One of last month’s interesting news items was Yahoo spinning off its Hadoop unit (as Hortonworks). Colleague Ashwin Jayaprakash pointed out that the in-thing in startups nowadays is to offer, usually cloud-based, event and data processing services based on Hadoop or similar principles: Datameer, HStreaming, Kitenga, Dataspora, Cloudera, Goldenorb…
One observation we are noting at TIBCO: many of the “big” use cases for complex event processing (using TIBCO BusinessEvents) is processing (big) amounts of event data in (near) real-time. For sure, data warehouses have their uses – especially for running batches of predictive analytic operations to find new outliers of performance. But its the real-time business processing (and real-time decisions) that seem to have the biggest impact.



It seems “big data” is the buzzword of 2011… some other relevant comments from:
- Opher Etzion (IBM R&D) – also commenting on Twitter’s Storm platform (targeting real-time but not yet hooked into any particular continuous query library)
- Mark Palmer (Streambase) – commented on the technologies for real-time big-data in finance, striking quite a few acords with comments on this blog in recent times (eg analytics, case management etc).