eviction

Best-effort high availability for single-replica Kubernetes workloads: no downtime from announced node disruptions (drains, upgrades, consolidation, spot reclaims) without a standing second replica.

Pseudo-LRU implementation using 1-bit per entry and achieving Full-LRU performance.

Gracefully evict Kubernetes Pods before they get OOM killed

A Time-aware Least Recently Used cache implementation in Go with configurable eviction policy

Deterministic cache toolkit - typed keys, TTL windows and observable eviction

A high-performance, event-driven buffer library for Node.js and browsers, with configurable overflow strategies

"vcache" is a library that provides a concurrent-safe in-memory cache to store key-value pairs.

A CPU cache-line hash table with zero-cost eviction. 64-byte buckets, LFU+LRU hybrid, 16-shard concurrency, per-entry TTL. 6.5-12x faster than moka on 4-thread workloads.

A LRU cache for memory paging and content caching.

Kubernetes operator: scale spare capacity ahead of voluntary node drains, then gate pods/eviction until Ready

MaskAhead: KV Cache Selection & Eviction for Block Diffusion LMs

Thread-safe, fixed-size, in-memory LRU cache that supports eviction, expiration, and out-of-band updates

DIY Form tools for MOTenantHelp including Eviction Defense Document Engine (EDDE)

kubectl plugin to evict pods

An Easy Java Cache Library

Generic in-memory cache with TTL and size-based eviction — TypeScript-first, zero dependencies.

Eviction algorithm laboratory for RESP2-compatible KV stores.

High efficient embedded Cache library for Golang

70k+ evictions plotted with react-map-gl sourced from nyc open data

REAL: REtrieval-reAsoning and Logic-constructed Attention Behaviors for Long-Context KV Cache Compression