The Hive think tank recently held a panel discussion on Stream Processing Systems. Panelists debated best approaches to messaging platforms, stream systems, and accompanying data stores. Other topics included scalability, latency, fault-tolerance/reliability/High Availability, ease and simplicity of deployment, maturity and popularity, enterprise features, and security. There was a lot covered!

Ben Lorica, O’Reilly Media’s Chief Data Scientist, moderated the conversation, comprised of the who’s who of real-time data systems


  • Karthik Ramasamy, Engineering Manager, Twitter and Apache Heron
  • M. C. Srivas, CTO/co-founder, MapR
  • Nikita Shamgunov, CTO/co-founder, SingleStore
  • Ram Sriharsha, Architect for Spark and Data Science, Hortonworks
  • Jay Kreps, CEO/Co-founder, Confluent

Stream Processing Questions Answered

  • Does Spark streaming have more improvements than storm?
  • Why does Twitter have its own messaging system?
  • What are customer concerns about latency and streaming?
  • What changes are happening with Kafka?
  • What are the overall upcoming trends in the stream processing space?
  • When it comes to security systems, how do you prevent specific data access to consumers?

Watch the Panel Recordingwatch-the-panel-recording

About the Hive Think Tankabout-the-hive-think-tank

The Hive, founded by T. M. Ravi and Sumant Mandal is an incubation studio focused on data driven applications in enterprise, IoT and on-line segments. The Hive works actively with founders to help incubate, fund and launch data-driven startups. The Hive Think Tank is the largest community of data experts with over 8000 ‘databees’. It brings together a unique community of entrepreneurs, business and technology practitioners who are focused on the use of Big Data approaches.