FiLiP: A File Lifecycle-based Profiler for hierarchical storage

The increasing gap between computing speed and storage latency leads to possible I/O bottlenecks on massively parallel computers. To mitigate this issue, hierarchical provides multi-tiered configurations where each tier has its own physical characteristics associated performance. Selecting the most appropriate file placement policy is difficult there our knowledge no tool that systematically statistics metrics for optimal selection. In paper, we present FiLiP (File Lifecycle Profiler), a software which better understanding of access by applications consequences movements across storage. After description FiLiP’s main features architecture, highlight usefulness using three intensive simulation HPC applications: NEMO, S3DIO NAMD three-tiered burst buffer.