Riot Games serves more than 100 million players worldwide. Its debut title, League of Legends, is now the most-played PC game in the world and a key driver of the rise of esports. Beyond League, Riot enjoys rising demand for a growing set of games, such as VALORANT, and multimedia projects, such as Arcane, an original series.

To deliver seamless, uninterrupted gameplay experiences around the world, Riot uses cloud infrastructure—but it didn’t always. Over about three years, Riot moved from on-premises datacenters to Amazon Web Services (AWS) in a journey that was nothing short of legendary.  

Riot launched League of Legends back in 2009. Amazon debuted the first commercial cloud only three years before. Considering the cloud was such a new technology, it’s little surprise that Riot kept all its IT infrastructure in datacenters. “Let’s be honest,” said Riot’s head of global infrastructure and operations, Brent Rich, on stage at AWS re:Invent 2023, “we didn’t trust that anyone else could meet our bar for making live games great.”

In Rich’s words, “that worked for about a decade.” By then, it was taking “forever” for Riot to get things done. This risked affecting player experience, and with aspirations to be the most player-focused game company in the world, Riot would not tolerate being encumbered by aging technology. “We knew that if over 100 million players worldwide didn’t see Riot investing in and loving our own game, they would leave, even if it is free to play,” said Rich. 

By that same time, leaders at Riot had grown more confident that the cloud was a viable option, and that AWS had the broad set of services that Riot needed to run its games. So, Riot decided to go “all in” on cloud, transitioning away from the work of managing datacenters to focus more on building amazing games and player experiences.

Through meticulous planning across teams from Riot, Slalom, AWS, and AWS Professional Services, Riot completed the journey to the cloud with finesse. It retired or migrated nearly all its infrastructure, including the workloads that power League of Legends.

The overall migration program consisted of several different initiatives, including: 

• Decommissioning 16 datacenters across 10 cities that supported gameplay and in-game purchasing for millions of players worldwide

• Exiting the home that was the heartbeat of game development at Riot: a datacenter in Las Vegas with more than 1,200 servers and 300 terabytes of data, as well as multiple game source code repositories spanning multiple Riot teams 

• Improving how Riot’s teams provision new, cloud-based IT resources through an automated deployment model

Riot hit every target date for leaving its datacenters, thanks in large part to an automation runbook that Slalom created for Riot’s transition to the cloud. The runbook was the product of much observation and discussion we orchestrated to understand, deconstruct, and document the intricacies of the engineering procedures for League of Legends. Riot’s teams successfully used the runbook to recreate on-premises environments on AWS efficiently, which helped them compress launch windows over the course of the migration. 

More important for players, Riot’s migration has helped it improve latency to ensure that every click, every move, and every interaction is instantaneous for every player around the world. Riot’s done this by making the most of the global infrastructure of AWS and carefully considering different locations for its servers, with support from Slalom and AWS. For example, Riot was able to achieve latency improvements that were 36% to 53% better than expected in one region. This has translated to a smoother, more immersive gaming experience for players.

With AWS’s powerful tools at its fingertips and new, cloud-friendly processes in place, Riot is also accelerating feature releases like never before. That means more frequent game fixes and faster responses to players’ needs and desires. 

“This transition to AWS technology has not only uplifted the Riot player experience globally, it has also established a foundational environment for integrating cloud solutions into game development,” said Tracy Ruthkoski, a director at Slalom.

According to Rich, the entire migration would not have been such a triumph without the teamwork of Riot, Slalom, and AWS. In an AWS blog post about the work, Rich underscored the significance of strategic partnerships in navigating complex challenges: 

“There are some things that are very hard to do without a strategic partner. Working with AWS and our integration partner, Slalom, for example, we now have a shared automation runbook for League of Legends that is incredibly detailed. If we have to stand something up in a new region, we can now do that in a matter of weeks. Those partnerships are invaluable.”