B1: Carbon-aware Multi-site Workflow Scheduling Under Uncertainty

Continued from B1: Scheduling and Adaptive Execution of Data Analysis Workflows across Heterogeneous Infrastructures

Description

Data centers are significant contributors to carbon emissions, which are highly influenced by a center’s spatiotemporal variability of energy sources and the time of executing a workload. We reduce the carbon footprint of multi-site DAW executions by investigating methods that characterize the energy and carbon emissions of sites over time to create task and machine profiles used for infrastructure-aware and carbon-aware mapping and scheduling under uncertainty.