482.1465560		January 1967
Workflow scheduling algorithms in cloud environment - A survey No authors listed 2014 Recent Advances in Engineering and Computational Sciences (RAECS) https://doi.org/10.1109/RAECS.2014.6799514		March 2014
GridSim: a toolkit for the modeling and simulation of distributed resource management and scheduling for Grid computing No authors listed Concurrency and Computation: Practice and Experience https://doi.org/10.1002/cpe.710		November 2002
CloudSim: a toolkit for modeling and simulation of cloud computing environments and evaluation of resource provisioning algorithms No authors listed Software: Practice and Experience https://doi.org/10.1002/spe.995		August 2010
Obtaining dynamic scheduling policies with simulation and machine learning No authors listed Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis on - SC '17 https://doi.org/10.1145/3126908.3126955		January 2017
Versatile, scalable, and accurate simulation of distributed applications and platforms No authors listed Journal of Parallel and Distributed Computing https://doi.org/10.1016/j.jpdc.2014.06.008		October 2014
Developing accurate and scalable simulators of production workflow management systems with WRENCH No authors listed Future Generation Computer Systems https://doi.org/10.1016/j.future.2020.05.030		November 2020
WfCommons: A framework for enabling scientific workflow research and development No authors listed Future Generation Computer Systems https://doi.org/10.1016/j.future.2021.09.043		March 2022
Exploring portfolio scheduling for long-term execution of scientific workloads in IaaS clouds No authors listed Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis https://doi.org/10.1145/2503210.2503244		November 2013
Self-tuning systems No authors listed IEEE Software https://doi.org/10.1109/52.754053		January 1999
Online Tuning of EASY-Backfilling using Queue Reordering Policies No authors listed IEEE Transactions on Parallel and Distributed Systems https://doi.org/10.1109/TPDS.2018.2820699		October 2018
Workflow scheduling in heterogeneous computing systems : A survey No authors listed 2017 International Conference on Computing and Communication Technologies for Smart Nation (IC3TSN) https://doi.org/10.1109/IC3TSN.2017.8284499		October 2017
LogGOPSim: simulating large-scale applications in the LogGOPS model No authors listed Proceedings of the 19th ACM International Symposium on High Performance Distributed Computing - HPDC '10 https://doi.org/10.1145/1851476.1851564		January 2010
DISSECT-CF: A simulator to foster energy-aware scheduling in infrastructure clouds No authors listed Simulation Modelling Practice and Theory https://doi.org/10.1016/j.simpat.2015.05.009		November 2015
Fostering Energy-Awareness in Simulations behind Scientific Workflow Management Systems No authors listed 2014 IEEE/ACM 7th International Conference on Utility and Cloud Computing https://doi.org/10.1109/UCC.2014.11		December 2014
A Survey of Data-Intensive Scientific Workflow Management No authors listed Journal of Grid Computing https://doi.org/10.1007/s10723-015-9329-8		March 2015
CloudNetSim++: A toolkit for data center simulations in OMNET++ No authors listed 2014 11th Annual High Capacity Optical Networks and Emerging/Enabling Technologies (Photonics for Energy) https://doi.org/10.1109/HONET.2014.7029371		December 2014
FogNetSim++: A Toolkit for Modeling and Simulation of Distributed Fog Environment No authors listed IEEE Access https://doi.org/10.1109/ACCESS.2018.2877696		January 2018
A taxonomy and survey on scheduling algorithms for scientific workflows in IaaS cloud computing environments: Workflow Scheduling Algorithms for Clouds No authors listed Concurrency and Computation: Practice and Experience https://doi.org/10.1002/cpe.4041		December 2016
The self-tuning dynP job-scheduler No authors listed Proceedings 16th International Parallel and Distributed Processing Symposium https://doi.org/10.1109/IPDPS.2002.1015662		January 2002
Portfolio-Based Selection of Robust Dynamic Loop Scheduling Algorithms Using Machine Learning No authors listed 2014 IEEE International Parallel & Distributed Processing Symposium Workshops https://doi.org/10.1109/IPDPSW.2014.183		May 2014
PSINS: An Open Source Event Tracer and Execution Simulator for MPI Applications No authors listed Lecture Notes in Computer Science https://doi.org/10.1007/978-3-642-03869-3_16		January 2009
On the validity of flow-level tcp network models for grid and cloud simulations No authors listed ACM Transactions on Modeling and Computer Simulation https://doi.org/10.1145/2517448		October 2013

Similar Records

An exploration of online-simulation-driven portfolio scheduling in Workflow Management Systems

Journal Article · 2024 · Future Generations Computer Systems · OSTI ID:2426051

McDonald, Jesse; Dobbs, John; Wong, Yick Ching; +2 more

Machine Learning Based Online Performance Prediction for Runtime Parallelization and Task Scheduling

Conference · 2008 · OSTI ID:951680

Li, J; Ma, X; Singh, K; +3 more

Runtime Scheduling Policies for Distributed Graph Algorithms

Conference · 2018 · OSTI ID:1526308

Firoz, Jesun S.; Zalewski, Marcin J.; Lumsdaine, Andrew; +1 more

On the Feasibility of Simulation-Driven Portfolio Scheduling for Cyberinfrastructure Runtime Systems

Citation Formats

References (24)

Similar Records

Related Subjects