Online gaming is all the rage, as evidenced by the massive buzz accompanying this week’s release of the first-person shooter “Titanfall” for the Xbox One.
Unlike most shoot ‘em ups, “Titanfall” doesn’t feature a single-player campaign that can still run even if the console’s offline; the entire game is multiplayer, and necessitates an online connection. Given the game’s buzz, that meant Microsoft (which produces the Xbox One and maintains the network behind Xbox Live) needed to anticipate a huge wave of players swamping its Azure cloud infrastructure from the moment of the game’s release.
But Microsoft might not have prepared quite as much as it needed to: within hours of players activating “Titanfall” for the first time, reports of issues signing into Xbox Live began to percolate throughout the Web. Those problems seemed to resolve fairly quickly—the day after the game’s release, everything seemed to be running smoothly—but the lesson remains: when it comes to launching a massive cloud-based product, even the most intensive preparations can’t prevent some manifestation of Moore’s Law.
It’s not as if Microsoft wasn’t anticipating the crunch. For the past several months, the company has touted how effectively Azure could handle the computing demands of the world’s gaming communities; for nearly that long, “Titanfall” was positioned as the first demonstration of that capability. “We’re trying to figure out how many people will be playing and trying to make sure the servers will be there for that,” Jon Shirling, one of the engineers for “Titanfall” creator Respawn, told Engadget. “We just say [to Microsoft], here are our estimates, aim for more than that, plan for problems and make sure there are more than enough servers available—they’ll know the whole time that they need to bring more servers online.”
This early hiccup aside, at least Microsoft can tell itself the “Titanfall” debut didn’t end up a repeat of the disastrous “SimCity” launch in 2013, in which Electronic Arts and Maxis failed to anticipate the demand on its servers for the city-building game; the resulting meltdown was a weeklong fiasco that tarnished what should have been a hit for everybody involved. And therein lies the lesson for developers, software companies and cloud providers: prepare for massive demand, especially when your company has the resources to deal with it.