Visible to Intel only — GUID: GUID-3E5EFD34-2A77-48FF-9870-70A53126521F
Visible to Intel only — GUID: GUID-3E5EFD34-2A77-48FF-9870-70A53126521F
Window
This is the benchmark for measuring the overhead of an MPI_Win_create/MPI_Win_fence/MPI_Win_free combination. In the case of an unused window, a negligible non-trivial action is performed inside the window. It minimizes optimization effects of the MPI implementation.
The MPI_Win_fence function is called to properly initialize an access epoch. This is a correction as compared to earlier releases of the Intel(R) MPI Benchmarks.
See the basic definitions and a schematic view of the pattern below.
Window Definition
Property |
Description |
---|---|
Measured pattern |
MPI _Win_create/MPI_Win_ fence/MPI_Win_free |
Reported timings |
t=Δt(M) (in μsec) as indicated in the figure below. |
Reported throughput |
None |
Window Pattern
![](/content/dam/docs/us/en/user-guide-benchmarks/2021-8/6180D50F-C960-4A38-B971-7FEA214B800F-low.png)