Customer Case Study:

Riot Games

Rising to the Challenge: Derive Greater Insight From Data

Riot Games is best known for their League of Legends game whose annual tournament, the League of Legends World Championship, draws top esports teams from around the world. The tournament is the most widely viewed and followed esports tournament and one of the most popular gaming events globally.

With more than 150 million registered users, League generates vast amounts of data for Riot Games. With an ever-expanding Databricks lakehouse on their AWS S3 cloud storage service, Riot struggled to derive value from all that data. “We needed a tool to help us find the information — the needles in the haystack — with value in our data lake,” says Michael Leslie, Senior Data Architect at Riot Games.

Riot’s analysts needed the right data to derive insights using their Tableau business intelligence platform (BI); meanwhile, their engineering teams struggled to answer analysts’ questions about the data. Engineers found themselves answering the same questions repeatedly, through Slack discussions and direct requests. They needed a way to document information about the data: one accessible place for analysts to self-serve and self-educate.

Giving Analysts the Power of Data at Their Fingertips

Initially, Riot Games focused on finding a solution that would help analysts discover trusted data for analytics. Ultimately, when Riot chose Alation, they gained a data intelligence platform that helps both their analysts and engineers become smarter about data.

Today, Riot analysts and data engineers usually begin their data search with Alation. To encourage adoption, Leslie’s team started with a single data set in their primary data lake and then expanded curated data sets. First, analysts use keyword searches to discover the data they need. They can then use Alation to explore data in depth — down to the schema, table, and column level — or trace lineage to ensure usability for their purpose. They can then run queries and create reports in Tableau directly from Alation.

Analysts are not the only users of Alation at Riot Games. Engineers use Alation to catalog and curate metadata from thousands of tables and datasets they create every day. Having the metadata in Alation reduces the number of direct inquiries to the engineering team from analysts – freeing up engineers’ time for more meaningful, rewarding projects.

Synching with the Engineers to Improve Workflow

Leslie calls Riot Games an engineering-first organization. “Most of our Alation users are engineers, and a lot of them are building tables and data sets in Databricks,” he notes.

Riot Games used scheduled metadata extraction to pull metadata from tables created in their Databricks environment. Speeding up this daily task was a key goal. “We wanted to build an experience where an engineer could create something and it’s there within the catalog for them to see within minutes,” says Leslie. Since the engineers needed a custom solution, Leslie reached out to the Alation support team for their expertise.

To help, Alation Professional Services created an event-based metadata extraction (MDE) API that updates the Alation catalog whenever engineers create or change a table in Databricks. When AWS Glue captures the event, the Alation API receives an alert and syncs the catalog with Databricks. Engineers can add their annotations to the newly created table within minutes, and move on with their work.

Seeing the Results: Higher Productivity and Efficiency

Leslie believes that Alation makes Riot Games a better, more interesting place to work for both analysts and engineers. “Riot Games has improved the efficiency and workflow of both engineers and analysts who use Alation,” he says. “Engineers can focus on creating and managing data pipelines, and less time answering questions, while analysts focus on deriving insights from data, instead of searching for and asking engineers about it.”

Alation helps analysts quickly find the data they need through the simple, Google-like Alation search. Once they find what they need in Alation, they can create Tableau reports to gain insights from that data. Leslie notes that lineage features enable users to gain trust in the data. “The lineage feature in Alation is really important to us,” he says. “It provides a graphical representation of all the lineage between the data and the report results.”

Analysts no longer need to wait for engineers to provide information and context about the data they’re using; they can tap directly into engineers’ knowledge through the Alation catalog. “Alation creates an end-to-end relationship between the data producers and consumers,” says Leslie.

“Alation creates an end-to-end relationship between the data producers and consumers at Riot Games.”

Michael Leslie
Senior Data Architect, Riot Games

Engineers at Riot Games have also improved their workflow and efficiency with the Alation MDE API that updates the catalog whenever a change is made to Databricks. Adding annotations is now part of the engineers’ immediate workflow whenever they create a table — and they love not having to wait for 12- to 24 hours to check that box.

The engineers within the League of Legends data central group are also actively building their own engineering library that plugs into Databricks (which auto-updates in Alation thanks to the MDE API). They’re now adding extended metadata into the table descriptions to better answer questions about the data. “They wanted to do more and automate more,” says Leslie. “This is one reason I pursued the metadata extraction API: to help drive further adoption and integrate engineers more fully into our ecosystem.”

As for the future? Riot Games plans to integrate data-use policies directly into the catalog, so people get just-in-time guidance on how to use data compliantly. Because Alation Professional Services created the near real-time MDE API, Riot engineers will be able to quickly tag and annotate data with security and sensitivity information as part of their workflow. Riot Games’ Alation catalog is poised to become the source of truth about sensitive data, adding an extra layer of protection around that data and helping the company avoid regulatory risk.

About Riot Games

Founded in 2006 by Brandon Beck and Marc Merrill, and led by CEO Nicolo Laurent, Riot is headquartered in Los Angeles, California. In 2009, Riot introduced their debut title, League of Legends, to worldwide acclaim. League is now the most-played PC game in the world and a key driver of the explosive growth of esports. Riot has over 2,500 employees, called Rioters, in more than 20 offices around the globe. They have been featured on numerous lists including Fortune’s “100 Best Companies to Work For,” “25 Best Companies to Work in Technology,” “100 Best Workplaces for Millennials,” and “50 Best Workplaces for Flexibility.”

• AWS S3
• AWS Glue
• Databricks
• Tableau