// the find
tigrisdata-archive/tigrisdb
Tigris is a globally distributed, multi-cloud object storage service with built-in support for the S3 API. It uses Dynamic Data Placement and Access-Based Rebalancing to deliver low-latency access worldwide — without the need to manage replication or caching.
This is the archived codebase of Tigris's original database product — a document store with transactions, built on FoundationDB. The company pivoted to S3-compatible object storage; this repo is a dead end. Only worth looking at if you're studying how to layer a document database on top of FDB.
The query layer is well-structured — filter, sort, aggregation, and vector search each live in their own package with tests. Schema inference and migration logic are non-trivial and worth reading if you're building schema-on-write over a KV store. The metrics and quota subsystems are production-grade: per-namespace rate limiting, Datadog integration, and granular per-operation instrumentation. Multi-protocol mux (gRPC + HTTP) in server/muxer is clean.
The repo is archived and the company has moved on — there is zero chance of bug fixes, security patches, or answered issues. The README is four lines and a redirect; the actual architecture is undocumented beyond scattered RFCs. FoundationDB as the storage backend adds a significant operational dependency that most teams will not want to run. Secondary index support was clearly still in progress (there's an RFC doc but the implementation shows signs of being unfinished).