A new optimization phase for scientific workflow management systems
A new optimization phase for scientific workflow management systems
Scientific workflows have emerged as an important tool for combining computational power with data analysis for all scientific domains in e-science. They help scientists to design and execute complex in silico experiments. However, with increasing complexity it becomes more and more infeasible to op...
Saved in:
Personal Name(s): | Holl, Sonja (Corresponding Author) |
---|---|
Zimmermann, Olav / Hofmann-Apitius, Martin | |
Contributing Institute: |
Jülich Supercomputing Center; JSC |
Published in: |
2012 IEEE 8th International Conference on E-Science : [Proceedings] - IEEE, 2012. |
Imprint: |
IEEE
2012
|
Physical Description: |
1 - 8 |
DOI: |
10.1109/eScience.2012.6404479 |
Conference: | 2012 IEEE 8th International Conference on E-Science (e-Science), Chicago (IL), 2012-10-08 - 2012-10-12 |
Document Type: |
Contribution to a conference proceedings |
Research Program: |
Computational Science and Mathematical Methods Grid Technologies and Infrastructures |
Publikationsportal JuSER |
Scientific workflows have emerged as an important tool for combining computational power with data analysis for all scientific domains in e-science. They help scientists to design and execute complex in silico experiments. However, with increasing complexity it becomes more and more infeasible to optimize scientific workflows by trial and error. To address this issue, this paper describes the design of a new optimization phase integrated in the established scientific workflow life cycle. We have also developed a flexible optimization application programming interface (API) and have integrated it into a scientific workflow management system. A sample plugin for parameter optimization based on genetic algorithms illustrates, how the API enables rapid implementation of concrete workflow optimization methods. Finally, a use case taken from the area of structural bioinformatics validates how the optimization approach facilitates setup, execution and monitoring of workflow parameter optimization in high performance computing e-science environments. |