Records · kv:2610.00023v1

Reproduction of valkey-io/valkey: valkey-benchmark throughput

K-Veritas Team

Submitted by K-Veritas Team

Published 2026-10-08

Independent reproduction · Benchmark or leaderboard entry · Databases

Reproduction of: original code

Abstract

Built Valkey from source at the pinned commit (make, default settings, jemalloc) and ran the bundled valkey-benchmark against a local server with persistence disabled: 2,000,000 requests, 50 parallel clients, pipeline depth 16, for SET, GET, LPUSH and LPOP, reporting requests per second and latency percentiles. Server and benchmark ran on the same machine (Intel Xeon W-2145, 16 threads).

Sealed report

Cite this record

@misc{kveritas261000023v1,
  title={Reproduction of valkey-io/valkey: valkey-benchmark throughput},
  author={K-Veritas Team},
  year={2026},
  howpublished={K-Veritas Records},
  note={kv:2610.00023v1},
  url={https://kveritas.org/records/2610.00023v1},
}

References

1. [1] Mamadou K. Keita and Christopher Homan. Computer Science Conferences Should Require Nonrepudiable Experimental Results. NeurIPS 2026 Position Paper Track. arXiv:2605.08586.

Sealed with K-Veritas [1]. Listed, not endorsed. A record shows these results came from the sealed code, unchanged. Whether the work is sound is for the reader to judge.