BrewFSBrewFS0.1
Test MethodAgent WorkloadsBenchmarksCompareArchitectureRust NativeUse CasesDocsBlogRoadmap中文
GitHubGet Started
Benchmark · RandRW

RandRW Benchmark

BrewFS delivers 92% more fully drained mixed random-I/O throughput versus JuiceFS in the July 24, 2026 Redis + RustFS snapshot (218.60 vs 113.86 MiB/s).

BrewFS
1.92×
Fully drained mixed random-I/O throughput
JuiceFS
1.0×
JuiceFS baseline under the same configuration
This result applies to the tested RandRW workload configuration. It should not be interpreted as a universal claim that BrewFS is faster in every scenario.

Test Configuration

Published for reproducibility

Every value below is part of the benchmark contract. This snapshot uses the published runner profile and records its effective parameters with the raw artifacts.

Metrics captured: IOPS, throughput, average latency, P95, and P99 — for both cold-cache and warm-cache states.

Benchmark TypeRandRW
ComparisonBrewFS vs JuiceFS
Workload ModelAgent-style random read/write
Filesystem AccessFUSE
Data BackendRustFS, S3-compatible
Metadata BackendRedis
Block Size4 MiB
Concurrency8 fio jobs, iodepth=1
Per-job size / runtime512 MiB / 20 s (time-based)
Read / Write Ratio70 / 30
Cache StateDurable prefill, drain, cache-preserving remount
Benchmark Toolfio with io_uring
Mount Profilewriteback-throughput-profile

Reproduce it

Recommended fio profile

This fio shape matches the published mixed random-I/O workload; use the repository runner to reproduce mount, prefill, drain, and cache handling.

agent-randrw.fio
[agent-randrw]
rw=randrw
rwmixread=70
bs=4m
iodepth=1
numjobs=8
size=512m
runtime=20
time_based=1
direct=0
group_reporting=1

Charts

What the benchmark measures

Overall RandRW Performance

Relative throughput, higher is better

BrewFS · drained total
1.92×
JuiceFS · drained total
1.0×

IOPS & Throughput

Mixed 70/30 read/write, relative

BrewFS
JuiceFS

Latency (lower is better)

avg · P95 · P99

avg
P95
P99

Multi-agent Scaling

Aggregate throughput by concurrent agents

1
2
4
8
16

The fully drained comparison is the published result. IOPS, latency, and multi-agent scaling panels remain illustrative until matching results are published.


Conclusion

RandRW is a better proxy for agent workloads

Agents continuously mix small reads, partial writes, metadata access, and artifact creation — not pure sequential read/write. BrewFS is optimized for this shared, random, and concurrent filesystem access.

BrewFSBrewFSGitHub

Rust-native filesystem for multi-agent collaborative read/write workloads.

Product
What is BrewFSArchitectureRust NativeUse Cases
Agents
Agent WorkloadsShared WorkspaceRandRW Benchmark
Benchmarks
OverviewRandRWCompare JuiceFS
Community
GitHubRoadmapCommunityDocsBlog

© 2026 BrewFS · Apache-2.0 · Built in Rust 🦀

BrewFS is an early, fast-evolving open-source project. Performance claims refer to specific agent-style RandRW benchmarks; production readiness should be validated per workload.