Events as a Basis for Workflow Scheduling
Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
This paper evaluates using event-based scheduling as a basis for dynamic workflow management. To do this, the WorkflowRunner is introduced as a tool for conducting event driven scheduling in a robust manner. It is evaluated in comparison to Slurm and the WorkflowRunner is found to schedule analysis ~2.5x quicker, in the majority of use cases. An example workflow is also presented, demonstrating how this style of scheduling allows for complete modification of the workflow structure at runtime, something very difficult to achieve in traditional workflow management systems. These developments are expected to be of particular use to dynamic scheduling systems, as well as for distributed, or heterogeneous systems such as those looking to accommodate human-in-the-loop interactions.
Original language | English |
---|---|
Title of host publication | Proceedings of WORKS 2022 : 17th Workshop on Workflows in Support of Large-Scale Science, Held in conjunction with SC 2022: The International Conference for High Performance Computing, Networking, Storage and Analysis |
Number of pages | 8 |
Publisher | IEEE |
Publication date | 2022 |
Pages | 52-59 |
ISBN (Electronic) | 9781665451918 |
DOIs | |
Publication status | Published - 2022 |
Event | 17th IEEE/ACM Workshop on Workflows in Support of Large-Scale Science, WORKS 2022 - Dallas, United States Duration: 13 Nov 2022 → 18 Nov 2022 |
Conference
Conference | 17th IEEE/ACM Workshop on Workflows in Support of Large-Scale Science, WORKS 2022 |
---|---|
Land | United States |
By | Dallas |
Periode | 13/11/2022 → 18/11/2022 |
Bibliographical note
Publisher Copyright:
© 2022 IEEE.
- Dynamic, Heterogeneous, MEOW, Workflows
Research areas
ID: 344725898