A platform for applications that need low-latency computation on large data sets. It stores and indexes your structured, text, and vector data so that queries, selections, and processing as well as machine learning model inference can be executed quickly at any scale within service time. Functionality can be customized and extended using application components hosted in Vespa.