Skip to content

Commit 403857f

Browse files
authored
chore: remove h200-trt runner type (#185)
* remove h200-trt runner type * bug fix * undo changes in plot perf * debug summarize py * argv[2] not even used in summarize.py -- removing from workflow * fix race condition in e2e-tests.yml
1 parent a2cf0b0 commit 403857f

8 files changed

Lines changed: 9 additions & 20 deletions

File tree

.github/README.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -96,7 +96,7 @@ full-sweep --precision fp8 --seq-lens 1k8k --config-files .github/configs/nvidia
9696

9797
**Test all TRT configs on H200 runners:**
9898
```
99-
full-sweep --framework trt --runner-type h200 h200-trt --config-files .github/configs/nvidia-master.yaml --runner-config .github/configs/runners.yaml
99+
full-sweep --framework trt --runner-type h200 b200-trt --config-files .github/configs/nvidia-master.yaml --runner-config .github/configs/runners.yaml
100100
```
101101

102102
**Quick smoke test of all configs (highest TP, lowest concurrency only):**

.github/configs/CONFIGS.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -49,4 +49,4 @@ Notes:
4949

5050
## Runners
5151

52-
The `runners.yaml` config represents the available runners in the repository. The keys are the runner *types* (i.e., the GPUs as well as some specific combinations like `h200-trt`) whereas the value is a list of *runner nodes*. This config is used to verify the master configs.
52+
The `runners.yaml` config represents the available runners in the repository. The keys are the runner *types* (i.e., the GPUs as well as some specific combinations like `b200-trt`) whereas the value is a list of *runner nodes*. This config is used to verify the master configs.

.github/configs/nvidia-master.yaml

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -144,7 +144,7 @@ dsr1-fp8-h200-trt:
144144
image: nvcr.io#nvidia/tensorrt-llm/release:1.1.0rc2.post2
145145
model: deepseek-ai/DeepSeek-R1-0528
146146
model-prefix: dsr1
147-
runner: h200-trt
147+
runner: h200
148148
precision: fp8
149149
framework: trt
150150
# For all sequence lengths, EP=TP
@@ -258,7 +258,7 @@ gptoss-fp4-h200-trt:
258258
image: nvcr.io#nvidia/tensorrt-llm/release:gpt-oss-dev
259259
model: openai/gpt-oss-120b
260260
model-prefix: gptoss
261-
runner: h200-trt
261+
runner: h200
262262
precision: fp4
263263
framework: trt
264264
# For all sequence lengths, EP=TP, DP_ATTENTION=false

.github/configs/runners.yaml

Lines changed: 0 additions & 11 deletions
Original file line numberDiff line numberDiff line change
@@ -14,17 +14,6 @@ h200:
1414
- 'h200-nv_1'
1515
- 'h200-nv_2'
1616
- 'h200-nv_3'
17-
h200-trt:
18-
- 'h200-cw_0'
19-
- 'h200-cw_1'
20-
- 'h200-nb_0'
21-
- 'h200-nb_1'
22-
- 'h200-nb_2'
23-
- 'h200-nb_3'
24-
- 'h200-nv_0'
25-
- 'h200-nv_1'
26-
- 'h200-nv_2'
27-
- 'h200-nv_3'
2817
b200-trt:
2918
- 'b200-nv_0'
3019
- 'b200-nv_1'

.github/workflows/collect-results.yml

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -26,7 +26,7 @@ jobs:
2626
pattern: ${{ inputs.exp-name && format('{0}_*', inputs.exp-name) || '*' }}
2727

2828
- name: Print summary
29-
run: python3 utils/summarize.py results/ ${{ inputs.exp-name || 'all' }} >> $GITHUB_STEP_SUMMARY
29+
run: python3 utils/summarize.py results/ >> $GITHUB_STEP_SUMMARY
3030

3131
- name: Aggregate results
3232
run: python3 utils/collect_results.py results/ ${{ inputs.exp-name || 'all' }}

.github/workflows/e2e-tests.yml

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -55,7 +55,7 @@ jobs:
5555
secrets: inherit
5656

5757
calc-success-rate:
58-
needs: test-sweep
58+
needs: collect-results
5959
if: ${{ always() }}
6060
runs-on: ubuntu-latest
6161

.github/workflows/full-sweep-test.yml

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -56,7 +56,7 @@ jobs:
5656
5757
set -x
5858
# Build runner type filters based on inputs
59-
RUNNER_TYPES="${{ inputs.use_h100 && 'h100' || '' }} ${{ inputs.use_h200 && 'h200' || '' }} ${{ inputs.use_h200 && 'h200 h200-trt' || '' }} ${{ inputs.use_b200 && 'b200 b200-trt' || '' }} ${{ inputs.use_mi300x && 'mi300x' || '' }} ${{ inputs.use_mi325x && 'mi325x' || '' }} ${{ inputs.use_mi355x && 'mi355x' || '' }}"
59+
RUNNER_TYPES="${{ inputs.use_h100 && 'h100' || '' }} ${{ inputs.use_h200 && 'h200' || '' }} ${{ inputs.use_b200 && 'b200 b200-trt' || '' }} ${{ inputs.use_mi300x && 'mi300x' || '' }} ${{ inputs.use_mi325x && 'mi325x' || '' }} ${{ inputs.use_mi355x && 'mi355x' || '' }}"
6060
6161
# DSR1 doesn't support H100, so exclude it
6262
DSR1_RUNNER_TYPES=$(echo $RUNNER_TYPES | sed 's/\bh100\b//g' | xargs)

utils/matrix-logic/generate_sweep_configs.py

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -824,7 +824,7 @@ def main():
824824
test_config_parser.add_argument(
825825
'--runner-type',
826826
required=True,
827-
help='Runner type (e.g., h200-trt, h100)'
827+
help='Runner type (e.g., b200-trt, h100)'
828828
)
829829
test_config_parser.add_argument(
830830
'--runner-config',
@@ -847,7 +847,7 @@ def main():
847847
test_config_parser.add_argument(
848848
'--runner-type',
849849
required=True,
850-
help='Runner type (e.g., h200-trt, h100)'
850+
help='Runner type (e.g., b200-trt, h100)'
851851
)
852852
test_config_parser.add_argument(
853853
'--model-prefix',

0 commit comments

Comments
 (0)