Simulation-Driven Portfolio Scheduling for Scientific Workflows

• João Antonio Soares UFPEL
• Gerson Geraldo H. Cavalheiro UFPEL

DOI: https://doi.org/10.5753/eradrs.2025.6801

Abstract

This paper proposes portfolio scheduling approach for scientific workflow management systems (WMS). The goal is to improve scheduling decisions by dynamically selecting suitable scheduling policies during workflow execution. The portfolio integrates multiple task scheduling and resource allocation strategies, considering both computational and network characteristics.

References

Casanova, H., Wong, Y. C., Pottier, L., and da Silva, R. F. (2023). On the feasibility of simulation-driven portfolio scheduling for cyberinfrastructure runtime systems. In Klusáček, D., Julita, C., and Rodrigo, G. P., editors, Job Scheduling Strategies for Parallel Processing, pages 3–24, Cham. Springer Nature Switzerland.

Deng, K., Song, J., Ren, K., and Iosup, A. (2013). Exploring portfolio scheduling for long-term execution of scientific workloads in iaas clouds. In Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis, SC ’13, New York, NY, USA. Association for Computing Machinery.

McDonald, J., Dobbs, J., Wong, Y. C., Ferreira da Silva, R., and Casanova, H. (2024). An exploration of online-simulation-driven portfolio scheduling in workflow management systems. Future Generation Computer Systems, 161:345–360.