Cloudera COO advises Australian and Asia Pacific enterprises – Don’t hesitate, take the Big Data plunge

Cloudera's Chief Operating Officer (COO) Kirk Dunn was in Sydney, Australia last week. BigInsights Chief Executive Raj Dalal had an interesting conversation with the man on what Cloudera is about, where Big Data was headed, and the key takeaways for professionals and enterprises in Australia and in the whole of the Asia Pacific belt.

Kirk was of the view that enterprises or professionals should hesitate no more to take the Big Data plunge. His advice, especially to those located in Australia and the Asia Pacific region, was that, yes, there's a lot of noise in the market which may add to the hesitation; yet, enterprises can start extremely small, get a good insight, and understand where they are going to be able to leverage upon in their operations before taking full advantage of Cloudera and Hadoop and Big Data.

"My recommendation is - go look at the annual report of your organisation, find out what the CEO or the top guy has said, the 3 to 4 key things for that year vis-a-vis the company, and then apply this new tool to those opportunities. Start small; it is a very organic system - you can start with ten nodes and you can get a very good insight," he explained.

Pointing out that Cloudera was essentially designed to be a software company, the COO said the company's energy came from two sources:  from the base of Apache Hadoop system, which is the core of what Big Data is based on. The 2nd, now equally important if not more revolutionary, the way software companies are going to be built in the future, is how Cloudera was going to add value on top of the open source core. "So we are going to build products that actually allow organisations to gain insights from asking questions they have always been wanting to ask but have never been able to. Features that we will build on include those that allow companies to do data backup and disaster recovery. We will also enable them to do data lineage to allow them to find out where the data has been, who has touched it, what actions have been put on it, and carry that on for all of time. Other areas like running real time SQL on top of Hadoop...and then a variety of other management capabilities, so that when someone is running a Map Reduce job or search on top of the platform, they can track what's going on and make sure that resources are being allocated," the COO said. "So it is really a two pronged, hybrid approach  - the first, very, very clear on the open source side, the second, on the value-added side software that people then pay us for but adds value to the operations."

Kirk also spoke about the recent launch of the Enterprise Data Hub product at the Strata Conference + Hadoop World 2013 in New York. This product, he said, did away with the notion that all your data was going to reside in one location and then you will want to put different kinds of data analytics engines on top of that platform.

"So, you do not want to have multiple capabilities. Instead, you want to have one hub where you an execute different kinds of analytical capabilities. That's a big area where we think things are going to grow, especially in certain verticals like telecom,  banking or healthcare. So apps are written around and on top of the platform for companies to decrease costs and increase their topline.