- ABOUT
- VIDEOS
- TOPICS
- TESTIMONIALS
- COMMUNITY
- EVENTS
- NOMINATIONS & LISTS
- SPONSORSHIP
- ABOUT
- VIDEOS
- TOPICS
- TESTIMONIALS
- COMMUNITY
- EVENTS
- NOMINATIONS & LISTS
- SPONSORSHIP
(US and Canada) Ricardo Crepaldi, Director of Business Intelligence, Big Data and Integration, BASF Germany talks with Dr. Michael Stonebraker, Founder, CTO and Visionary, Hopara, and Ken Smith, CEO, Co-Founder Hopara.
Dr. Michael Stonebraker, Founder, CTO and Visionary, Hopara shares that the major disruptor in the industry is the cloud. He thinks everyone should have a plan to move everything they possibly can to the cloud as quickly as they can. Why? Because it’s cheaper. He shares a quick vignette to support his statement about why it's cheaper: Microsoft Azure's data centers are shipping containers in parking lots, in places with only cheap power walls and roofs for security. So, if one compares that with the raised-floor data centers in Berlin or Frankfurt, they are simply cheaper. They also offer a resource on demand: pay for what you need and don't pay when you don't use it. Hence, that’s almost guaranteed to be cheaper. He shares that the cloud will become even cheaper over time, and that is why he is a fan.
Stonebreaker believes decision support should move aggressively and as quickly as possible. He shares that the OLTP will take longer, but it should move if it can or it will be weighed down by internal legacy systems.
The second disruptor, according to Stonebraker, is the data-driven enterprise. He believes CEOs are waking up to the fact that data is the real value addition and everyone should be empowered to use it efficiently, quickly, and painlessly. He thinks making that happen will be a challenge, but it is imperative if enterprises are going to succeed at being data-driven.
He thinks a lot of CDOs have their feet nailed to the floor, so to speak. For example, the average data scientist spends between 80% and 90% of their time finding the data they want to analyze, integrating it, cleaning it, and empowering data scientists. That percentage needs to be knocked down, Stonebraker says. Realistically, data scientists spend Monday to Thursday and half of Friday doing monotonous work and spend less than half a day a week doing the job for which they are hired.
He shares further that enterprises will realize they are full of data silos — hundreds of them — and if they don't allow their CDOs to see into all of these data silos, they have no chance of figuring out the silo integration strategy to assist data scientists. Therefore, it's essential to empower the CDO to do a much more systematic view of data integration.
Stonebraker also shares that if you want to cross-sell between your various business units, you've got to integrate customer databases from the various silos. Many CDOs do not have read access to every bit of data in the enterprise, but that should be required.
Stonebraker believes the cloud is an opportunity to change many aspects of the enterprise business, but there has to be a global strategy, which should come from the CDO. And if you don't empower CDOs to do that, you'll be losing the opportunity to modernize. So, he concludes, enterprises need to beef up their CDOs’ job descriptions and give them access to every bit of enterprise data.