Disaggregation-Aware Scheduler

A simulation framework for job scheduling in HPC systems with disaggregated memory, developed with LBNL.

A simulation framework for exploring job scheduling and memory-allocation strategies in HPC systems with disaggregated memory resources, developed during my internships at Lawrence Berkeley National Laboratory.

The simulator underpins results published at CLUSTER’24 and IPDPSW’24, and is used in ongoing collaborations on memory-disaggregated system design.

Source: github.com/Artlands/DisaggregationAwareScheduler