Red Hat's Greg Kleiman, director of storage and big data, meets up at Strata+Hadoop World 2015 with Brandon Draeger, director of marketing and strategy, big data solutions at Intel, and David Tishgart, director of marketing at Cloudera to discuss what interesting new trends they are seeing at the show. One revelation in particular will be most surprising...see the video
Greg: What are you hearing at the show?
Brandon: This morning there were some exciting keynotes, including more customer stories. It was less about pure technology talks and more about how big data is changing business and people’s lives.
David: I’d echo what Brandon said. And an interesting change from years past is that people aren’t asking what Cloudera or Hadoop is, they know who we are, and we can finally talk about the use case and the value we provide for customers instead of going over the technologies.
Greg: We’re seeing a lot more customers struggling with integrating their data. We just announced our JBoss data virtualization connecter with Impala, on Cloudera. Are you hearing the same thing about data integration?
David: Data integration is key, right? Because at the end of the day what customers want is get value from the data. Analytics, analytics with high concurrency, get data out to business users quicker. But it all starts with integration. How do I get data out of my existing platform into Hadoop?
Brandon: I think this past year has been a step forward in enterprise readiness in Hadoop. We’re seeing a lot of the obstacles begin to come down. We’re hearing three categories of issues — rapid performance and scalability, manageability, security. I think we’ve taken quantum leaps around these spaces. It’s a maturity issue, that customers are asking about Hadoop performance shows they are on the maturity curve.
Greg: The other trend we saw this morning — we’re seeing big data getting moved to the cloud. We just released OpenStack 6 with Sahara infrastructure. Are you hearing much about cloud? Are customers asking about that?
David: Funny, our name is Cloudera, we are living up to that cloud promise. It’s gonna be a big year for us! I think the reason for that is, the way we like to talk about it, data will live where it is formed. It doesn’t make sense for this sensor data/high ingestion/streaming stuff to be all come down into your data center and be stored there. It’s really expensive, they’re taking their data out and keeping it there and bringing the high volume Hadoop workloads into the clouds where the data lives, instead of bringing the data to you. It’s about limiting the data movement and doing the workloads there, and its more important that we’re optimized for the cloud.
Brandon: We’re seeing customers embracing cloud as an operational model run on-premise deployments as well. Hadoop is a workload on OpenStack and the work you’re doing on Sahara fully supports a lot of that.
Greg: Thank you both for your time!