In RuneScape, the online multiplayer role-playing game from developer Jagex, players can take on fantastical quests in a mythical world.
In Cambridge, the city where the studio is based, the tasks aren't quite as mystical, but as Jagex head of IT Barry Zubel told Computing, their completion will reap rewards for both developers and players as the studio prepares to launch its next blockbuster title: Transformers Universe.
"Our biggest project is virtualisation and that's going to be a big winner for us," he said, describing how "we've historically not virtualised anything operationally" due to the limits it would put on game performance – but that's changing.
"The way we've managed to get that through is we initially used virtualisation in our test and QA environments, and that's worked really well," said Zubel. Benefits from this included a reduced need for physical servers and greater flexibility.
But while many of the servers on which the games run aren't likely to be virtualised, a lot of the infrastructure behind them will go that way.
"That doesn't mean we're going to be running all our front-end game servers virtualised, that's still unlikely to happen. But it does mean a lot of the back-end infrastructure is moving towards virtualisation," he said, going on to explain how a change in the perception of virtualisation has made it easier to sell the concept to his team.
"Historically, it's been something that developers haven't trusted but now we've actually been able to prove to them it works and is reliable. It's actually very cost-effective and it allows you to make the best use of the physical hardware that you've got," he said.
Those servers, Zubel explained, are designed to be available even in the event of a component failure, so as to not disrupt people playing Jagex games.
"We design everything to be N+1 redundant, and I mean everything. Each of our sites themselves, we can physically lose an entire site and it won't affect the playability of RuneScape. What will happen is people will get disconnected then redirected to one of our other sites," he said, adding that services can be moved to ensure services are rapidly spun up again.
"We have the ability to move the applications around our front-end servers very, very quickly. So if we did physically lose a server, we can have that up and running again in a minute."
Zubel also discussed configuration management at Jagex, describing the systems engineering process as "absolutely number one the best thing we have".
"From being able to have a server physically installed to being operational, you're probably at less than eight minutes now, that's all down to configuration management," he said. "We've written an entire ecosystem of add-ons and scripts to our configuration management system, just to do things that are very specifically for our games."
Those games hinge on providing stable online connections to players, which is why Jagex has more than 500 physical servers spread across the globe.
"The closer we can push the game servers to our customers, the better the connectivity," said Zubel, who added that while RuneScape is able to function on high latency connections, the upcoming Transformers Universe isn't so tolerant, so Jagex has been attempting to install servers nearer to players.
"Transformers is a little less forgiving because of the game that it is, so we've been looking at placing servers in places where players will be," he said.
As well as servers, the everyday running and continued development of Jagex games rely heavily on business intelligence and analytics, as Zubel explained.
"BI is a very important part of the business and actually drives a lot of the things that we do.
"We have a lot of hardware allocated internally for analytics clusters, it's a very big part of what we do. It has driven a lot of changes in RuneScape and it is also set up to drive a lot of changes in Transformers as well."
[Please turn to page 2]
Sometimes, the power of the mainframe is the most cost effective answer. Computing's Peter Gothard puts Computing's readers' questions on the future of the mainframe to IBM's Z13 expert Steven Dickens.
This Dummies white paper will help you better understand business process management (BPM)