Cloud-Native Database Architecture

If you follow the cloud database market closely, you are sure to hear the phase "cloud native" often. But what does it mean? And why should we care?

The Cloud Native Computing Foundation, an industry group that advances cloud-native computing through open source initiatives, defines cloud native as including "containers, service meshes, micro-services, immutable infrastructure, and declarative APIs." 

That's a high-level description of the software components, but I wanted to understand how cloud native applies to data warehouses. So I invited Mark Cusack, CTO of Yellowbrick Data, to join me on the Cloud Database Report Podcast. We talked about data architecture, Kubernetes, elasticity, and the separation of storage and compute, which Cusack described as "table stakes" to modern cloud data warehouses. 

• Listen to the Podcast: Yellowbrick CTO Mark Cusack - What is a Cloud-Native Data Warehouse?

Earlier this month, when Yellowbrick announced $75 million in Series C funding, CEO Neil Carson used the occasion to tout Yellowbrick's cloud-native architecture—and to ding competitor Snowflake. 

“Time and time again, customers are telling us that they can no longer tolerate the unbounded expense of consumption-based billing, especially vendors that force advanced purchase of ‘credits’ like Snowflake,” Carson said. “Customers much prefer our modern Kubernetes-based, truly cloud-native architecture which allows a choice of pricing models.” 

Notice Carson's choice of words — truly cloud native! A bit of one-upmanship, no doubt, but also a very clear signal that cloud-native database architecture will increasingly become a point of competitive differentiation as more organizations build data warehouses that span hybrid, edge, and multi-clouds.