# twitter/fatcache

Memcache on SSD

Repository: https://github.com/twitter/fatcache
Canonical: https://ross.abutalabs.com/products/fatcache
Language: C
License: Apache-2.0
License Family: permissive
Archived: true
Last push: 2021-11-01T18:32:51+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 4991, "days_push": 1766, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1299, forks 173 (observed 2026-08-28T04:04:17.394126+00:00)

## What it is
Fatcache is a networked, memcached-protocol-compatible cache that uses SSD storage as an extension of DRAM, allowing terabyte-scale caching at lower cost. It batches small writes to avoid SSD write amplification and is now retired in favor of Twitter's Pelikan.

## Use cases
- cache large datasets that don't fit in memory
- reduce DRAM cost for terabyte-scale memcached workloads
- run a memcached-compatible SSD-backed cache
- maintain high cache hit rates for big data workloads
- extend an in-memory cache tier with SSD capacity

## When to choose
- you need a memcached-protocol cache larger than DRAM allows and can accept slightly higher latency
- you want lower cost per cached byte than pure in-memory caching

## When to avoid
- you need actively maintained software - the project is retired
- your working set fits comfortably in RAM
- you need persistence - fatcache is still a volatile cache
- consider Pelikan or modern alternatives instead

## Facets
- artifact type: service
- maturity: abandoned
- function: caching, database
- domain: databases, performance, backend, big-data
- platform: c, self-hosted
- tags: memcached-protocol, ssd-cache, distributed-cache, key-value-store, retired, linux

## Member repositories
- twitter/fatcache (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:17.394126+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:53:28.073577+00:00, confidence not recorded.
  - readme: https://github.com/twitter/fatcache (fetched 2026-08-28T04:04:17.394126+00:00, sha 9b9807c4eb28)
- Data as of 2026-08-30T08:39:29.467469+00:00.