Optiva

Migrating thousands of globally distributed machines to Google Cloud Platform—Without the downtime

"Pythian had the best price and the best approach, with a world-class team of really outstanding resources, which they made available to us on an elastic basis, so we didn’t have to pay for the whole team the whole time. We also valued their deep relationship with Google—that stamp of approval was meaningful for us."

 

Business need

Following multiple acquisitions, Optiva had 10 data centers around the world housing more than 3,500 machines running more than 50 operating systems—many of them extremely out of date. Some of the machines also had licensing issues that made it difficult to migrate them to GCP. Optiva needed to move all of its workloads to the cloud within six months. And extended downtime was not an option: its internal teams required constant server access to do their jobs.
 

Solution

Before a contract was even signed, Pythian consulted with Optiva to understand its IT environment, workloads, organizational structure and ambitions for the cloud—and to identify potential challenges that might arise during the cloud journey. This in-depth analysis allowed Pythian to vet its assumptions about how the migration would work so it could hit the ground running once the contract was finalized.  Thanks to a globally distributed team, Pythian has the capabilities to work around the clock to ensure the migration progresses as smoothly as possible. As a first step, Pythian analyzed all of the machines and categorized them based on how difficult they would be to migrate. Based on that analysis, Pythian could use  automated workflows to quickly and easily migrate the straightforward machines, leaving its team free to dedicate as much time as possible to those with complicated technical or licensing issues. After determining the optimal migration pattern for Optiva, Pythian leveraged live migration technology continuously and incrementally replicated and synchronized data from source machines to target machines. This process allowed Pythian to migrate the company’s workflows with minimal to no downtime—and zero impact on live production activity. Far from a simple “lift and shift,” Pythian dove deep into the technical details, applying cloud-native best practices to tune and enhance the company’s servers and operating systems to ensure optimal performance in the GCP environment.
 

Result

Pythian’s in-depth architecture and discovery phase allowed the migration to happen swiftly, which let Optiva take immediate advantage of GCP capabilities and make improvements in areas like automation, security and monitoring. After six months, Optiva’s annual data center costs were down to just $4 million, with further savings expected to bring costs all the way down to $1.5 million—a reduction of 90% from its original data center budget of $15 million. The additional cost-saving benefits of GCP—such as paying only for resources used rather than continuously for always-on servers, or deploying workloads to larger GCP data centers that are billed at lower rates—will help Optiva save even more money in the long term. Moving most of its assets into GCP also allowed Optiva to consolidate its remaining physical servers from 10 data centers into just one. This has greatly simplified the IT team’s job: there is now a single support ticket management system instead of four, seven operating systems instead of 50 and four remote desktop protocol (RDP) systems instead of 120. And now that Optiva is less reliant on third-party hardware, it can also reduce its total cost of ownership (TCO) and accelerate its refresh cycles.
 

Technologies

  • Google Cloud Platform
  • Google Compute Engine
  • CloudEndure