Dynamic Resource Reallocation and Process Malleability
Study of the integration of malleability techniques in HPC applications.
Management of Virtualized GPUs in Cloud Environments
Exploring new ways of using GPU Scheduling as a Service (GSaaS) in computational containers and virtual machines.
Scientific applications run on supercomputers where thousands of nodes are shared among users. When those applications start, their resources remain allocated until the job ends. We have detected two potential approaches in resource managing, with which we increase the global throughput and provide a better utilization of the underlying resources.