September 28th – 30th 2015 saw the Apache Big Data Conference in Budapest. The setting was the palatial Corinthia Hotel in the centre of the city with events split between keynote speeches, breakout events, and "BarCamp" sessions where all comers were welcome to suggest a topic.
It was no suprise to see the big hitters: Apache Hadoop, Apache Spark, and Apache Kafka making a big impact at the IBM-sponsored show. If featured real-life use-cases and innovative combinations of technologies used in tandem to solve Big Data problems. There was also room to showcase Apache incubator projects such as Geode (a distributed, in-memory database) and Zeppelin (a web-based notebook for analytics) jostling for position with the established projects.
Amongst the discussion of offline analytics, there was still room for the operational data stores with one afternoon dedicated to Apache CouchDB. Firstly Benjamin Young (a.k.a bigbluehat) espoused the benefits of offline-first storage to service application data locally and sync with the cloud at a later date.
IBM Cloud Data Services' Mike Wallace then talked about the 'awkward bits' that crop up when using CouchDB and encouraged developers to join the project and become committers:
— Glynn Bird (@glynn_bird) September 29, 2015
The last talk of the afternoon was originally scheduled to be Hands on CouchDB 2.0 by Joan Touzet, the Apache PMC member and CouchDB Developer, but she couldn't make it. It took 3 people to replace her. I talked about some of the new features coming in CouchDB 2.0 (clustering, MongoDB-style queries, etc), Mike Wallace lifted the lid on the Erlang internals, and Michelle Phung showcased the all-new CouchDB Dashboard – a huge improvement over the 1.x version.
For an event recap, visit the official website.