Speaker
Description
High-performance data management systems are foundational to modern scientific facilities, particularly in high-energy physics (HEP) and nuclear physics (NP) where experiments generate massive datasets. The Large Hadron Collider produces 5 petabytes daily, while the High-Luminosity LHC upgrade will require 10× greater capacity by 2030. Individual experiments document their solutions, yet systematic cross-institutional comparative analysis enabling evidence-based infrastructure planning remains absent. We address this gap by developing an integrated analytical framework that combines workflow patterns and data acquisition models with a comprehensive technology taxonomy spanning data transfer, management, orchestration, optimization, authentication, monitoring and validation layers. We apply this methodology across nine leading HEP and NP facilities within WLCG, ESnet and GÉANT, analyzing technology deployments that handle over 500 petabytes annually. We also identify three emergent architectural patterns: data lake federations for reducing replication costs, software-defined networking for time-critical workflows, and FAIR-compliant metadata systems for cross-experiment integration. Our framework enables facilities to map their specific requirements to validated technology combinations, providing concrete design guidance for HL-LHC-era infrastructure planning based on workflow characteristics and resource constraints.