Speaker
Description
Abstract: The transition to the HL-LHC era brings unprecedented computing demands and a rapidly shifting hardware landscape. Since its successful deployment in 2023, HEPScore has become the standard CPU benchmark for WLCG sites, replacing the legacy HEP-SPEC06. The journey to HEPScore was a major collaborative effort, involving software developers, data analysts, site personel, and the WLCG Deployment Task Force, to create a representative benchmark based entirely on real-world HEP applications.
Today, as GPU applications and GPU availability over the Grid rise, the community requires a comparable standard to measure raw performance and assess the cost-benefit ratio of GPU resources. Building on the proven methodology of HEPScore23, the HEPiX Benchmarking Working Group is intensifying the development of a GPU equivalent, again utilizing representative HEP GPU workloads as payloads.
This next phase introduces complex new challenges, including heterogeneous architectures, strong workload-specific optimizations, and a fast-evolving hardware and software ecosystem. This contribution reviews the collaborative journey and technological solutions that made CPU HEPScore23 a reality, analyzes the novel challenges of GPU standardization, and presents our strategy and current development status for establishing a robust GPU benchmark for the WLCG.
Speaker Bio: Dr. Robin Hofsaess is a Fellow within the LCG section at CERN, where his work focuses on the development and evolution of HEPScore, WLCG's benchmarking framework. He joined CERN in early 2025 after completing his Ph.D. in Physics at the Karlsruhe Institute of Technology (KIT). During his doctoral research with the CMS experiment and at GridKa, the German WLCG Tier-1 center, his work centered on investigating optimizations for the integration of HPC centers into the WLCG. Today, Dr. Hofsaess leverages his expertise in distributed computing and performance optimization to help the HEP community navigate the transition toward heterogeneous architectures and GPU benchmarking for the HL-LHC era.