During the last day of Strata Conf 2014, theCUBE co-hosts John Furrier and Jeff Kelly interviewed Ankur Gupta, General Manager with MetaScale, talking about the evolution of the industry in general and the evolution of MetaScale in particular.

“Since the last time we spoke about Metascale, we announced our Hadoop appliance. We saw that the market needed a ready-to-go Hadoop appliance that could come with fully-managed services and would be implemented quickly, so that the companies would not have to go through the pain of putting the hardware together, understanding which Hadoop software/distribution they need to use, and then implement that hardware in their data center,” said Gupta.

“What is the big trend that’s lifting up your value proposition?” asked Furrier.

“We’ve identified several types of customers: 1) the companies in the early phase of Hadoop, which are exploring their Big Data strategies. Here our experience comes really handy. We help companies with their strategies and with developing their big data centers of excellence, how to use Hadoop and build use-cases around it; 2) then there are the advance stage customers who already have Hadoop infrastructure and are now interested in putting big data science on top of Hadoop, utilizing it to manage their data better, getting more out of the different sources data coming to them; 3) companies that started their Hadoop journey because their internal IT staff said they could do it but who, after some investigations, don’t know what to do about it,” explained Gupta.

One of the reports published by Wikibon singled out that one of the reasons companies fail in implementing Hadoop is because they don’t have talent or they don’t have use-cases or an efficient use of their infrastructure, noted Gupta.

Deploying Hadoop, Big Data

 .

“What are the verticals you see?” asked Furrier.

“Because we have a retailing background, it sometimes gets very hard for us to work with other retailers. They see value in what we bring to the table, but they might not feel comfortable working with a company originating from another retailer,” confessed Gupta. “We see a lot of traction in healthcare – managing the data and making sense of it seems more important for them. The financial sector uses Hadoop not only to manage data but for all the batch processes, to reduce their expensive EDW footprints. That’s the demand we see in the financial sector.”

“With all these platform wars going on, and you having a platform as well, what does it take to have a successful platform?” asked Furrier. “What are some of the myths regarding the way the customers think the platforms should look like?”

“This is one big problem we saw in the market, which is why we launched our Hadoop appliance. There’s a differenciation between a BU (Business Unit), who wants to use Hadoop to make sense out of the data, and a Central IT who is busy with the daily tasks. They are not always building the Hadoop infrastructure. BU wants it but they don’t have the IT to manage it, while Central IT doesn’t necessarily have time or resources to build it.

“This is when our Hadoop appliance comes into play. It’s a ready-to-go Hadoop infrastructure with three main benefits: 1) it comes with our own reference architecture and our knowledge of what has been successful in the market, while managing Hadoop in large enterprises; 2) it’s a complete 24/7 managed services. A company doesn’t need to worry about taking care of the infrastructure or about buying or hiring resources to manage the infrastructure. It’s our job to fully manage it; 3) it’s as easy to use and deploy as plug and play.”

• The appliance model for Hadoop

Kelly wanted to focus more on the idea of an appliance model in the Hadoop landscape. When you think about Hadoop you think about scale-out commodity boxes, when thinking about applicance people think about exadata first.

“How does that appliance work in the Hadoop paradigm?” asked Kelly.

“We define our appliance as a box, but a very flexible one. It comes with our own reference architecture and the hardware we think it’s the best, but from there on it’s very flexible. Our appliance doesn’t necessarily come with our distribution; we give customers choice to go either with Hortonworks, Cloudera, Apache or we could recommend one for their use-cases,” replied Gupta.

Apart form the flexibility, Gupta highlighted the easiness of the appliance. The expansion of the infrastructure is not as difficult as it used to be.

Advantages of MetaScale appliance:

 .

• distribution neutral
• management neutral
• capacity neutral

 .

Big Data : Who needs it?

 .

Picking up on the stress mentioned by Gupta between the IT and the business side, Kelly wanted to know who the main customer of MetaScale is: the business person or the IT person?

“One way or another we end up working with an IT person, which is good. When we are selling Hadoop and big data services, the user is primarily someone from business. At the end of the day they are looking to get value out of their investment; the value is related to the use-cases – either making better sense of data or better managing the data. Very often we work directly with IT, who are either trying to reduce their EDW footprint, or they’re looking for a more efficient way to manage their existing EDWs.”

As the industry is evolving, it’s a great side to see the business driving more of the conversation, stated Kelly. Up until now the market was too focused on the tech. “What are your plans for this year?” he asked.

“We grew up 200 percent in 2012 and 2013 and we expect a similar growth this year. We see a lot of demand and we are looking to expand. As Hadoop becomes more and more mainstream, the companies embracing it will become more self-sufficient. There will be more tools. We see a lot of growth in the professional market. The market is becoming so crowded. We advise our clients to think long term, as the data is only going to get bigger. We try to prepare them for the upcoming digital storm,” stated Gupta.

“What’s going on around Big Data? What’s most newsworthy?” inquired Furrier.

“Data is becoming even bigger. The buzz word today is ‘enterprise data hub’ and I think in the future we’ll see some interesting results from the medical and the financial fields,” predicted Gupta.