F3 Lightning: How Meta Accelerates Ads ML Features to Production at Scale
Meta's Facebook Feature Framework (F3) compiles ads ranking features into one merged DAG so shared work runs once, but that graph was also the release unit, so one bad definition blocked every other feature as the catalog grew past 20,000 features per request. F3 Lightning makes each feature an independently compiled, validated, deployed, and rolled-back configuration, then rebuilds shared execution at service startup: stable identifiers derived from each computation's meaning, one flattened execution plan that deduplicates common nodes, a single per-request memory buffer, and dense integer slot mapping so request-path reads are direct array indexes. Comparable extraction features now reach production about 3.6x faster than traditional F3 with no increase in infrastructure compute cost.
- Published
- 2026-11-03
- Scale
- 20,000+ features per ads ranking request, ~3.6x faster feature releases to production, no increase in infrastructure compute cost
- Techniques
- per-feature release shared execution plan stable computation identifiers shared memory buffer dense index mapping
- Stack
- F3 DAG feature graphs Ads ML