Fast, simple, scalable storage
that protects your data
SeaweedFS Enterprise adds recovery from human mistakes, cost-saving erasure coding, and priority support to the open-source distributed store you already run — S3 objects, Iceberg tables, and POSIX files from one cluster.
Cisco
NVIDIA
Proton
TrueNAS
Ænix
What sets SeaweedFS apart
The headline features — one at a time as you scroll.
Cold metadata should weigh nothing.
Sealed directories pack cold entries into compressed, erasure-coded volume chunks — the filer store shrinks 18–58×, measured on a live cluster, and everything stays readable in place.
See the metadata mathUndo anything. To the exact second.
Snapshots only protect the moments you scheduled. SeaweedFS records every change in a retention window — roll a folder, bucket, or object back to any second, overwrites and renames included, with a preview before anything moves.
See how recovery worksA real mount. Flat memory. Any scale.
mount -t seaweedvfs is a native Linux filesystem, not FUSE.
The client keeps no per-file state — its memory is sized by in-flight
I/O, not file count, so the daemon’s heap measures the same at
33M files as at 500K while FUSE mounts grow into gigabytes and OOM.
Full protection at 1.2×. Not 3×.
Tune erasure coding to your hardware: a 20+4 ratio survives four simultaneous disk failures on 60% less raw capacity than triple replication — and automatic repair rebuilds lost shards on its own.
Tune your EC ratioOne cluster. Every protocol.
S3 objects, POSIX files, HDFS jobs, and Iceberg tables served from the same cluster and the same copy of the data — no gateways, no sync jobs between systems, no second storage bill.
Deploy one clusterA lakehouse with no extra parts.
The built-in Apache Iceberg REST catalog serves S3 table buckets directly — no separate catalog service — and compacts Parquet files and cleans up snapshots in the background. Query from Spark, Trino, Dremio, or DuckDB.
Explore Iceberg tablesThree things Enterprise adds
Add a license key to your existing cluster — same deployment, same APIs, same tools — and get production data protection, lower storage cost, and operational control. Free for dev & test under 25TB.
Data Recovery
Undelete files within a configurable retention window. SeaweedFS delays garbage collection and replays the filer log so operators can restore from the Admin UI — no snapshot needed. Learn more →
Point-in-Time Recovery
Roll a folder, bucket, or object back to exactly how it looked at a chosen moment — reverting overwrites and restoring deletes across a whole subtree, with a preview before anything changes. Learn more →
Self-Healing Storage
Automatically detects and removes corrupted data entries, keeping your data intact even after an unexpected server shutdown. Learn more →
Automatic EC Repair
Detects and rebuilds missing or corrupted erasure-coding shards on its own, keeping data available through disk and node failures. Learn more →
EC Bitrot Scrub
Periodically verifies every erasure-coded shard against a checksum — catching silent disk corruption, even in cold parity shards reads never touch, and (optionally) quarantining bad shards so EC rebuild regenerates a clean copy. Learn more →
Customizable EC Ratio
Tune erasure coding to your hardware. A 20+4 ratio gives full failure protection at just 1.2× storage overhead — far less than 2–3× replication. Learn more →
zstd Compression
New data is compressed with zstd by default — typically 10–30% smaller than gzip and several times faster to decode, with no format change and nothing to migrate. Learn more →
EC Volume Vacuum
Automatically compacts erasure-coded volumes to reclaim the space left behind by deleted data — no manual maintenance windows. Learn more →
Remote Volume Vacuum
Reclaims deleted-byte cost in cloud-tiered volumes by compacting the remote object on a worker — no volume-server disk needed, and it runs in the background. Learn more →
Sealed Directories
Packs a cold directory's entries into compressed, volume-stored chunks — roughly 18–58× less filer metadata for that subtree. Seal it frozen and read-only, or mutable and still writable with automatic re-compaction. Learn more →
Admin UI with OIDC
Sign operators in through your existing identity provider with admin or read-only roles mapped from their groups — no separate passwords. Learn more →
Multi-Tenancy & S3 QoS
Isolate tenants and cap per-tenant S3 request rates so one noisy workload can't starve the rest of the cluster. Learn more →
Priority Support & License Fallback
Email, Slack, and Telegram support with a response within 24h on business days — and a license fallback that keeps your cluster serving if a license check is ever delayed. Learn more →
Built on open-source SeaweedFS
Enterprise rides on the battle-tested open-source core: a single distributed engine that serves as your S3 object store, Iceberg table warehouse, and POSIX filesystem — all from the same cluster, with your data still portable and yours.
Object Store
S3-compatible distributed object store. Scales linearly from gigabytes to petabytes with sub-millisecond lookups.
Iceberg Catalog & Maintenance
A built-in Apache Iceberg REST catalog over S3 table buckets — no separate catalog to deploy. It compacts Parquet files and cleans up in the background. Query from Spark, Trino, Dremio, DuckDB, and more.
POSIX File Mount
Mount over FUSE or the kernel mount — 100% POSIX (hardlinks, permissions, atomic renames), with peer-to-peer chunk sharing between clients to accelerate repeated reads.
O(1) Disk Reads
Every file resolves to a single disk seek, so read latency stays flat and sub-millisecond whether you store millions or billions of files.
Append-Only Writes
Writes only ever append — never overwrite in place. Sequential I/O keeps ingestion fast and makes data immutable and recoverable by design.
Linear Scalability
Add nodes to scale capacity and throughput predictably. Proven at petabyte scale with billions of files.
HDFS & Kubernetes
Works as a drop-in HDFS replacement and runs natively on Kubernetes with a first-class Helm chart and CSI driver.
Add a License Key. Keep Everything Else.
SeaweedFS Enterprise extends the open-source cluster you already run. There is no migration and no new architecture — just a license key.
- Same deployment, same APIs, same tools
- Protection runs in the background automatically
- No vendor lock-in — your data stays portable
- Free for dev & test under 25TB with no feature gates
Open Source vs. Enterprise
Same architecture and APIs. Enterprise adds data protection, efficiency, and support on top of the foundation you already trust.
| Capability | Open Source | Enterprise |
|---|---|---|
| S3 / FUSE / HDFS / Iceberg APIs | ✓ | ✓ |
| Erasure Coding | ✓ Fixed ratio | ✓ Customizable (e.g. 20+4 = 1.2×) |
| Storage Compression | ✓ gzip | ✓ zstd by default + gzip |
| Sealed Directories (filer metadata reduction) | — | ✓ |
| Data Recovery (undelete) | — | ✓ |
| Point-in-Time Recovery | — | ✓ |
| Self-Healing Storage | — | ✓ |
| Automatic EC Repair & Vacuum | — | ✓ |
| Remote Volume Vacuum (cloud-tier reclaim) | — | ✓ |
| EC Bitrot Scrub (silent corruption) | — | ✓ |
| Admin UI with OIDC | — | ✓ |
| Multi-Tenancy & S3 QoS | — | ✓ |
| Support | Community | Email / Slack / Telegram — within 24h |
Ready to Protect Your Data?
Evaluate up to 25TB free for development and testing. No credit card required. Add a license key to your existing cluster and turn on protection in minutes.