voldemort/voldemort

Health v2 · maintenance only

Adoption not part of the score

2686 stars · 580 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Voldemort is a distributed key-value storage system modeled on Amazon's Dynamo, providing automatic replication, partitioning, versioning, and multi-datacenter support. It was primarily developed and used by LinkedIn, which discontinued production usage in 2018 and the project is no longer under development.

Use cases

• store key-value data across a replicated distributed cluster
• run a Dynamo-style NoSQL database with no single point of failure
• serve batch-computed data from Hadoop at low latency
• replicate data across multiple geographically distant datacenters
• use pluggable serialization like Protocol Buffers or Thrift for rich values

When to choose

• you need a proven Dynamo-inspired distributed key-value store on the JVM
• you want multi-datacenter replication with versioned data and no central coordinator
• you need to serve Hadoop batch-computed read-only datasets

When to avoid

• you need an actively maintained project - it is abandoned since LinkedIn stopped using it in 2018
• you need ACID transactions or relational queries
• you want a modern alternative - consider Venice or other maintained NoSQL stores

Facets

service · maturity abandoned

database caching serialization databases microservices big-data jvm self-hosted key-value-store dynamo-clone nosql distributed-storage eventual-consistency linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="voldemort/voldemort")