The latest cloud infrastructure announcements, technical solutions, and enterprise cloud insights.

Rethinking How to Build a Cloud

Kyle York
VP of Product Strategy, Oracle Cloud Infrastructure and GM, Oracle Dyn Global Business Unit

Oracle Cloud Infrastructure has faced many challenges with its late entry into infrastructure as a service (IaaS), but that late entry has also come with a significant benefit: we've been able to hire the best and brightest people from the market leaders.

Most of the people building Oracle's cloud have worked on at least one other cloud. Their experiences have guided us as we've purpose-built the industry's first truly enterprise-grade cloud. Let's take a look at our approach and how it has evolved.

In the Beginning

Our initial goal was to create a high-scale cloud that would be a better fit for enterprise workloads such as Oracle applications and databases. Many Oracle customers wanted to move these mission-critical workloads to the cloud, but they found that it wasn't easy. And in some cases, it was downright impossible.

To address their needs, the Oracle Cloud had to be a virtual infrastructure that looked like their on-premises environments, but it also needed the scalability of a public cloud. That combination would make it possible to more easily migrate not just applications but entire on-premises systems—virtualization, storage, and management and security software—to the cloud.

The ability to move and improve these systems is one of our most important differentiators in the market. A lot of hard work and ingenuity went into making it happen.

Oracle Cloud Infrastructure engineers spent the first year or so designing and building the foundation, which included data centers, automation, storage, and networking—plus the tools to build on top of that foundation. They all took a fresh perspective on what they had learned from working at companies such as Amazon, Google, Microsoft, and Facebook, and they rethought how to build a cloud.

The three guiding principles of this approach, which we still follow today, are:

The next step was to build out the core infrastructure pieces and keep iterating. The first Oracle Cloud Infrastructure offering was bare-metal compute, and then came virtual machine instances, followed by integrated database services, and we're adding new services today.

To Infrastructure and Beyond!

With everything we do, we're trying the answer this critical question: how do we support and improve our customers' mission-critical systems?

A current area of focus is security and compliance. Oracle Cloud Infrastructure holds a PCI DSS Attestation of Compliance for more than a dozen services, and an attestation for HIPAA's requirements around security, breach notification, and privacy. We also recently announced a web application firewall, DDoS protection, cloud access security broker support, and a key management service for increased cloud security.

Additionally, Oracle is making advancements in next-generation cloud infrastructure. We've announced a streaming service to receive, process, and archive all infrastructure and platform events in near real time. And we're the first public cloud with generally available compute instances powered by AMD EPYC processors. These enable us to offer 64 cores per server—more than any other cloud in production—which is a great fit for database, big data analytics, and high-performance computing (HPC) workloads.

Join the discussion

Comments ( 1 )
  • Alexandre Lacreuse Thursday, January 10, 2019
    Thanks for your article. Predictable pricing is Key Success Factor.
Please enter your name.Please provide a valid email address.Please enter a comment.CAPTCHA challenge response provided was incorrect. Please try again.Captcha