Patrick Toole & Christopher Rocca - BigDataNYC 2013 - theCUBE - #BigDataNYC
summary
Hadapt presents a relational abstraction layer that lets analysts run SQL on HDFS without writing MapReduce or Hive jobs, pushing queries down to data nodes for wide parallelism and on‑disk scale rather than in‑memory limits. The platform natively queries semi‑structured clickstream and key–value blobs, reduces ETL by exposing flexible schema, and embeds machine‑learning routines callable from SQL. Early adopters include advertising, online retail and financial services, and deployments report practical unification of analytics stacks.
AI-written summary · about
