Anti-Cheating Tool for Massive Multiplayer Games Using Amazon Aurora and Amazon ML Services – Percona Live ONLINE Talk Preview

by Yoav Eilat, Yahav Biran

Percona Live Online Agenda Slot: Tue 19 May • New York 4 p.m. • London 9 p.m. • New Delhi 1:30 a.m. (Wed) Level:  Intermediate


Multiplayer video games are among the most lucrative online services. The overall games industry worldwide generated an estimated $174B in 2019, according to IDC. With this popularity, cheating becomes a common trend. Cheating in multiplayer games negatively impacts the game experience for players who play by the rules, and it becomes a revenue issue for game developers and publishers. According to Irdeto, 60% of online games were negatively impacted by cheaters, and 77% of players said they would stop playing a multiplayer game if they think opponents are cheating.

Current methods for detecting and addressing cheating become difficult and expensive to operate as cheaters respond to the evolution of anti-cheating techniques. This session will show an effective method for game developers to continuously and dynamically improve their cheat-detection mechanisms. It uses Amazon Aurora and Amazon SageMaker for cheating detection, but can be adapted to other databases with similar capabilities. We’ll utilize the recently-launched Aurora machine learning functionality, which allows game developers to add ML-based predictions using the familiar SQL programming language without building custom integrations or learning separate tools. We’ll show which ML algorithms are useful for cheat detection and how an anti-cheat developer can write a single SQL query that handles the inputs and outputs for the algorithm.

Why is your talk exciting?

Machine learning is everywhere these days, or at least that’s how it feels when you work at Amazon. Some types of ML get a lot of attention, like self-driving cars, or services that take a JPEG and tell you if it’s a dog or a cat. But if you think about it, a vast amount of the world’s information is plain old tabular data in traditional relational databases. What about running ML on that data? Who knows what amazing insights and secrets are lurking inside?

We’ll look at a cool video game example where we’re looking for cheaters, e.g. people who write bots to play on their behalf. We’ll show which ML models can detect these cheats and how to more easily run the analysis from your application, using tools that we’ve built. You should be able to run it on other databases if they have similar ML capabilities.

Who would benefit the most from your talk?

Application developers and database administrators who don’t know a whole lot about machine learning but would like to start.

What other presentations are you most looking forward to?

We’re curious what a complete online event will be like.  We’re looking forward to see how it compares to the traditional kind of conference.

Yoav Eilat

Yoav is Senior Product Manager for Amazon Relational Database Service and Amazon Aurora. He joined AWS in 2016 after holding product roles for several years at Oracle and other technology companies. At AWS, he managed the launches of Aurora PostgreSQL, Aurora Serverless, Aurora Global Database, and other major features. Yoav currently focuses on new capabilities for the MySQL and PostgreSQL database engines.

See all posts by Yoav Eilat »

Yahav Biran

Yahav is a Senior Solutions Architect in Amazon Web Services Game-Tech. Dr. Biran designs reliable, resilient, and secured cloud-based information systems for AWS Cloud customers. He received his Ph.D. (Systems Engineering) from Colorado State University in 2017. He was previously a Technical Program Manager at Microsoft Azure; there he built Azure Compute Services. He has more than fifteen years of professional experience in information systems. His research seeks an objective and overall higher fidelity approach to optimize large scale game-tech scenarios.

See all posts by Yahav Biran »


We invite you to our forum for discussion. You are welcome to use the widget below.

✎ Edit this page on GitHub