Go backward to Traces Go up to Top Go forward to Profile and Counts

Summary of Data Collection

1. Use profile and count information to obtain parameter values needed to complete performance models.
2. Measure execution times for a range of processor counts and problem sizes, and compare results with predictions.
3. If observed and modeled performance differ, use profile and count information to verify basic assumptions of model.
   • Message counts and volumes, load imbalances, replicated computation).
4. If aspects are still unexplained, use tracing and study performance on a few processors. Increase processor number as understanding improves.