It’s an axiom of modern application performance: The more advanced an app, the more demands it places on infrastructure. This is true in big data, trading, cybersecurity and a range of other use cases. It’s also true in the world of chess, a game that enjoyed renewed popularity a year ago following the Emmy award-winning Netflix miniseries The Queen’s Gambit, which follows the rise of a chess prodigy from basement chess matches to global acclaim.
Today we’re talking about online chess, specifically Chess.com, the world’s largest online chess site. The Chess.com application is quite the workhorse; it has over 65 million members and more than 250,000 concurrent users playing more than 10 million games online. Every day.
SEE ALSO: Deploy a Cloud-native Java App on Kubernetes Using Spring and K8ssandra
Whether face-to-face or online, the gaming experience is everything for chess players. Any disruption, be it noise, room temperature or (in the case of online chess) sluggish software, can ruin a player’s concentration and change the outcome of the match.
Chess.com faced this very issue, a victim of its own massive success and exploding playerbase. With millions of users, the application often produced delays lasting several seconds between players’ turns. This interrupted the flow of the game, made it less enjoyable and hurt user loyalty. Furthermore, in-house Java developers couldn’t fix the issue, despite their extensive efforts to tune the Java programming language code used to build the Chess.com application.
The cause of these application delays? So-called “Stop-The-World” garbage collection pauses. That is, pauses in the Java code that results from a process which frees up memory on the infrastructure running the application. Garbage collection is a responsibility of the Java Virtual Machines (JVM) that are used to run Java application code.
That’s where Azul comes in. Chess.com uses Azul Platform Prime, a highly performant Java runtime used by capital markets firms, manufacturers, ecommerce sites and digital advertisers — anyone who needs better application and infrastructure performance. The use of Azul Platform Prime checked the garbage collection glitches in Chess.com’s Java applications and in its Java-based technologies, such as Jetty. In doing so, it eliminated application stalls altogether, and improved reliability and players’ experience. Azul’s solution was also a win for developers, increasing productivity by swapping their underlying JVM without any changes or tuning to their actual Java code.
SEE ALSO: OutOfMemoryError: Kill Process or Sacrifice Child – Causes and Solutions
Azul Platform Prime improves the performance and scalability of Java ecosystems with a highly optimized runtime that ensures consistent response times, reduces system stalls, and delivers better customer experiences — all with far less infrastructure. It boosts the number of transactions from the same hardware and speeds up Java performance even as loads increase. As a result, it can reduce capital expenses for servers by as much as 50%, cut operating expenses for cloud services with no over-provisioning and drive continuous value.
In the words of Chess.com CEO Erik Allebest: “Because Azul Platform Prime powers our Java-based platform, we don’t have to worry about application speed or infrastructure costs. We can focus on our members’ experience and rest assured that Azul Platform Prime will handle thousands of moves a second with the best performance possible on the world’s best chess website.”