update

Author: Binyang2014

.azure-pipelines/templates/nccl-test.yml

@@ -79,8 +79,8 @@ steps:
     name: PyBench
     displayName: Run collective benchmarks
     remoteScript: |
-      mpirun --allow-run-as-root -tag-output -x GPU_MAX_HW_QUEUES=8 -np 8 python3 ./python/mscclpp_benchmark/bench_collective.py --collective allreduce --dtype float8_e4m3b15 --accum-type float32 --autotune
-      mpirun --allow-run-as-root -tag-output -x GPU_MAX_HW_QUEUES=8 -np 8 python3 ./python/mscclpp_benchmark/bench_collective.py --collective allreduce --dtype float16 --symmetric-memory --autotune
+      mpirun --allow-run-as-root -x GPU_MAX_HW_QUEUES=8 -np 8 python3 ./python/mscclpp_benchmark/bench_collective.py --collective allreduce --dtype float8_e4m3b15 --accum-type float32 --autotune
+      mpirun --allow-run-as-root -x GPU_MAX_HW_QUEUES=8 -np 8 python3 ./python/mscclpp_benchmark/bench_collective.py --collective allreduce --dtype float16 --symmetric-memory --autotune
 
 - template: stop.yml
   parameters:

.azure-pipelines/templates/rccl-test.yml

@@ -62,8 +62,8 @@ steps:
     name: PyBench
     displayName: Run collective benchmarks
     remoteScript: |
-      mpirun --allow-run-as-root -tag-output -x GPU_MAX_HW_QUEUES=8 -np 8 python3 ./python/mscclpp_benchmark/bench_collective.py --collective allreduce --dtype float8_e4m3b15 --accum-type float32 --autotune
-      mpirun --allow-run-as-root -tag-output -x GPU_MAX_HW_QUEUES=8 -np 8 python3 ./python/mscclpp_benchmark/bench_collective.py --collective allgather --dtype float8_e4m3b15 --autotune
+      mpirun --allow-run-as-root -x GPU_MAX_HW_QUEUES=8 -np 8 python3 ./python/mscclpp_benchmark/bench_collective.py --collective allreduce --dtype float8_e4m3b15 --accum-type float32 --autotune
+      mpirun --allow-run-as-root -x GPU_MAX_HW_QUEUES=8 -np 8 python3 ./python/mscclpp_benchmark/bench_collective.py --collective allgather --dtype float8_e4m3b15 --autotune
 
 - template: stop.yml
   parameters: