RustFS vs S3 in EU: MinIO Is Dead. Here's Your Escape Route.

Update, September 2026: RustFS 1.0 is GA

RustFS 1.0.0 was released on 16 September 2026 as the first general-availability version. The project now marks its core object storage (S3 API, multipart uploads, versioning, object lock, replication, encryption) as ready for production. Some features are still in preview, including S3 Tables and MinIO on-disk compatibility.

Parts of this article were written during the alpha phase, before GA. We updated them in October 2026.

Why MinIO Died (And Why This Matters)

MinIO didn't fail because the technology broke. It failed because the company decided to stop investing in the open-source version.

In December 2025, MinIO Inc. made an official announcement: the project was entering "maintenance mode." That's corporate-speak for: "we're not building this anymore, but we won't formally kill it."

What does maintenance mode actually mean?

• No new features (you get security patches, maybe, if they feel like it)
• No PR reviews (community contributions sit in limbo)
• No guaranteed security updates (if a vulnerability is discovered, you're on your own)

Before that, in October 2025, MinIO locked the web console behind a $96,000/year enterprise license. The free version has a limited UI. Suddenly, managing your storage clusters became expensive.

This wasn't an accident. MinIO Inc. made a deliberate choice: push users toward the commercial offering, or push them out.

The result? If you're running MinIO today, you're on legacy software. And if you're planning to migrate from S3 to MinIO in 2026, you're migrating to a dead end.

Why S3 Is Expensive 

Everyone complains about S3 pricing. They're right to complain. But they're complaining about the wrong thing.

S3 isn't expensive because AWS is greedy. S3 is expensive because its pricing model is unpredictable. You pay for storage (reasonable). You pay for requests (fine). But then you pay for egress—and here's where it compounds: that €0.09 per GB of data leaving AWS adds up in ways you didn't anticipate.

Store 15 TB. Transfer 200 TB per month (typical for B2B SaaS backups). Here's what happens:

Storage: 15 TB × €0.023/GB-month = €345

Egress: 200 TB × €0.09/GB = €18,000

API calls: 10M requests × €0.0004 = €4,000

---

MONTHLY: €22,345

YEARLY: €268,140

By year three, that bill is €370,000. You didn't add workloads. AWS just found new ways to charge you.

The second problem is sovereignty. Your data lives in US infrastructure, governed by US law. You say "we're GDPR compliant" because AWS says so. But you don't control the data center. You don't control the region. You don't control the exit path. You're renting compliance, not owning it.

Neither of these problems is catastrophic for small teams. But they compound.

RustFS: The MinIO Replacement That Actually Works

RustFS is a new object storage server written in Rust. It speaks the S3 API. It can hold your data the same way S3 does. And here's the killer feature: it can literally replace MinIO in your infrastructure without changing a single line of code.

Drop-in at the API level. Your applications keep working without code changes: same S3 API, same SDKs. Existing MinIO data moves by copy (bucket replication or mirroring), not by swapping the binary: MinIO on-disk compatibility in RustFS is still in preview and off in the default build, and objects encrypted by MinIO cannot be read in place.

Performance:

• 2.3x faster than MinIO for small object payloads (4KB—typical for backups, logs, metadata)
• Full S3 compatibility (versioning, WORM compliance, encryption, replication)
• Entire binary is under 100MB (runs on ARM, VPS, data centers)
• Apache 2.0 licensed (no AGPL surprises)

The catch: RustFS reached GA only on 16 September 2026 (version 1.0.0). The project marks its core S3 engine as production-ready, but some features are still in preview, including S3 Tables and MinIO on-disk compatibility. Its production track record is short compared to Ceph.

So why talk about it?

Because RustFS represents the future. MinIO is dead. SeaweedFS is production-ready but was built for different priorities (billions of small files, not backup scenarios). And Garage is too niche (geo-distributed homelab storage, AGPL licensing).

RustFS fills the gap: it's the spiritual successor to MinIO, built for the same use case (S3 replacement), but better designed and actually maintained.

Timeline reality:

• 29 April 2026: alpha moves to beta
• 16 September 2026: RustFS 1.0.0, the first GA release. Core S3 engine production-ready
• Next: preview features (S3 Tables, MinIO on-disk compatibility) move to stable. Check the release notes before you rely on them

If you're on S3 today and hate the costs, your options are:

1. Self-host RustFS 1.0 (free licence, your team runs ops and on-call)
2. Switch to SeaweedFS or Ceph (longer production track record, different architecture)
3. Use managed RustFS (we run ops: 24/7 support, patches, 99.9% SLA)

The Real Numbers: TCO Comparison

Scenario: German SaaS company, 15 TB stored, 200 TB monthly egress (typical B2B SaaS backup scenario)

AWS S3 (2026 pricing)

Storage: 15 TB × €0.023/GB-month = €345

Egress: 200 TB × €0.09/GB = €18,000

API calls: 10M requests × €0.0004 = €4,000

---

MONTHLY: €22,345

YEARLY: €268,140

RustFS Self-Hosted (DIY)

Infrastructure: €7,200/year

DevOps salary (1 FTE): €60,000/year

---

YEARLY: €67,200

But there's a catch: You're hiring a full-time DevOps engineer just to run RustFS. If you self-host, you're:

• Running a young storage engine (GA since September 2026) in production on your own
• Responsible for security patches
• On the hook if something breaks at 2am

RustFS Managed (Hostzero)

Usage-based pricing: €0.03/GB/month

One-time setup fee: €3,000

For 15 TB scenario:

15,360 GB × €0.03/month × 12 months = €5,529.60/year

---

FIRST YEAR: €8,529.60 (€5,529.60 + €3,000 setup)

YEARS 2+: €5,529.60/year

Includes: ops, support, security patches, monitoring, backups, SLA 99.9%, 24/7 response.

Your data is yours: zero lock-in, full autonomy, migrate anytime.

Why Managed is Actually Cheaper:

DIY requires hiring a full DevOps engineer (€60k/year).

Managed RustFS costs €5.5k/year for the same infrastructure + professional ops.

What you get:

- 24/7 support (we take on the operational risk: patches, upgrades, monitoring, on-call)

- Security patches (immediate response)

- SLA 99.9% uptime guarantee

- Your data stays yours (zero lock-in, full autonomy, migrate anytime)

- Your team focuses on product, not ops

The Numbers Over Time

The math is brutal. At 15TB, Managed RustFS is about 31x cheaper than S3 in year one (€268k vs €8.5k, setup included).

Managed is cheaper than DIY from year one (€8.5k vs €67k).

By year two, managed is 53x cheaper than S3, and you haven't spent a dime on hiring.

Migration Path: From S3 to RustFS (The Realistic Timeline)

Total time: 6 weeks. Downtime: Zero (if done correctly).

Week 1-2: Planning

• Audit your S3 buckets (total size, request patterns, data sensitivity)
• Classify your data (which stays hot, what's archive)
• Plan your RustFS infrastructure (how many nodes, where physically)
• File GDPR verification request if needed (Hostzero handles this)

Week 3-5: Migration

• Deploy RustFS in parallel to S3 (don't touch production yet)
• Enable dual-write: new data goes to both S3 and RustFS simultaneously
• Copy existing data over a dedicated network link (~2 weeks for 15TB on Hetzner connection)
• Test your application against RustFS endpoints (make sure S3 API calls work)
• Run checksums to verify data integrity (crucial step, don't skip)

Week 6: Cutover

• Switch application reads to RustFS
• Keep S3 as read-only fallback for 30 days (safety net)
• Monitor RustFS performance (it will be faster—lower latency to Frankfurt)
• Decomission S3 after 30 days

Decision Framework: When to Migrate, When to Wait

Migrate to RustFS NOW if:

• Your AWS bill exceeded €200k/year (math is obvious)
• You have EU data residency requirements (GDPR strict interpretation)
• You want an exit strategy from AWS ecosystem lock-in
• You're currently on MinIO and need a migration path (DIY or managed)

Choose Ceph or SeaweedFS instead if:

• You need block or file storage, not only S3
• You need a production track record longer than a few months
• You depend on features RustFS still has in preview (S3 Tables, reading MinIO data in place)

Stay on S3 if:

• You're storing less than 5TB and growing slowly (cost difference is immaterial)
• You need AWS ecosystem integration (Lambda, Glue, SageMaker, etc.)
• You value "set and forget" simplicity over cost optimization
• Your compliance requirements don't demand EU data residency

The Real Talk: Should You Use RustFS in 2026?

Yes, in production, with a staged rollout.

Since version 1.0 (September 2026), RustFS is GA and its core S3 engine is supported for production traffic. It is still younger than Ceph or SeaweedFS: fewer runbooks, fewer backup integrations, fewer engineers who have run it at scale.

A sensible path:

• Start with backups, logs and other non-critical buckets
• Test in staging against your real S3 calls before you switch reads
• Keep preview features (S3 Tables, MinIO on-disk compatibility) out of production for now

Your options:

• Self-host RustFS 1.0, if you have the ops team for it
• Use SeaweedFS or Ceph, if you need a long production track record or block storage
• Use managed RustFS (we run the cluster in Frankfurt, with 24/7 support and a 99.9% SLA)

Why Managed RustFS Matters

The gap between "free but risky" (DIY RustFS) and "stable but expensive" (S3) is real. Managed RustFS closes that gap.

Here's what changes when you use managed:

• We run the operations — If RustFS has a security issue, we patch it immediately. You don't.
• We do the ops — Your team builds product. We manage infrastructure.
• You get an SLA — 99.9% uptime guarantee. With DIY, you get whatever your contractor can deliver.
• You're never locked in — Your data is portable. Take it and run RustFS yourself anytime. No lock-in.

For about €5.5k/year at 15 TB (€0.03/GB/month, plus €3,000 one-time setup), you get:

• Frankfurt data center (GDPR native, low latency, renewable energy)
• 24/7 support (actual humans, not chatbots)
• Monitoring + alerting (you see your data, we see everything)
• Backups + disaster recovery (tested and ready)
• Security patches (applied immediately)
• Migration support (we handle the data movement)

The Bottom Line

MinIO is dead. It's not coming back. MinIO Inc. made a business decision, and now they're pushing users out.

Your options:

1. Stay on S3 and accept the cost (€268k/year is just the starting point)
2. Move to SeaweedFS now (mature, proven, but different architecture)
3. Need block storage or production maturity now? Run on a Ceph cluster we operate in Frankfurt (S3, block and file storage on one mature open-source cluster)
4. Self-host RustFS 1.0, if you have the ops team for it
5. Use managed RustFS today (get the future now, with support + SLA)

If you've been on MinIO for years, or planning to migrate from S3, the answer is clear: RustFS 1.0 is GA. For most teams in 2026, managed RustFS is the shortest path: a production-ready core, and we carry the operations.

Ready to Escape S3?

Talk to an engineer. 

30-minute call, no sales pitch.