What inspired you to go with Starburst?

With our data so spread out, it was proving to be a lengthy task to access it. Our ETL pipeline and deployment process was averaging a two-month timeline just to get data to our end users, which was inefficient and time-consuming, holding us back from insights that could drive our business forward.

We realized that it would be far more efficient to embrace a decentralized Data Mesh architecture through Starburst as its abstraction layer. As a result, we were able to give our end users the flexibility they needed by allowing them to access all data from one single point, thereby reducing our time-to-insight from months to minutes. With Starburst, data users can access the data faster and more efficiently. No other product in the market has the capabilities that Starburst has!

Where does Starburst fit in your overall data strategy? Which critical technologies are you integrating with Starburst?

Starburst is an essential part of our overarching Data Mesh architecture that gives our users the much needed flexibility to access data through a single point rather than having to go around to ten different data sources. Plus, when we can give the right access to business users, it helps Sky accelerate its overall business goals.

We integrate Starburst with all of our technologies, including GCP, AWS, and Tableau. By doing this, if users have a query that they need to perform, they can access all the data and information they need without having to dig through all of our technologies. Starburst integrates particularly well with Tableau and helps us run more accurate business reports while also providing faster access to data.

What are your three favorite Starburst features?

My three favorite features are:

• Caching tables: We cache tables from our legacy system in the cluster, so that it’s possible to query those tables in Starburst itself
• Security and Apache Ranger: We can easily set policies for individual users and filter access at the table, column, or row level
• Integration with BI tools like Tableau: Users gain fast access to new data sources without having to learn a new tool or wait for ETL

Which business outcomes are you most proud of?

With Starburst, we are able to do things that we couldn’t even think of before. Our time-to-insight has been drastically reduced. We have gone from months to minutes in terms of querying and insights. In the future, with Starburst Stargate, we also hope to leverage the cluster capability to make accessing data across borders seamless while remaining compliant with data governance regulations.

What success have you been able to measure?

When we initially began using Starburst, we had a series of metrics on which we measured the success of the application. These were connecting to different data sources on-premise and on cloud, accessing data quickly, and being able to have fine-grained security access. Starburst met all of these metrics and was an immediate success.

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

Decentralized access is definitely the future. We have learned from the past that a centralized data approach just doesn’t work. It is expensive and it is difficult to manage the constant copying and moving of data that is required.

We are currently in the process of creating data products, which Starburst is really helping with. Previously, without a single point of secure data access, creating a data product was not possible. With the abstraction layer that Starburst provides across different data sources, it has become our analytics engine for the Data Mesh.