Martin James, regional vice president of Northern Europe for DataStax, emphasised the importance of having access to flexible infrastructure: “Gaming is a great example of a ‘real time’ business.
If you are playing online, you can’t put up with poor performance – at best, it reduces the range of services that a game can support and at worst it stops games being playable altogether.
“For gaming networks, being able to predict which games are going to succeed is difficult too,” he said. “So preparing how to scale up the infrastructure behind games is a challenge. Instead of committing to specific data centres or hardware platforms, games companies have to be able to scale up almost instantly in response to demand and manage their costs in a linear way, rather than having to over-invest.”
The point about cost is a vital one. By only paying for what they use in a flexible, elastic way, companies will be able to deal with increases in demand without the risk of putting themselves out of business.
Then we come to the issue of reliability. Thanks to the growth of cloud computing, mitigating outages has become a key concern for data centre providers in all industries and gaming is no exception.
“The importance of the data centre in online gaming is only really felt when there is a crash,” said Greg McCulloch, CEO of Aegis Data. “Last year both Sony and Microsoft experienced outages to their online gaming platforms which left millions of users unable to access online features.
“It is vital therefore, that the data centre, which typically sits at the back-end of these immersive experiences, is able to withstand massive fluctuations in power and usage. Spikes are naturally going to be experienced at the end of the day when kids are home from school and adults are back from work, or when a popular game series brings out its latest edition.
“Data centres must be multifaceted in their services for online gamers. The growth in Massive Multiplayer Online (MMO) games, like World of Warcraft, take up huge services in the data centre and if that goes down then there are many disgruntled customers. The more resilient a data centre, the better.”
Pokemon Go provides the perfect example of this issue. The hugely popular app struggled to cope with demand when it was first released in July 2016, with widespread server issues leaving disgruntled users unable to connect.
Developer Niantic Labs turned to Google’s Cloud Platform to help it fix these scalability issues, but was blighted by similar problems again earlier this year at the inaugural Pokemon Go Fest in Chicago.
“Latency and unreliable multiplayer servers are key reasons for players to abandon games,” suggested Nelson Rodriquez, Director Media Industry at Akamai Technologies. “We’ve interviewed thousands of players, and found that poor multiplayer experiences are leading causes of churn.
“Players want to know their games are fair, and they want to be able to count on the back-end infrastructure. They don’t want to have to think about how a game infrastructure was architected. They just want to play, without interruption.”
The reality is that gaming platforms today simply can’t afford drops in connectivity. Customers now demand more than ever from the services they use and won’t hesitate in moving to a competitor if outages become a regular occurrence.
Facebook's messaging service has hit 5 billion downloads, amid reports that it has backed off plans to insert adverts into…