Changes

Jump to: navigation, search

GPU621/VTuners

No change in size, 21:44, 5 December 2022
Parallelism
|}
[[File:Vtune Roadmap.png|500px400px|frame]]
[[File:Effective-gpu.png|500px400px|frame]]
Spin and Overhead Time
enables us to investigate the concurrency problems in the application and time-dependent the performance of each thread. In the figure below in the lower half part of the window is the timeline view. As shown in brown colour which indicates the CPU time. Not until ~12 second, the mater thread was split into 8 threads and the first five were off-loaded, while the last threes (TID: 14500, 16268, 28576) were waiting (shown in light green colour) and the last two even waited all the way end which weakened parallelism. When brown band (CPU Time) concurrently happened to multiple threads, it means high level of parallelism.
[[File:Effective-CPU-Utilization-Histogram.png|500px400px|frame]]
== Platform and I/O ==
117
edits

Navigation menu