Leadership in Enterprise Data Mesh with Jack Steele from Zetaris

Listen to this episode on Anchor FM

Are you struggling with your data centralization journey? This episode of the DATAcated show will help with ideas to move forward from data silos. Host Kate Strachnyi talks with Jack Steele, Chief Product and Community Officer at Zetaris - The Networked Data Platform. They discuss the changes in the data industry and how data is used. Listen to hear how organizations can build on what they already have and do something different.

You will want to hear this episode if you are interested in...

  • Jack’s 20-year data industry career [02:26]
  • What is data mesh? [06:07]
  • No need to start again [11:32]
  • Self-serve data platform [18:57]
  • Implementing data mesh [21:57]
  • The future of data mesh [26:33]
  • Implementing Zetaris [31:32]
  • The effect on cost [36:38]

Connecting data silos

Jack has been in the data industry for 20 years. Most of his time has been spent in large companies trying to find ways to bring data together using classical data warehousing architectures. Six months ago, Jeff joined Zetaris to lead its product stream. Zetaris takes an entirely different approach by connecting to the source of data. As a virtualization-first platform, which enables the company to move incredibly quickly. They made possible cases that had been too expensive or slow through the centralization journey. Instead of starting from scratch, Zetaris accesses existing data silos and connects them.

Zetaris was founded about five years ago based on the idea that data needs to be accessed where it lives. The company pivoted from using previous deep technologies to using Spark with Apache Ignite. Bringing together those two open-source technologies meant that suddenly small companies could do things that large companies were struggling to do.

Data accountability

Many companies have been on their data centralization journeys for a long time but are at the point where data lakes haven’t delivered. Data mesh is the complete opposite of data centralization and is all about building data products in a federated way. It’s more than just a technology approach; it’s a way of managing and running an organization to enable data value and enable data owners to own their data products. In addition to the technology side, businesses need to transform their organizational approaches to make the most of the data mesh.

Some of the challenges of data mesh are being helped by other things in large organizations. Agile and bringing multi-disciplinary teams together is similar to the idea of moving ownership of data into each operating unit, which ensures accountability is in place. Before this, the Chief Data Officer was in the middle of the organization, responsible for data but accountable for none of it. The big problem was that the centralized team somehow had to be accountable for making sure reports were running correctly and that all the data was exactly right, yet the entire organization was generating it.

Scaling with data

The most important thing to explain to CEOs is that they don’t need to start again. Data mesh isn’t meant to replace anything. Instead, it networks previous work to drive more value while adding the extras like data governance. Businesses have to pivot quickly and can’t afford to wait for traditional approaches. Waiting months for a new data set to be brought in, curated, and added isn’t realistic to keep up with the speed of how businesses are moving. Data mesh reduces friction and helps businesses with their data needs to achieve their scaling goals.

Resources & People Mentioned

Connect with Jack Steele

Connect with DATAcated

Subscribe to the DATACATED On Air podcast