Open-source object indexer for S3-compatible storage
Your objects. Indexed.
One fast metadata API and lifecycle event stream across storage providers. Stop reimplementing pagination loops and provider glue. Start shipping.
Without Traverso
- ListObjectsV2 pagination loops in every service
- Provider-specific event shapes nobody maintains
- Bucket listings that timeout at scale
- Custom sync layer held together by cron and prayer
With Traverso
- One index-backed API for browse, lookup, and stats
- One normalized lifecycle event for all consumers
- Predictable query latency from a real index
- Scheduled reconciliation with partition checkpoints
How it works
From raw bucket state to reliable internal data.
Connect
Register bucket URIs, credentials, and partition strategy.
Index
Scheduled scans and opportunistic updates build the index.
Query
Serve metadata lookups from a fast, stable API surface.
Emit
Publish normalized lifecycle events to your queues.
Capabilities
Built for the messy reality of multi-provider storage.
Multi-provider
AWS S3, Backblaze B2, Wasabi, RStor. One adapter interface.
→Partition-safe scans
Coordinator shards work. Workers checkpoint. Timeouts re-queue.
→Opportunistic freshness
App-side writes update the index instantly via event endpoint.
→Scoped API keys
Bucket-level scopes, per-key rate limits, operations gating.
→Normalized events
Created, updated, deleted, transitioned, restored. One schema.
→SST v4 infra
Aurora, Lambda, SQS, SNS. Deploy with one command.
→Make storage indexing the boring part.
Deploy Traverso. Point it at your buckets. Let your team build on one clean contract.