Fueling innovative software
July 15-18, 2019
Portland, OR

Transforming legacy infrastructure into globally scalable regions

Sean Kane (New Relic)
11:00am11:40am Wednesday, July 17, 2019
The Next Architecture
Location: D135/136
Secondary topics:  Cloud Native
Average rating: ***..
(3.88, 8 ratings)

Who is this presentation for?

  • Engineers and technical leadership




What happens when a company has outgrown its very first data center or cloud region? Ideally, it just uses its existing tools to stand up as many new regions as it needs. But in reality we know that a lot of compromises are made in a company’s early years, and your first attempt at building infrastructure probably looks a lot like the typical monolithic application—something that got you through the early years but really needs to be rethought and divided up into manageable components.

Sean Kane details how a very small engineering team inside New Relic spent the better part of the company’s 10th year in business implementing a plan that would take New Relic from a single US data center to having multiple global presences with multiple providers. The resulting software stack, which is largely made up of open source tools, not only allowed the company to streamline standing up and managing new regions but also improved its annual disaster recovery exercises significantly by providing a 90% reduction in engineering hours and a 25% reduction in the operating expenses for the event.

Sean shares what the company did and what it learned, demonstrating that this type of radical organization shift is achievable by anyone with just the right amount of shared vision, technical prowess, and trust.

Prerequisite knowledge

  • General technical and business experience

What you'll learn

  • Understand that large, critical organization change is possible from small, focused teams
  • See how a minimal but well-considered set of tools and processes can open up many doors
  • Learn to build systems that encourage the behavior you want from users and to iterate, iterate, iterate
Photo of Sean Kane

Sean Kane

New Relic

Sean Kane is the lead site reliability engineer at New Relic. A longtime system administrator and operations engineer, Sean has worked in a range of industry segments, including biotech, defense, entertainment, and hardware and software engineering in locations ranging from Alaska to Pakistan over his 20-year career. He’s the coauthor of Docker: Up and Running and provides Docker-related training with O’Reilly. In his spare time, Sean enjoys photography and sharing with his children the endless wonders of science, the great outdoors, and rappelling down skyscrapers. If you are looking for a conversation starter, Sean graduated from the Barnum & Bailey Clown College, completed two summer internships with the CIA, and built the first website in the state of Alaska, as well as the original USPS site.