Skip to content

(Single-card) Model perf tests #6438

(Single-card) Model perf tests

(Single-card) Model perf tests #6438

Triggered via schedule November 24, 2024 17:00
Status Cancelled
Total duration 1h 8m 49s
Artifacts 6

perf-models.yaml

on: schedule
build-artifact  /  ...  /  build-docker-image
1m 12s
build-artifact / build-docker-image / build-docker-image
Matrix: build-artifact / build-artifact
Matrix: models-perf / models-perf
Fit to window
Zoom out
Zoom in

Annotations

7 errors and 40 notices
models-perf / llm_javelin GS
The runner has received a shutdown signal. This can happen when the runner service is stopped, or a manually started runner is canceled.
pcie-cards-are-being-used-startup
Tenstorrent cards seem to be in use - this shouldn't be happening. Please let the infra team know by filing an issue with the CI job link and tagging them. Rebooting
models-perf / llm_javelin GS
The operation was canceled.
models-perf / llm_javelin N300 WH B0
The run was canceled by @shwetankTT.
pcie-cards-are-being-used-cleanup
Tenstorrent cards seem to be in use. Killing PIDs and exiting unsuccessfully. This can happen if a test hung and is normally an issue with the test, rather than infra.
models-perf / llm_javelin N300 WH B0
Process completed with exit code 1.
models-perf / llm_javelin N300 WH B0
The operation was canceled.
printing-out-smi-info-cleanup
Touching and printing out SMI info
disk-usage-before-startup
Disk usage is 90 %
disk-usage-after-startup
Disk usage is 90 %
printing-smi-info-startup
Touching and printing out SMI info
reset-successful-startup
tt-smi reset was successful
printing-out-smi-info-cleanup
Touching and printing out SMI info
successful-reset-cleanup
tt-smi reset was successful
reset-successful-cleanup
tt-smi reset was successful
disk-usage-before-startup
Disk usage is 33 %
disk-usage-after-startup
Disk usage is 33 %
printing-smi-info-startup
Touching and printing out SMI info
printing-out-smi-info-cleanup
Touching and printing out SMI info
successful-reset-cleanup
tt-smi reset was successful
reset-successful-cleanup
tt-smi reset was successful
disk-usage-before-startup
Disk usage is 59 %
disk-usage-after-startup
Disk usage is 59 %
printing-smi-info-startup
Touching and printing out SMI info
reset-successful-startup
tt-smi reset was successful
disk-usage-before-startup
Disk usage is 33 %
disk-usage-after-startup
Disk usage is 33 %
printing-smi-info-startup
Touching and printing out SMI info
reset-successful-startup
tt-smi reset was successful
printing-out-smi-info-cleanup
Touching and printing out SMI info
successful-reset-cleanup
tt-smi reset was successful
reset-successful-cleanup
tt-smi reset was successful
printing-out-smi-info-cleanup
Touching and printing out SMI info
successful-reset-cleanup
tt-smi reset was successful
reset-successful-cleanup
tt-smi reset was successful
disk-usage-before-startup
Disk usage is 33 %
disk-usage-after-startup
Disk usage is 33 %
printing-smi-info-startup
Touching and printing out SMI info
reset-successful-startup
tt-smi reset was successful
printing-out-smi-info-cleanup
Touching and printing out SMI info
attempting-reset-cleanup
Attempting to reset card(s).
successful-reset-cleanup
tt-smi reset was successful
reset-successful-cleanup
tt-smi reset was successful
disk-usage-before-startup
Disk usage is 90 %
disk-usage-after-startup
Disk usage is 90 %
printing-smi-info-startup
Touching and printing out SMI info
reset-successful-startup
tt-smi reset was successful

Artifacts

Produced during runtime
Name Size
TTMetal_build_grayskull
299 MB
TTMetal_build_wormhole_b0
299 MB
perf-report-csv-cnn_javelin-grayskull-bare_metal
333 Bytes
perf-report-csv-cnn_javelin-wormhole_b0-bare_metal
494 Bytes
perf-report-csv-other-grayskull-bare_metal
1.11 KB
perf-report-csv-other-wormhole_b0-bare_metal
700 Bytes