Multi-agent Scaling

How BrewFS behaves with 1 to 16 concurrent agents.

This benchmark measures how BrewFS behaves with 1, 2, 4, 8, and 16 concurrent agents sharing the same workspace: aggregate throughput, per-agent latency, and metadata contention.

Read/write mix scenarios: 70/30 · 50/50 · 30/70.

Multiple agents running together amplify random IO, metadata pressure, and concurrent workspace updates — which is exactly the pattern BrewFS is optimized for.

Metadata-heavy Benchmark

Isolating the metadata path across pluggable backends.

Reproducibility

Every benchmark publishes its full contract.

Get started

Benchmarks Overview RandRW Benchmark Metadata-heavy Benchmark Multi-agent Scaling Reproducibility Benchmark Methodology

Benchmarks