Last week, together with my colleagues Tünde and Kris, I had the honor of attending the Cloudera Technical Summit in Las Vegas. The Summit, held annually in Vegas, is accessible only to premier Cloudera partners and is the only event in the year where Cloudera partners get together with Clouderans and are treated as they were Clouderans (well, except for the color of the conference badge).
Key takeaways from Cloudera Tech Summit 2015
For obvious reasons, I cannot talk about the content of the sessions, but a few things struck me:
Bigger is better
Europe is still trailing behind the US with regard to Hadoop adoption. Companies are still diffident and do not embrace Big Data so quickly. Clusters tend to be smaller (but, hey, everything in Europe is smaller when compared to the U.S., particularly in Vegas) and this limits the ability to just collect data and make it bloom.
Integration of Hadoop into production applications
Integration of Hadoop into production applications is happening, and it is happening a lot, mostly because of products born out of the Cloudera foundry (think of Flume, Sqoop), but not only (Solr and Kafka come to mind).
Cloudera Search is hot
A proven technology such as Solr integrated with Cloudera gives engineers an edge when it comes to bringing search solutions into production.
Near real-time
Even though I mostly work in the financial sector, I learned that we can actually never say Real Time when speaking about Big Data in Finance. This is mostly due to latencies in all Big Data products and that the nature of financial transactions is very different from hard real-time guarantees. The recently published article about [Cloudera and Santander](http://blog.cloudera.com/blog/2015/08/inside-santanders-near-real-time-data-ingest-architecture/ ‘Inside Santander’s near real-time data ingest architecture’ ) corroborates this, as they always say Near Real Time.
Moving up
Cloudera gives a lot of thought to its product and strategic roadmap, which is leading to the adoption of its products by more and more enterprise clients.
Scaling up
Spark's future is bright but Impala's future (mostly thanks to projects such as Ibis) is just as bright if not brighter.
On top of that, the conference venue and all Clouderans were a blast: they not only are incredibly smart and talented but also incredibly accessible and helpful. Hopefully until next year.
Can’t wait until next year? Xebia will join Cloudera on the first-ever Cloudera Sessions in the Dutch capital Amsterdam on October 15th. Cloudera Sessions are the place to get up to speed with the latest developments and to hook up with like-minded data scientists.