How do you design a process historian strategy for a large facility?

Design considerations: data collection rate (balance resolution vs storage - 1 second for control loops, slower for steady values), compression settings (exception-based or compression algorithms), retention periods (raw vs summarized data, regulatory requirements), server architecture (redundancy, scalability), client access (thin client, web), data contextualization (asset model, time in state), and integration with other systems (LIMS, CMMS, ERP). Best practices: consistent tag naming conventions, documented calculations, regular archive maintenance, and disaster recovery planning. Key applications: troubleshooting, performance monitoring, optimization, and compliance reporting.