ShannonStore Key Features
Discover the core features of modern data infrastructure S3-compatible object storage for the cloud era.
Full S3 API Support
Fully supports S3 APIs — CRUD, Multipart Upload, Bucket Versioning, plus Bucket Policy, CORS, Lifecycle, Tagging, Replication, Object Lock, and presigned URLs (SigV4). Seamlessly integrates with existing S3 tools and SDKs.
High Availability
Architecture with no single point of failure (SPOF), ensuring data durability and availability.
Zero-Downtime Horizontal Scaling
Add or remove nodes without downtime. The cluster keeps up as your data grows, with no manual rebalancing.
Security
Policy-based access control plus encryption at rest and in transit keep your data safe.
Data Efficiency
Erasure Coding delivers high durability at a fraction of the storage cost of replication, with automatic recovery from disk failures.
Automatic Data Integrity
A background scrubber silently detects and repairs corrupted data so silent failures never reach your users.
High-Performance Network Processing
An asynchronous network layer absorbs heavy traffic without breaking a sweat.
Streaming Optimization
Large object streaming and Range Request optimization support.
WORM / Object Lock (Compliance)
S3 Object Lock with Governance/Compliance retention and Legal Hold. Locked objects cannot be deleted or overwritten by anyone during retention — ideal for audit and regulated workloads.
Object Lifecycle (ILM) with Tiering
A leader-driven scanner auto-expires objects (days or seconds), prunes noncurrent versions, and aborts stale multipart uploads. Transition rules move cold data to a registered Storage Class (another ShannonStore / MinIO / AWS S3), and RestoreObject pulls it back on demand — full hot/cold lifecycle without leaving the S3 API.
Storage Classes & Transparent Tiering
Register named Storage Classes with their own EC profile and tier destination, then assign per bucket. Lifecycle Transition silently moves objects out; GET on a tiered object returns 403 InvalidObjectState until RestoreObject — the S3 spec, end-to-end.
Topology-Aware Placement
A 6-pass cascade spreads each erasure-coded shard across distinct zones, racks, hosts, nodes, and disks — survive a full rack or zone loss, not just a disk. Weighted HRW handles heterogeneous disks; a rebalance worker migrates chunks when topology changes.
O(1) Bucket Stats
Per-node bucket counters update on every PUT/DELETE; cluster totals come from a single RPC fan-out — no RocksDB fullscan per page load even on multi-petabyte buckets.
Bucket & Site Replication
Async per-bucket replication and multi-cluster active-active site replication. Each site re-encrypts with its own KMS, and bidirectional replication converges without an infinite loop — ideal for cross-region DR.
Open Data Ecosystem
Integrates with open-source data ecosystems including Iceberg, Trino, Spark, and Flink, as well as the CCL data ecosystem — Ontul, NeorunBase, and ItdaStream.
Use Cases
Private Cloud Storage
Build secure on-premises S3 storage.
Data Lake
Analytics storage for Spark, Flink, Trino, and more.
Backup & Archiving
Cost-effective long-term storage for large-scale data.
Application Storage
Image, log, and file storage for web/mobile apps.
Compliance Archive (WORM)
Immutable audit and regulated data with Object Lock retention and legal hold.
Multi-Region DR / Site Replication
Disaster recovery and availability across clusters with active-active site replication.
Considering ShannonStore for your infrastructure?
Simple. Scalable. Reliable.
Object storage that grows with your data — no downtime when you scale, all the way to petabyte-scale deployments.