Explore the next frontier of data

Read the latest news and opinions from our experts

 

Featured Post

Recent Posts

The Future of Analytics: In Conversation With Matt Fuller

 

Product Blog

Datanova is just next week. More than 2,000 data and analytics leaders will join us to learn more about how to unlock the value of data with Starburst. As part of the two-day virtual conference, members of our product team will provide an overview of our product vision and direction. I sat down with Matt Fuller, a Starburst Co-Founder and the VP of Product, to discuss the current state of analytics and the future of the space.

Good afternoon Matt, thanks for joining us.

As companies have shifted massively to the cloud in the past decade, could you give us your opinion on the current state of analytics in the cloud and the challenges companies face?

Hi Mike, and welcome to the team. It’s an exciting time to be joining Starburst. 

From the start of the development of the data economy, organizations have leveraged many different solutions to solve the various problems in their businesses. Marketing may be leveraging a data lake to address one area while the Finance team might connecting to a data warehouse or another technology to solve their issues. New technologies have also led to investments in databases, data lakes, object stores and more. In addition, companies have seen the move to the cloud as a necessary step to reduce costs and provide broader access to data across an organization. However, all of these developments come with both monetary and time commitments. These developments also have the potential to silo data off from parts of the organization.

There’s a growing movement to rethink how organizations access, manage, and consume data. How is Starburst approaching the ‘Data Democratization’ movement?

Starburst was built upon the idea of providing a single point of secure access to all your data. Starburst provides you with the tools to connect to all of your underlying data sources and access the data directly. Starburst is based on Trino  (formerly known as PrestoSQL), a high performance MPP SQL-based query engine. We provide you with the means to access all of your data using the tools you already know and love.

In the next 12 months, how do you see Starburst opening the door for more parts of the organization to leverage the power of Trino?

There are many challenges for data consumers today. You might need to wait months for data engineers in the organization to create a data pipeline to fulfill their new requests. Cost constraints may limit their ability to move to new faster technology. And knowledge barriers may impose a significant hurdle in enabling everyone in the organization to access new and better insights. Creating a simplified way to access a wide variety of data sources is our goal. We’re working on a number of new initiatives to focus on data consumers and enabling easier access to all of the organization’s data.

Speaking about opening new areas of the organization to Starburst and Trino: Where do you see connectivity on the platform moving towards in the next 12 months?

We’re always adding more connectors. Right now, we see, as a priority, working on more connectors for SaaS applications such as Salesforce. There is a lot of data hidden or locked behind SaaS applications that make it very difficult to run analytics on without complex pipelines and even manual steps to extract the data. But providing connectivity to these data sources, we enable one to do analytics on data in time that was not previously possible. 

In the past couple LTS releases of Starburst Enterprise, you’ve added capabilities for dynamic filtering and aggregate pushdown, primarily in RDBMS connectors. Is Starburst  looking to extend this to all/other connectors? 

We are always improving performance. Yes, recently we added more SQL pushdown for all our relational database connectors. By pushing more SQL down, you can eliminate unnecessary data movement, thus improving performance. The concept of dynamic filtering is a fancy way of saying we can create additional filterings during query processing based on the data we see. This allows us to filter away unnecessary data earlier in the processing, greatly improving the performance of queries with joins across different data sources. Yes, we’re continuing to extend this to other relevant connectors. 

There are’s also challenges to opening broader access to data within an organization. Recently, Starburst has partnered with Immuta, Privacera and Okera. How important is secure data access when building new features into the products?

Out of the box, Starburst provides security capabilities such as RBAC and ABAC, column level security, row level filtering, and data masking. At our core, we take security very seriously. From the start, we built our product with enterprise security features in mind. However, we recognize you may have additional data governance requirements that a purpose built product from our partners. Therefore we have integrations with the many of the leaders in this space. 

What are some of the challenges surrounding deploying Trino in the cloud?

Running infrastructure software at scale that is reliable and performs well is challenging. There’s the high cost of administration from provisioning and monitoring to scaling and optimizing. As well as managing costs. Keeping up to date with the latest version of Trino and accessing the latest features available. Ensuring the right users have the proper permissions is mission critical in an organization. You may read blogs about large technology companies who have been very successful at it. But keep in mind many of these companies have virtually unlimited resources when it comes to managing and maintaining their Trino clusters.

Any spoilers for what’s coming at Datanova?

There are’s a lot of exciting things in development. Our ultimate goal is to bring the power of Starburst to a broader audience. Empowering data consumers with a single point of access to all of their data will help organizations make better and faster decisions. We’re very excited to share what we’re working on with everyone at Datanova. I guess you’ll need to attend to see for yourself. 

Thank you so much for your time Matt.

You can find out more about our product direction and vision next week at Datanova on February 9th and 10th. Our “When Worlds Collide: Starburst Product Vision” session will provide an overview of Starburst’s product direction on February 9th. Also look forward to our new product announcement session on February 10th.

 

Mike Marolda

Product Marketing Manager

New call-to-action

Your Comments :

data-mesh-email-signature

From Facebook

Read more of what you like.