Barre/ZeroFS
ZeroFS: A log-structured filesystem for S3. ZeroFS serves S3-compatible buckets as POSIX filesystems over NFS and 9P, or as raw block devices over NBD. observed · 2026-08-28
Health v2 · maintenance only
81/100
- Activity 99
- Release rhythm 87
- Longevity 29
How is this computed?
round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-03. Adoption (stars, forks) is never an input.
- gap_med: 1.0
- age_days: 415
- days_rel: 7
- days_push: 7
- n_releases_24m: 151
Adoption not part of the score
3003 stars · 113 forks observed · 2026-08-28
What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded
ZeroFS is a log-structured filesystem written in Rust that turns S3-compatible object storage (Amazon S3, GCS, Azure Blob, or any S3-compatible store) into a POSIX filesystem. It serves files over NFS and 9P and raw block devices over NBD from a single userspace process, with built-in compression (zstd/lz4) and encryption (XChaCha20-Poly1305).
Use cases
- mount an s3 bucket as a posix filesystem
- use object storage as a block device for vms
- store kubernetes persistent volumes on s3
- run a shared filesystem backed by cloud object storage
- encrypt and compress files stored in s3
- replace expensive cloud block storage with s3
- access s3 files from macos or windows over nfs
When to choose
- you want POSIX-conformant filesystem semantics on top of S3 without running a database
- you need encrypted, compressed storage where data is encrypted before it leaves the machine
- you need block devices (NBD) or Kubernetes CSI volumes backed by object storage
- you have workloads with hundreds of millions of small files and want near-raw-S3 throughput
- you want high availability with leader/standby failover over the same bucket
When to avoid
- you need strict fsync durability guarantees - NFS COMMIT semantics can return before data reaches S3 (use 9P instead)
- you need a simple one-off file sync tool rather than a full filesystem server
- your workload requires very low latency metadata operations and you have no local disk for caching
- you cannot run a persistent userspace server process and only want ad-hoc bucket access
Facets
application · maturity active
file-system object-storage cryptography compression caching backup cloud-computing self-hosted infrastructure-as-code files microservices windows bsd rust self-hosted cloud s3 nfs 9p nbd posix-filesystem log-structured lsm-tree block-device csi-driver fuse object-storage-gateway encrypted-storage storage linux macos docker kubernetes
5 sources
- readme: https://github.com/Barre/ZeroFS · fetched 2026-08-28 · 85b7a8c91889
- homepage: https://www.zerofs.net · fetched 2026-08-29 · 78b79a93b65b
- site_page: https://www.zerofs.net/docs · fetched 2026-08-29 · 10d6024cdac6
- site_page: https://www.zerofs.net/docs/quickstart · fetched 2026-08-29 · b6dc08b353cb
- registry_crates: https://crates.io/api/v1/crates/zerofs · fetched 2026-08-29 · d93a723f96f6
Member repositories
| Repository | Role | Health v2 |
|---|---|---|
| Barre/ZeroFS | main | 81 |
For agents
Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem