Skip to content
CI

nsys-jax: roll back to 2024.4.1 and handle single-gpu profiles of single module executions better #3197

Sign in to view logs

nsys-jax: roll back to 2024.4.1 and handle single-gpu profiles of single module executions better

nsys-jax: roll back to 2024.4.1 and handle single-gpu profiles of single module executions better #3197

Run time

Learn about OS pricing on GitHub Actions
Job Run time
metadata
1s
bump-manifest
18s
arm64 / test-distribution (extra-only-distribution.sh)
7s
amd64 / test-distribution (extra-only-distribution.sh)
5s
arm64 / test-distribution (mirror-only-distribution.sh)
9s
amd64 / test-distribution (mirror-only-distribution.sh)
9s
arm64 / test-distribution (upstream-only-distribution.sh)
5s
amd64 / test-distribution (upstream-only-distribution.sh)
5s
arm64 / test-distribution (local-patch-distribution.sh)
6s
amd64 / test-distribution (local-patch-distribution.sh)
10s
arm64 / build-base / build-base
7m 8s
amd64 / build-base / build-base
4m 22s
amd64 / build-jax / build-jax
12m 47s
arm64 / build-jax / build-jax
23m 37s
amd64 / build-equinox / build-equinox
3m 10s
amd64 / build-levanter / build-levanter
4m 30s
amd64 / build-grok / build-grok
3m 8s
amd64 / build-gemma / build-gemma
5m 13s
amd64 / build-triton / build-triton
10m 52s
amd64 / build-maxtext / build-maxtext
5m 44s
amd64 / build-upstream-pax / build-upstream-pax
6m 13s
amd64 / build-upstream-t5x / build-upstream-t5x
6m 22s
amd64 / test-jax / jax-V100-unit-test
21m 3s
amd64 / test-jax / jax-A100-unit-test
14m 17s
amd64 / test-pallas / pallas-V100-unit-test
2m 25s
amd64 / test-pallas / pallas-A100-unit-test
4m 50s
amd64 / test-jax / runner / launch-slurm-runner
2h 44m 17s
amd64 / test-pallas / runner / launch-slurm-runner
2h 43m 34s
amd64 / test-levanter / levanter-V100-unit-test
19m 22s
amd64 / test-levanter / levanter-A100-unit-test
15m 17s
amd64 / test-levanter / runner / launch-slurm-runner
2h 54m 28s
amd64 / test-gemma / gemma-V100-unit-test
3m 55s
amd64 / test-gemma / gemma-A100-unit-test
3m 58s
amd64 / test-gemma / runner / launch-slurm-runner
2h 43m 22s
amd64 / test-maxtext / single-process-multi-device (1, 1, 2, 4)
42m 13s
amd64 / test-maxtext / maxtext-multinode (1, 1, 1, 1)
39m 13s
amd64 / test-maxtext / maxtext-multinode (1, 1, 8, 1)
43m 34s
amd64 / test-maxtext / maxtext-multinode (1, 1, 1, 8)
40m 48s
amd64 / test-maxtext / maxtext-multinode (1, 1, 4, 2)
47m 15s
amd64 / test-maxtext / maxtext-multinode (1, 2, 2, 2)
45m 30s
amd64 / test-maxtext / maxtext-multinode (1, 4, 2, 2)
2h 58m 8s
amd64 / build-rosetta-pax / build-rosetta
6m 54s
amd64 / test-upstream-pax / pax-multi-node (1DP1FSDP1TP1PP, 1, 1, 1, 1, 4)
2h 51m 34s
amd64 / test-upstream-pax / pax-multi-node (8DP1FSDP1TP1PP, 1, 8, 1, 1, 4)
2h 48m 1s
amd64 / test-upstream-pax / single-process-evaluation (1, 8, 1, 1)
2h 55m 40s
amd64 / test-upstream-pax / pax-multi-node (1DP8FSDP1TP1PP, 1, 1, 8, 1, 4)
3h 0m 59s
amd64 / test-upstream-pax / pax-multi-node (2DP1FSDP1TP4PP, 4, 2, 1, 1, 4)
2h 53m 19s
amd64 / test-upstream-pax / pax-multi-node (4DP1FSDP2TP1PP, 1, 4, 1, 2, 4)
3h 17m 56s
amd64 / test-upstream-pax / pax-multi-node (16DP1FSDP1TP1PP, 1, 16, 1, 1, 4)
3h 44m 38s
amd64 / test-upstream-pax / pax-multi-node (2DP1FSDP2TP4PP, 4, 2, 1, 2, 4)
3h 41m 2s
amd64 / test-upstream-pax / single-process-multi-device (1, 8, 1, 1)
3h 14m 21s
amd64 / test-upstream-pax / pax-multi-node (LLaMA_eval, 1, 1, 8, 1, 4, true, --model-type LLaMA70BProxy --evaluate)
3h 10m 47s
amd64 / test-te / te-V100-unit-test
52m 30s
amd64 / test-upstream-pax / single-process-multi-device (1, 1, 2, 4)
3h 5m 18s
amd64 / test-te / te-A100-unit-test
40m 11s
amd64 / test-te / runner / launch-slurm-runner
3h 36m 17s
amd64 / test-te-multigpu / te-multi-gpu (2) / te-2GPU
1h 7m 20s
amd64 / test-te-multigpu / te-multi-gpu (8) / te-8GPU
1h 3m 47s
amd64 / test-te-multigpu / te-multi-gpu (4) / te-4GPU
53m 17s
amd64 / build-rosetta-t5x / build-rosetta
7m 38s
amd64 / test-upstream-t5x / t5x-multi-node (8G2N, 8, 2)
3h 53m 59s
amd64 / test-upstream-t5x / t5x-multi-gpu (1P8G, 8)
3h 22m 28s
amd64 / test-upstream-t5x / t5x-multi-node (8G2N_fmha, 8, 2, --enable-fmha 1)
3h 50m 1s
amd64 / test-upstream-t5x / t5x-multi-gpu (1P2G_fmha, 2, --enable-fmha 1)
3h 27m 26s
amd64 / test-triton / triton-V100-unit-test
4m 21s
amd64 / test-triton / triton-A100-unit-test
4m 30s
amd64 / test-triton / runner / launch-slurm-runner
3h 52m 14s
arm64 / build-triton
0s
arm64 / build-gemma
0s
arm64 / test-jax
0s
arm64 / test-pallas
0s
arm64 / build-maxtext / build-maxtext
17m 3s
arm64 / build-equinox / build-equinox
3m 56s
arm64 / build-upstream-t5x / build-upstream-t5x
17m 43s
arm64 / build-levanter / build-levanter
5m 58s
arm64 / build-upstream-pax / build-upstream-pax
25m 28s
arm64 / build-grok / build-grok
4m 2s
arm64 / test-triton
0s
arm64 / test-gemma
0s
amd64 / test-rosetta-pax / rosetta-pax-multi-node-te (1DP1FSDP1TP1PP_TE, 1, 1, 1, 1, 4)
4h 16m 36s
amd64 / test-rosetta-pax / rosetta-pax-multi-node (1, 8, 1, 1)
3h 59m 7s
amd64 / test-rosetta-pax / rosetta-pax-multi-node-te (8DP1FSDP1TP1PP_TE, 1, 8, 1, 1, 4)
4h 12m 8s
amd64 / test-rosetta-pax / rosetta-pax-multi-node (1, 4, 1, 2)
3h 59m 26s
amd64 / test-rosetta-pax / rosetta-pax-multi-node-te (1DP8FSDP1TP1PP_TE, 1, 1, 8, 1, 4)
4h 5m 15s
amd64 / test-rosetta-pax / rosetta-pax-single-node-dropout-te (1, 8, 1, 1)
3h 50m 43s
amd64 / test-rosetta-pax / rosetta-pax-multi-node-te (4DP1FSDP2TP1PP_TE, 1, 4, 1, 2, 4)
4h 6m 21s
amd64 / test-rosetta-pax / rosetta-pax-multi-node (4, 2, 1, 1)
4h 0m 28s
amd64 / test-rosetta-pax / rosetta-pax-multi-node-te (16DP1FSDP1TP1PP_TE, 1, 16, 1, 1, 4)
4h 9m 3s
amd64 / test-rosetta-pax / rosetta-pax-multi-node (4, 2, 1, 2)
3h 56m 15s
amd64 / test-rosetta-pax / rosetta-pax-multi-node-te (5B_fused_attn_1, 1, 1, 8, 1, 2, --model-type 5B)
4h 1m 6s
amd64 / test-rosetta-pax / single-process-multi-device-te (1, 8, 1, 1)
4h 3m 34s
amd64 / test-rosetta-pax / rosetta-pax-multi-node-te (5B_fused_attn_0, 1, 1, 8, 1, 2, --model-type 5B --disable-fused-attn)
4h 2m 33s
amd64 / test-rosetta-pax / single-process-multi-device-te (1, 1, 2, 4)
3h 54m 18s
amd64 / test-rosetta-pax / rosetta-pax-multi-node-te (LLaMA_eval_TE, 1, 1, 8, 1, 4, true, --model-type LLaMA70BProxy --evalu...
4h 13m 55s
amd64 / test-rosetta-pax / single-process-evaluation-te (1, 8, 1, 1)
4h 14m 30s
amd64 / collect-docker-tags
1s
amd64 / test-rosetta-t5x / vit-single-process-multi-device (8)
4h 27m 43s
amd64 / test-rosetta-t5x / multi-gpu-multi-node (2N8G-te-1, 8, 2, --gin.train/utils.DatasetConfig.pack=False --gin.train_eva...
4h 50m 12s
amd64 / test-rosetta-t5x / single-process-multi-device (1P1G_te-0, 1, --enable-te 0)
4h 19m 47s
amd64 / test-rosetta-t5x / multi-gpu-multi-node (2N2G_te-0, 2, 2, --enable-te 0)
4h 36m 52s
amd64 / test-rosetta-t5x / single-process-multi-device (1P8G_te-1, 8, --gin.train/utils.DatasetConfig.pack=False --gin.train...
4h 59m 7s
amd64 / test-rosetta-t5x / vit-multi-gpu-multi-node (1, 1)
4h 18m 12s
amd64 / test-rosetta-t5x / vit-multi-gpu-multi-node (1, 2)
4h 32m 29s
amd64 / test-rosetta-t5x / vit-multi-gpu-multi-node (8, 1)
4h 23m 24s
amd64 / test-rosetta-t5x / vit-multi-gpu-multi-node (8, 2)
4h 54m 10s
arm64 / test-levanter
0s
arm64 / test-maxtext
0s
arm64 / test-upstream-t5x
0s
arm64 / build-rosetta-t5x / build-rosetta
13m 41s
arm64 / test-upstream-pax
0s
arm64 / test-te
0s
arm64 / test-te-multigpu
0s
arm64 / build-rosetta-pax / build-rosetta
11m 33s
arm64 / test-rosetta-t5x
0s
arm64 / collect-docker-tags
1s
arm64 / test-rosetta-pax
0s
amd64 / test-te-multigpu / sitrep
5s
amd64 / test-maxtext / test-maxtext-metrics
14s
amd64 / test-maxtext / test-maxtext-summary
1s
amd64 / test-maxtext / test-maxtext-sitrep / sitrep
6s
amd64 / test-maxtext / test-maxtext-outcome
1s
amd64 / test-upstream-pax / test-upstream-pax-summary
1s
amd64 / test-upstream-pax / test-upstream-pax-metrics
20s
amd64 / test-upstream-pax / test-upstream-pax-sitrep / sitrep
12s
amd64 / test-upstream-pax / test-upstream-pax-outcome
1s
amd64 / test-upstream-t5x / test-upstream-t5x-summary
1s
amd64 / test-upstream-t5x / test-upstream-t5x-metrics
13s
amd64 / test-upstream-t5x / test-upstream-t5x-sitrep / sitrep
9s
amd64 / test-upstream-t5x / test-upstream-t5x-outcome
1s
amd64 / test-rosetta-pax / test-pax-rosetta-summary
1s
amd64 / test-rosetta-pax / test-pax-rosetta-metrics
22s
amd64 / test-rosetta-pax / test-pax-rosetta-sitrep / sitrep
9s
amd64 / test-rosetta-pax / test-pax-rosetta-outcome
1s
amd64 / test-rosetta-t5x / test-t5x-rosetta-summary
1s
amd64 / test-rosetta-t5x / test-t5x-rosetta-metrics
18s
amd64 / test-rosetta-t5x / test-t5x-rosetta-sitrep / sitrep
15s
amd64 / test-rosetta-t5x / test-t5x-rosetta-outcome
1s
make-publish-configs
4s
merge-new-manifest
0s
publish-containers (equinox, mock-jax-mealkit, 500, ghcr.io/nvidia/jax-toolbox-internal:103177853... / publish
26s
publish-containers (equinox, mock-jax, 900, ghcr.io/nvidia/jax-toolbox-internal:10317785303-equin... / publish
18s
publish-containers (gemma, mock-jax, 900, ghcr.io/nvidia/jax-toolbox-internal:10317785303-gemma-a... / publish
21s
publish-containers (base, mock-jax, 800, ghcr.io/nvidia/jax-toolbox-internal:10317785303-base-amd... / publish
14s
publish-containers (grok, mock-jax, 900, ghcr.io/nvidia/jax-toolbox-internal:10317785303-grok-amd... / publish
19s
publish-containers (gemma, mock-jax-mealkit, 500, ghcr.io/nvidia/jax-toolbox-internal:10317785303... / publish
13s
publish-containers (grok, mock-jax-mealkit, 500, ghcr.io/nvidia/jax-toolbox-internal:10317785303-... / publish
18s
publish-containers (jax, mock-jax, 1000, ghcr.io/nvidia/jax-toolbox-internal:10317785303-jax-amd6... / publish
15s
publish-containers (jax, mock-jax-mealkit, 500, ghcr.io/nvidia/jax-toolbox-internal:10317785303-j... / publish
27s
publish-containers (levanter, mock-jax-mealkit, 500, ghcr.io/nvidia/jax-toolbox-internal:10317785... / publish
14s
publish-containers (levanter, mock-jax, 900, ghcr.io/nvidia/jax-toolbox-internal:10317785303-leva... / publish
15s
publish-containers (maxtext, mock-jax, 900, ghcr.io/nvidia/jax-toolbox-internal:10317785303-maxte... / publish
26s
publish-containers (maxtext, mock-jax-mealkit, 500, ghcr.io/nvidia/jax-toolbox-internal:103177853... / publish
19s
publish-containers (t5x, mock-jax, 900, ghcr.io/nvidia/jax-toolbox-internal:10317785303-t5x-amd64... / publish
19s
publish-containers (pax, mock-jax-mealkit, 500, ghcr.io/nvidia/jax-toolbox-internal:10317785303-p... / publish
18s
publish-containers (pax, mock-jax, 900, ghcr.io/nvidia/jax-toolbox-internal:10317785303-pax-amd64... / publish
25s
publish-containers (triton, mock-jax, 900, ghcr.io/nvidia/jax-toolbox-internal:10317785303-triton... / publish
12s
publish-containers (triton, mock-jax-mealkit, 500, ghcr.io/nvidia/jax-toolbox-internal:1031778530... / publish
22s
publish-containers (t5x, mock-jax-mealkit, 500, ghcr.io/nvidia/jax-toolbox-internal:10317785303-t... / publish
23s
publish-containers (upstream-pax, mock-jax, 900, ghcr.io/nvidia/jax-toolbox-internal:10317785303-... / publish
33s
publish-containers (upstream-t5x, mock-jax, 900, ghcr.io/nvidia/jax-toolbox-internal:10317785303-... / publish
27s
publish-containers (upstream-t5x, mock-jax-mealkit, 500, ghcr.io/nvidia/jax-toolbox-internal:1031... / publish
28s
publish-containers (upstream-pax, mock-jax-mealkit, 500, ghcr.io/nvidia/jax-toolbox-internal:1031... / publish
13s
finalize / workflow-badge
14s
finalize / report
18s
finalize / upload-badge
11s
finalize / publish-badge
2s
7d 23h 26m 5s