Integrating the Perlmutter HPC system in the ALICE Grid

The Perlmutter HPC system is the 9th generation supercomputer deployed at the National Energy Research Scientific Computing Center (NERSC). It provides both CPU and GPU resources, offering 393216 AMD EPYC Milan cores with 4 GB of memory per core, for CPU-oriented jobs, and 7168 NVIDIA A100 GPUs. Due to the specific highly parallel and massive CPU and memory requirements of the native payloads running on supercomputers, there is always an idle part of the computing capacity. Conversely, Grid payloads require few CPU cores for a single task and can take advantage of the idle resources. This ‘backfill’ is advantageous both for the supercomputer operators, increasing the overall use efficiency of the machine and for the Grid users, allowing them to opportunistically use a substantial amount of CPUs. ALICE takes advantage of these conditions, the architecture of the Perlmutter supercomputer, and utilities offered by NERSC. This is achieved by deploying a standard Grid interface to Perlmutter via the NERSC SuperFacility API scheduling tool to submit and monitor ALICE Grid payloads. Perlmutter has been integrated into the ALICE Grid and it currently runs Monte Carlo simulation jobs. R&D continues to include analysis jobs by connecting to an EOS instance hosted at LBNL Tier 2 site. Integrated into the ALICE Grid, Perlmutter has proven to be a reliable resource contributor to ALICE, providing on average 8000 cores in 2024. This paper describes the path taken to integrate Perlmutter into the ALICE Grid and the modifications needed to integrate HPC facilities into the standard Grid infrastructure.