Command line: flow.py --with conda:packages={},python=3.12.0,disabled=False --with kubernetes:cpu=1,memory=4096,disk=10240,image=006988687827.dkr.ecr.us-west-2.amazonaws.com/obptask-python:v0.25.3,namespace=jobs-default,gpu_vendor=nvidia,tmpfs_tempdir=True,tmpfs_path=/metaflow_temp,hostname_resolution_timeout=600 --with environment:vars={} --quiet --metadata=service --environment=pypi --datastore=s3 --datastore-root=s3://obp-ri3f9l-metaflow/metaflow/ --event-logger=nullSidecarLogger --monitor=nullSidecarMonitor --no-pylint --with=argo_workflows_internal:auto-emit-argo-events=1 step do_smth --run-id argo-memraynumpyflow-p9clq --task-id t-1949d7f9 --retry-count 0 --max-user-code-retries 0 --input-paths argo-memraynumpyflow-p9clq/start/t-0c3b51d1
Start time: 2024-12-02 15:54:32.056000+00:00
End time: 2024-12-02 15:54:33.963000+00:00
Duration: 0:00:01.907000
Total number of allocations: 71071
Total number of frames seen: 596
Peak memory usage: 192.1 MiB
Python allocator: pymalloc

The table reporter provides a simple tabular representation of memory allocations in the target when the memory usage was at its peak.

You can find more information in the documentation.