Analyze, View, & Verify CPU & GPU Hardware Bottlenecks

This quick video shows how you can evaluate workload performance across the CPU and GPU with the Graphics Trace Analyzer, part of the Intel® Graphics Performance Analyzers (Intel® GPA) tool. See how to use the timeline view to identify CPU-GPU contention, and examine the GPU adapter queue and CPU context queue to further analyze whether the application is CPU or GPU bound.

Giselle Gomez, Technical Consulting Engineer, Intel Corporation

Giselle is a Technical Consulting Engineer in the Intel Visual Computing Products group, and works with the Intel® Graphics Performance Analyzers (Intel® GPA) team, ensuring customers have the learning materials to use the tool to its fullest potential. She joined Intel in 2016, starting as an intern supporting the Intel® XDK tool, and began her current role in 2017, shortly after graduating from Washington State University where she earned a BS in Computer Science.

For more complete information about compiler optimizations, see our Optimization Notice.