Experiences Running Spark in Production at Quantifind
Erich Nachbar, CTO of Quantifind, will talk about their experiences running Spark in production. Spark is currently used for batch processing of unstructured text (entity extraction, language detection, indexing, Tweet & topic classification) and realtime, in-memory data retrieval (statistical & metrics computations) exposed through REST. Using Spark in production can enable unique product features like driving complex, on-demand computations in realtime. After a short demo, we will explore architectural and operational aspects of such a system as currently running at Quantifind.
About: Erich Nachbar is CTO of Quantifind. Quantifind is analyzing unstructured user content like comments and product reviews to predict product revenue and intentful audience demographics for major companies in the entertainment industry. He specializes in distributed data storage and processing using modern technologies like Spark, Storm, Scala, Kafka and Cassandra.
Pizza and beer will be served at 6:30pm. Talk will start at 7PM. Thanks to Groupon for hosting this meetup.