CPU Profiler – AMD uProf CPU profiler follows a statistical sampling-based approach to collect profile data to identify the performance bottlenecks in an application. The profile data collection can be triggered by – OS timer, core PMC events, and IBS. AMD uProf offers user-friendly UI to view and analyze the profile data, thereby helps to optimize wide variety of applications, drivers, game engines, and so on.

OS Tracing – The OS and runtime libraries can be traced along with CPU sampling-based profiles to provide timeline views to analyze what is happening in the system when an application is running. Events can be collected and analyzed: OS scheduling event, System calls, POSIX threads library’s (pthread) thread synchronization APIs, block I/O calls, page fault, and memtrace.

MPI Tracing – MPI trace analysis can be used to analyze, compute, and message passing load imbalance among the ranks of a MPI application running on a cluster. The profile reports are generated for Point-to-Point and Collective API activity summary.

GPU Profiler – Provides GPU performance events-based kernel execution statistics.

GPU Tracing – Traces the HIP, ROCr, and KFD runtime APIs and GPU activities when a HIP based application is executing.

AMD uProf’ s ANALYZE page to view the Function Hotspots function level

AMD uProf’ s SOURCE page to view the performance data corelated to source lines

AMD uProf’ s Flame Graph window to identify the hot call-paths

AMD uProf’s System Analysis Timeline View to identify performance bottlenecks

AMD uProf’s MPI Communication Matrix to analyze message passing load imbalance