Driving the On-Demand Economy with Predictive Analytics
summary
Announces MemSQL 5 GA with a new LLVM-based execution engine that claims 10x faster than the prior release, plus native geospatial SQL for zoomable queries. Introduces an open-source pipeline tool called Streamliner that subscribes to Kafka topics, applies Python/Scala transformations or Spark MLlib models in-memory, streams scores and raw sensor data into MemSQL, and triggers business actions. Shows real-world scale: 200,000 wind turbines producing about 2M updates/sec, and describes persistent scoring, rapid ingestion, and real-time BI.
AI-written summary · about
chapters (4)
related talks
09mHow to be fast and have a good memory with Machine Learning
06mImplementing custom models for machine learning
05mTechnical convergence in enterprise database software
05mClose Encounters with the Third Kind of Database
05mTeaching databases to learn in the world of AI
06mThe Enterprise-Focused Startup
12mPower a Better Future
06mMachines and the magic of fast learning