Skip to content

MLPerf™ Storage Benchmark Suite

active 2023-08-252026-08-11 (UTC)

Complete coverage26,398 / 26,398 hourly files (100%) · 2 absent upstream2023-08-152026-08-18 (UTC)
Events
1.5K
Pushes
327
Pull requests
121
Issues
151
Stars
129
Forks
45

Activity over time

Daily event counts in the loaded window

Line chart, 1083 days from 2023-08-25 to 2026-08-11. Pushes: 327 total, peak 15 in a day. Pull requests: 121 total, peak 11 in a day. Issues: 151 total, peak 29 in a day. Comments: 333 total, peak 34 in a day. Stars: 129 total, peak 3 in a day.

  • Pushes
  • Pull requests
  • Issues
  • Comments
  • Stars

Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.

Top contributors

Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity

ContributorContributionsPushesPRsComments
FileSystemGuy4202223866
johnugeorge129454014
github-actions[bot]700070
zhenghh04545330
wvaske4921321
russfellows432439
glimchb400321
dslik3916110
idevasena36637
wolfgang-desalvador18305
garimajain0514025
hazemawadalla12074
LouisDDN9015
hanyunfan9008
sajikvr7001
gaowayne5003
noctarius5003
ram-sangle4022
Linzsd4001
shp7764003

Recent activity

Latest issues, pull requests and releases

  • Issue comment#732github-actions[bot]2026-07-09 01:08
    feat(#731): validate training datagen hierarchy and emit self-describing manifest
  • Issue comment#662FileSystemGuy2026-07-07 23:18
    test_rules_coverage drift tests wipe real coverage entries via monkeypatch.setattr
  • Issue#712gaikwadabhishek2026-07-07 23:06
    Object-store checkpoint write drops the `s3://` scheme for model-parallel shards (mp_rank > 0)
  • Issue#694daddypig882026-07-07 03:11
    VectorDB reportgen omits QPS/latency/recall metrics and VDB pass thresholds are unclear
  • Issue comment#572idevasena2026-07-05 23:09
    Is this a bug? RESOURCE_EXHAUSTED error when running single-host vectordb test
  • Pull request#673FileSystemGuy2026-07-04 23:47
  • Issue comment#663FileSystemGuy2026-07-04 23:24
    _ENV_RUNTIME_DENYLIST missing 4 MPI per-rank/per-invocation env vars, causing spurious SystemDriftError and defeating quantity: N dedup
  • Issue comment#640github-actions[bot]2026-07-02 04:23
    fix(#638): kvcache emits summary.json, matching submission loader
  • Pull request#609FileSystemGuy2026-07-01 18:17
  • Issue comment#613wolfgang-desalvador2026-07-01 07:18
    Fix report generation to avoid all results grouped in same JSON
  • Issue#616FileSystemGuy2026-07-01 01:28
    reportgen: label warmup run in stdout + remove --output-dir flag
  • Issue comment#587github-actions[bot]2026-06-29 23:49
    fix(#573): CAP-02 probe wire payload exceeds PIPE_BUF at scale — drop ranks array + cap failure messages
  • Issue comment#578russfellows2026-06-29 23:09
    client mem mandatory for datagen
  • Issue#278FileSystemGuy2026-06-27 02:21
    Resolve MPI deduplicatable data generation (mlpstorage integration)
  • Issue comment#278FileSystemGuy2026-06-27 02:21
    Resolve MPI deduplicatable data generation (mlpstorage integration)
  • Issue#323FileSystemGuy2026-06-27 02:18
    [Passing score blocker] `ParquetReader` involves a CPU time in `get_sample` 5x / 4x times longer than the I/O time
  • Pull request#563FileSystemGuy2026-06-27 01:13
  • Pull request#561FileSystemGuy2026-06-26 23:41
  • Pull request#557FileSystemGuy2026-06-26 23:08
  • Issue comment#556github-actions[bot]2026-06-26 22:53
    test(benchmarks): populate orgname/systemname/mode/vdb_engine in VectorDB tests
  • Issue comment#554github-actions[bot]2026-06-26 22:50
    test(code-image): refresh fixtures and assertions to md5-tree-v2
  • Pull request#547dslik2026-06-26 20:08
  • Issue comment#504russfellows2026-06-26 00:07
    Datagen fails when using the s3torchconnector
  • Issue comment#518FileSystemGuy2026-06-25 23:42
    Clarification on client host memory requirement for 1T model checkpointing – changed between v2.0 and v3.0?
  • Issue#528wolfgang-desalvador2026-06-25 22:56
    Training crashes with `RuntimeError: could not unlink the shared memory file /torch_*`

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 129 stars here means stars gained during the window, not the repo's star count.