In order to bridge the enormous gap between I/O and CPU speeds, a new model of in-situ workflows is emerging as a promising solution which means performing on-line analysis while and where the data is generated. Though, in-situ workflows are promising it also impose many runtime challenges. In order to avoid these runtime challenges, the scientist often end-up in over-provisioning of the resources required for their experiments which impacts cost, energy consumption and availability of resources for other jobs. Thus, there is a need for an autonomous system that can actively or pro-actively assess the workflow state frequently, i.e., pressure of system resources, and take appropriate actions to adapt the workflow to dynamic resource requirements and use minimal resources. In this presentation, I will talk about the work in progress on understanding the memory pressure of the workflow to achieve the above goal.
Slides not available for download yet
Swati Singhal is a third year Ph.D. student in Computer Science at the University of Maryland advised by Professor Alan Sussman. Her research interest focus on high performance parallel and distributed computing. Before moving to United States, she worked as a Software Engineer in High Performance Computing team at IBM Research India. She holds a masters in Computer Science from the University of Delhi, India.
Back to Top