Thought Leaders in Big Data: Zetaris CEO Vinay Samuel (Part 1)

A very interesting innovation in data virtualization.

Sramana Mitra: Let’s start by introducing our audience to yourself as well as your company.

Vinay Samuel: I’m the CEO and Founder of Zetaris. We are changing the way data analysis is done in large enterprises. The traditional model is to collect data from different source systems across the enterprise and the internet and bring it into a data warehouse or data lake.

What Zetaris has done is change that model. Rather than bringing data to a central point, why not publish a query to wherever the data is and then join the answers to those in real-time to enable what we call a virtual data warehouse or a virtual data lake? If you don’t have to centralize data, you take away months, if not years, in the preparation time for analytical projects.

You greatly reduce the data quality issues that are created when you do a typical data lake or data warehouse project. You bring down the the cost of migrating data either to the cloud or data center. There are huge benefits in the ability to do a decentralized data platform which is what Zetaris is all about.

Sramana Mitra: You interface with the traditional data repositories.

Vinay Samuel: Absolutely. One of the first things we had to do when we set up our network to data platform is to connect to anything. We connect to streams. We connect to data warehouses. We connect to old mainframe technologies. We connect to everything in your data landscape whether it’s inside or outside your company. All of a sudden, you get this single view based on what we call our virtual data pipelining and AI for data conforming or data joining.

Sramana Mitra: Let’s double-click down and do some use cases. Where are you finding traction?

Vinay Samuel: There’s the general use case where customers have data in their data centers. Large customers might have data warehouses or data lakes and they also have data in their cloud data warehouse. They might have data in Snowflake. We have customers who have data in every cloud. They have a full multi-cloud scenario.

The problem is to give the business access to that data in a joined view and enable the business to self-serve and query the data for analytics and machine learning. The general use case is to enable the business at the lowest cost and in the fastest time. That’s across every industry. We just call that use case creating a virtual data warehouse.

I’ll give you some industry use cases. I’ll talk about a large health research company. They’re one of the largest in the world. They’ve got data in their billing systems, clinician systems, and their research systems. That data is scattered across their data centers. Some of it is with third parties.

The problem is the time it takes for a researcher to get access to clean, conformed, and joined data was prohibitive. It would take weeks in some cases and months in most cases to bring their data together. That would involve the traditional ETL or data cleansing. We overlaid it on the existing systems. We connected to all of their legacy systems.

Very quickly, we created a single query-able view of that data. In the old days, we used to talk about the logical data model. We basically create a logical data model on top of their physical data structures that’s, all of a sudden query-able. The speed of their research and data preparation made a lot more cycles in their research process leading to better, deeper, and more accurate outcomes, ultimately, saving lives and creating a much better scenario. That’s in the health industry.

This segment is part 1 in the series : Thought Leaders in Big Data: Zetaris CEO Vinay Samuel
1 2 3