AWS has added vector, full-text and hybrid search, plus server-side aggregations, to ElastiCache Serverless for Valkey. The company says vector searches can reach up to 99% recall with latency as low as microseconds, and that updates can become searchable in as little as microseconds after writes complete.
The feature is available on Valkey 9 in commercial AWS Regions, AWS GovCloud (US) Regions and the AWS European Sovereign Cloud Region. Customers can create a new serverless cache or upgrade an existing one. AWS says the service is billed based on use and needs no infrastructure management.
Why it matters
Teams building search or retrieval features can query recently written data without managing search infrastructure, though AWS’s stated latency and recall figures are company claims.
