Pepijn a32a2c647b feat(streaming): full-matrix SLURM submitter + results summarizer ...

slurm/run_streaming_matrix.sh fans the benchmark matrix (sources {hub,bucket,
warmed_bucket} x modes {single,sarm} x decode {cpu,cuda}) out as isolated single-GPU
SLURM jobs, so an OOM in one config is contained and reported per-job by SLURM. Worker
count and shuffle buffer are bounded (lower for cuda, which holds a CUDA context + NVDEC
session per worker) to avoid host/VRAM OOM. Source/mode/decode/workers/buffer/account/
partition are env-overridable; SOURCES/MODES/DECODES select subsets.

benchmarks/streaming/summarize_results.py collapses the per-run JSONs into one comparison
table + summary.csv (frames/s/node, first-batch + p50/p95/p99 latency, cache hit-rate).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

2026-06-09 15:51:36 +02:00

..

benchmark_streaming_robocasa.sh

feat(streaming): multinode example, dataloading benchmark, distributed smoke test

2026-06-09 13:48:23 +02:00

run_streaming_matrix.sh

feat(streaming): full-matrix SLURM submitter + results summarizer

2026-06-09 15:51:36 +02:00

train_streaming_robocasa.sh

feat(streaming): multinode example, dataloading benchmark, distributed smoke test

2026-06-09 13:48:23 +02:00