Skip to content

SUMMARY.md

Latest commit

 

History

History
59 lines (46 loc) · 2.58 KB

SUMMARY.md

File metadata and controls

59 lines (46 loc) · 2.58 KB
Raw

High-Performance Rust Cache Libraries (Summary)

This list focuses on Rust cache libraries that explicitly describe themselves as high performance and/or publish benchmark results. It is not a definitive ranking, but a curated summary based on each project's own documentation.

Moka (crate: moka)

Moka is a fast, concurrent cache library inspired by Java's Caffeine. It provides thread-safe sync and async caches with TinyLFU admission + LRU eviction, supports entry count and weighted-size bounds, and includes TTL/TTI expiration policies. Benchmark notes and hit-ratio results are published in its wiki.

Key features/benefits:

  • High concurrency for reads and updates with thread-safe caches.
  • Near-optimal hit ratio via TinyLFU admission and LRU eviction.
  • Sync + async cache variants plus TTL/TTI and per-entry expiration.
  • Documented benchmark results in the wiki.

Docs: https://github.com/moka-rs/moka Benchmarks: https://github.com/moka-rs/moka/wiki#admission-and-eviction-policies

Quick Cache (crate: quick_cache)

Quick Cache positions itself as a lightweight, high performance concurrent cache optimized for low overhead. It uses the S3-FIFO policy, scales with threads, and publishes benchmark comparisons in its README.

Key features/benefits:

  • Low overhead relative to a concurrent hash table.
  • S3-FIFO policy for scan resistance and strong hit rates.
  • Good multi-thread scaling; async and sync atomic operations.
  • Built-in benchmarks comparing to other caches.

Docs: https://github.com/arthurprs/quick-cache Benchmarks: https://github.com/arthurprs/quick-cache#benchmarks

Mini Moka (crate: mini-moka)

Mini Moka is a fast, concurrent cache library and a smaller, simpler sibling of Moka. It offers thread-safe caches with high concurrency for reads/updates and a non-thread-safe option for single-threaded use.

Key features/benefits:

  • Fast, concurrent cache with simpler API and smaller footprint.
  • Full concurrency of retrievals with high expected update concurrency.
  • LFU admission and LRU eviction policies.

Docs: https://github.com/moka-rs/mini-moka

foyer (crate: foyer)

foyer is an efficient hybrid cache library combining in-memory and disk-backed cache. It emphasizes high concurrency, zero-copy in-memory caching, and pluggable algorithms, aiming for performance under heavy load.

Key features/benefits:

  • Hybrid in-memory + disk cache for larger working sets.
  • Zero-copy in-memory abstraction for better performance.
  • Concurrency-focused design with thread-safe primitives.
  • Pluggable algorithms and built-in observability integrations.

Docs: https://github.com/foyer-rs/foyer