What inspired you to go with Starburst?

The big reasons that inspired us to go with Starburst were data access, security, performance, and cost savings. Starburst provides our end-users the capability to access all the data and democratize it in a simple and easy fashion. Data democratization is very important for us and we wanted to provide fast data access while also maintaining security. Our users need to be able to access our raw data in order to explore it and use it in new and innovative ways. Starburst has been a huge enabler in democratizing our data, leading us closer to our goal of achieving a self-service infrastructure.

Where does Starburst fit in your overall data strategy?

Starburst has been integral to our architecture, serving as our sole tool for all query patterns and data access across our enterprise, providing us the capability to secure our data in the most organized and efficient fashion.

Which critical technologies are you integrating with Starburst?

We integrate Tableau, Google Big Query, Google Cloud Storage, Cloud SQL, and Auto-cloud ABMS with Starburst. By integrating with Tableau, our users have the ability to access our raw data and BI reports, which opens up new business capabilities, recommendation capabilities, and adds more value to the line of business.

What are your three favorite Starburst features?

My three favorite features are:

• Federated Data: The ability to tap into multiple data sources from one point of access is huge
• Security: We’re able to apply fine-grained access control to unstructured data sets
• Semantic Layer: Our users can seamlessly use the data without worrying about where it is coming from

Which business outcomes are you most proud of?

We are really proud of our ability to tap into the unknown, explore data that we have not been able to explore previously, and come up with recommendations that will change the dimensions of how we are doing business and how we are providing capabilities to our customers.

In the past, most of our data processing was driven by a traditional ETL approach. So, by the time we arrived at a decision, a day or two would have passed since the time we asked the questions of our data. With Starburst, we can access data sets directly and make sense of the data in near real-time. This gives us the ability to make changes if required along the way and reach decisions faster. For example, the historical data of our customers is easily accessed and integrated into our mainstream pipelines, which provides a more personalized booking experience and more accurate recommendations for our customers.

What success have you been able to measure?

We’ve definitely accelerated time-to-insight with Starburst. Plus, now we can see the current happenings of our organization and make more informed business decisions. We’ve also realized a lot of cost savings. We are reducing our storage costs by 5-10X by leaving data in its unstructured, raw form, and accessing it through Starburst.

Do you think decentralized data access and Data Mesh are the future of analytics at scale?

Absolutely, having a centralized system used to work well in the past, but a decentralized data inventory allows the data to be where the users are which gives them the capability to explore the unknown.

Data Mesh allows us to cater to both our data producers and our data consumers. It provides for domain-driven ownership for the producers while also providing a unified level of access for our consumers. Our biggest focus in incorporating a Data Mesh is migrating all our data from on-prem to the cloud, and, in the future, we will build out our capabilities to multi-cloud.