Cuda weekly benchmarks

Cuda weekly benchmarks #11

Sign in to view logs

Triggered via schedule February 1, 2025 01:30

soonum

main

Status Failure

Total duration 7h 42m 4s

Artifacts 1

benchmark_gpu_integer_weekly.yml

on: schedule

Run benchmarks (1xH100) / Prepare operations matrix

Run benchmarks (2xH100) / Prepare operations matrix

Run benchmarks (8xH100) / Prepare operations matrix

Run benchmarks (L40) / Prepare operations matrix

Run benchmarks (1xH100) / Setup instance (cuda-single-h100-benchmarks)

Run benchmarks (2xH100) / Setup instance (cuda-2-h100-benchmarks)

Run benchmarks (8xH100) / Setup instance (cuda-multi-h100-benchmarks)

Run benchmarks (L40) / Setup instance (cuda-l40-benchmarks)

Matrix: Run benchmarks (1xH100) / cuda-benchmarks

Matrix: Run benchmarks (2xH100) / cuda-benchmarks

Matrix: Run benchmarks (8xH100) / cuda-benchmarks

Matrix: Run benchmarks (L40) / cuda-benchmarks

Run benchmarks (1xH100) / Slack Notification

Run benchmarks (2xH100) / Slack Notification

Run benchmarks (8xH100) / Slack Notification

Run benchmarks (L40) / Slack Notification

Run benchmarks (1xH100) / Teardown instance (cuda-single-h100-benchmarks)

Run benchmarks (2xH100) / Teardown instance (cuda-2-h100-benchmarks)

Run benchmarks (8xH100) / Teardown instance (cuda-${{ inputs.profile }}-benchmarks)

Run benchmarks (L40) / Teardown instance (cuda-l40-benchmarks)

Annotations

6 errors

Run benchmarks (8xH100) / Setup instance (cuda-multi-h100-benchmarks)

Instance task failed (details: api call failed (reason: bad_request, message: Not Enough Stock of H100-80G-PCIe. Unable to launch virtual-machines.))

Run benchmarks (8xH100) / Setup instance (cuda-multi-h100-benchmarks)

Failure occurred while waiting for instance.

Run benchmarks (8xH100) / Setup instance (cuda-multi-h100-benchmarks)

Instance stop request has failed (HTTP status code: 500, body: Failed to stop backend instance (reason: backend provider not found with instance key: InstanceHashMapKey("99664d6ce0e1b3c353514967e5978ce56f4c6040", "87368f41-be8a-4ff6-9550-2a807bebb17d")))

Run benchmarks (8xH100) / Setup instance (cuda-multi-h100-benchmarks)

instance stop request failed

Run benchmarks (1xH100) / Cuda benchmarks (single-h100)

Process completed with exit code 2.

Run benchmarks (L40) / Cuda benchmarks (l40)

Process completed with exit code 2.

Artifacts

Produced during runtime

Name	Size
99664d6ce0e1b3c353514967e5978ce56f4c6040_integer_multi_bit_default_2-h100	26.2 KB