dgraph-io/ristretto

Health v2 · maintenance only

Adoption not part of the score

6983 stars · 444 forks observed · 2026-08-28

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

Ristretto is a high-performance, fully concurrent in-memory cache library for Go, built originally for Dgraph. It uses a TinyLFU admission policy paired with SampledLFU eviction to deliver high hit ratios and throughput, with cost-based eviction and optional metrics.

Use cases

• add a fast in-memory cache to a Go application
• cache database query results in memory
• reduce contention in a highly concurrent Go service
• cache with cost-based eviction for variable-size items
• improve cache hit ratios over a plain map or LRU
• monitor cache hit ratio and throughput metrics

When to choose

• you need a production-ready, concurrent memory-bound cache in Go
• hit ratio and throughput under heavy concurrency matter
• you want cost-based eviction for items of varying sizes

When to avoid

• you need a distributed or shared cache across processes (use Redis or similar)
• your project is not written in Go
• you need persistent or disk-backed caching

Facets

library · maturity stable

caching developer-tools performance databases go cross-platform in-memory-cache lfu-eviction tinylfu concurrent generics

Member repositories

For agents

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