Speaker
Description
The distributed computing system of the ATLAS experiment at the Large Hadron Collider (LHC) uses resources from several EuroHPC facilities through both allocated and opportunistic access. HyperQueue, a meta-scheduler developed at IT4Innovations, the Czech National Supercomputing Center, enables the experiment's workload to be adapted to the many-core architecture typical of modern HPC systems. This contribution highlights two HyperQueue use cases. The first involves running ATLAS jobs on LUMI, whose environment differs significantly from that of a standard grid site. We successfully obtained an allocation on LUMI - one of the most powerful European supercomputers, located in Finland - and used it for ATLAS production. The second use case concerns executing a mix of jobs with widely varying resource requirements. While running a single type of workload is simpler, such tasks may become scarce. The ability to run and combine diverse workloads greatly increases the number of available jobs and, consequently, improves ATLAS utilization.