dgraph-io/badger

Health v2 · maintenance only

Adoption not part of the score

15761 stars · 1313 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

BadgerDB is an embeddable, persistent, and fast key-value database written in pure Go, serving as the storage engine for Dgraph. It is designed as a performant Go-native alternative to RocksDB, optimized for SSDs and supporting concurrent ACID transactions with serializable snapshot isolation.

Use cases

• embed a fast key-value store in a Go application
• replace RocksDB or LevelDB with a pure Go alternative
• store metadata or state for distributed systems
• build a local persistent cache with high write throughput
• persist large datasets on SSDs from a Go service
• need ACID transactions in an embedded database

When to choose

• you are writing Go and want zero CGo dependencies
• you need an embedded KV store with ACID transactional guarantees
• your workload is write-heavy and runs on SSDs
• you want a proven store serving hundreds of terabytes in production

When to avoid

• you need SQL queries or relational schemas
• you need a network-accessible client-server database
• your dataset must fit in memory and you prefer an in-memory store
• you need rich secondary indexes or document queries out of the box

Facets

library · maturity stable

database caching serialization databases developer-tools backend go windows cross-platform key-value-store embedded-database lsmtree pure-go acid-transactions ssd-optimized linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="dgraph-io/badger")