vector_similarity_search complements cache

Source: Pattern·Target: Pattern

Summary

Vector search results for common queries can be cached with cache-aside to avoid repeated ANN index lookups for the same semantic query, trading some recall freshness for significant latency improvement.

Evidence

·vector_similarity_search lists cache_aside in related_patterns; embedding generation adds latency at query time (patterns/vector_similarity_search.yaml)
·cache_aside applicable_when read-to-write ratio is high and data is accessed repeatedly (patterns/cache_aside.yaml)

Evidence grounding

Grounded, 2 supporting items

Caching hot ANN queries avoids repeated embedding-generation and index-traversal cost.