Microsoft CEO Satya Nadella posted an intriguing video on LinkedIn this week extolling the virtues of what he called a “hyperscale limitless database” to meet the surging needs of the data-driven economy.
I am *clearly* no expert on database technology. But I surmise that the key takeaway of the video from Nadella and corporate VP of Azure Data Rohan Kumar is that Microsoft’s looking to revolutionize the way companies can gather, manage and analyze staggering amounts of data—not only in the cloud but also at the edge.
That’s a challenge that every major player in the Cloud Wars is attempting to overcome. And Microsoft appears to have made a huge step forward with its new Azure Database Hyperscale.
Nadella opens the 4-minute video with this overview:
Any modern application has to deal with data: the variety of data, the velocity with which data is being accumulated, and the sheer volume. And to make it possible for any application developer to build these modern applications, we have now completely rewritten the data platform on Azure.
One of the things I’m very excited about is how we’ve taken a cloud-first, ground-up approach to building a new platform where we’ve now thought about and really taken advantage of the new memory hierarchy and really made it possible for compute and storage to come together to support the real needs developers have.
Nadella then asks Kumar to describe “this hyperscale limitless data estate.” Kumar explains how the long-time constraints of relational databases always force developers to make trade-offs in the volume of data the app can handle, or in the capabilities of the app itself.
Those seem like lousy options, particularly in today’s data-intensive economy. So Kumar’s vision for how Microsoft hopes to transcend those limitations offers an intriguing look at what the future holds.
“Relational databases are really powerful,” Kumar explains, “but one of the challenges of scaling relational databases to very large volumes beyond a certain size—a few tens of terabytes or so—every relational database reaches its limits on compute, memory and storage, and that ends up impacting its performance and availability.
“More often than not, you look at the database architects and they spend a *lot* of time up-front developing their application workloads to think about how big the database is going to get, and then do trade-offs around database size and performance.
“Once a relational database reaches its limits, the only option developers have is to completely redesign and rewrite the application,” Kumar says.
“With the volume of data growth we’ve seen from intelligent edge devices and sensors, there’s a very clear need for a relational database to break through these limits. We have such a relational database that we’ve recently launched on Azure, called Azure Database Hyperscale.”
I’ll be keeping a close eye on the market reaction to this product, and on possible reactions from Microsoft’s competitors in the Cloud Wars.
Subscribe to the Cloud Wars Newsletter for in-depth analysis of the major cloud vendors from the perspective of business customers. It’s free, it’s exclusive, and it’s great!