You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository was archived by the owner on Jan 12, 2024. It is now read-only.
Copy file name to clipboardExpand all lines: AdvantageBenchmark/readme.md
+1-1Lines changed: 1 addition & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -6,7 +6,7 @@ This benchmark is intended to provide an easy way to verify the performance char
6
6
7
7
## Executing the benchmark
8
8
9
-
To execute the benchmark, compile each version of advantage.sln using `dotnet build .\advantage.sln -c Release` from their respective folders. Then the executable to run will be either `bin\Release\netcoreapp3.1\host.exe` in the privateBuild folder or `host\bin\Release\netcoreapp3.1\host.exe` in the releaseBuild folder. This executable takes parameters describing which test circuits to execute and how many loops to perform as integer arguments, such that `host.exe 1 1 5` will run 5 loops of test 1 and `host.exe 0 3 100` will run 100 loops of tests 0 through 3. Check the contents of `privateBuild\Program.cs` to see the tests that correspond to each identifier; for most machines, test 1 aka advantage 4x4 circuit is the best choice for benchmarking.
9
+
To execute the benchmark, compile each version of advantage.sln using `dotnet build .\advantage.sln -c Release` from their respective folders. Then the executable to run will be either `bin\Release\net6.0\host.exe` in the privateBuild folder or `host\bin\Release\net6.0\host.exe` in the releaseBuild folder. This executable takes parameters describing which test circuits to execute and how many loops to perform as integer arguments, such that `host.exe 1 1 5` will run 5 loops of test 1 and `host.exe 0 3 100` will run 100 loops of tests 0 through 3. Check the contents of `privateBuild\Program.cs` to see the tests that correspond to each identifier; for most machines, test 1 aka advantage 4x4 circuit is the best choice for benchmarking.
10
10
11
11
The benchmark can also be run via runTest.ps1 or runTest.sh, which performs a sweep across configured environment variables that adjust the number of threads used and gates fused in simulating the circuit. See the definition of the script used on your platform to understand how it configures the `OMP_NUM_THREADS` and `QDK_SIM_FUSESPAN` environment variables.
0 commit comments