If you haven’t read about the Pokemon Go craze that’s taking the world by storm, you’re probably busy playing the augmented-reality game and have set aside your regular pursuits as you pursue another monster.
Launched this month, the game immediately exploded in popularity (it now has more users on Android devices than the Twitter app) and increased parent company Nintendo’s market value by $17 billion in just over a week.
But with a user base that continues to grow – it’s available in 26 countries and counting – Niantic, the Nintendo subsidiary that created the game, can’t keep pace with high player demand and has suffered at least 17 outages because of overloaded servers.
Niantic won’t comment on its infrastructure, but if the brilliant folks there were to reach out for advice, we’d refer them to a blog post we published in 2014. It’s one of our most popular ones and still applies in these Pokemon Go days of 2016: Five Things They Never Told You About Downtime.
Niantic undoubtedly will figure out how to meet user demand before it’s too late, before players have had enough of interrupted service and decide no game is worth waiting for. So rather than dish out advice to Niantic, it makes sense for us to offer advice to the creator of the next video game phenomenon that will keep us glued to our screens and require a tremendous amount of computing resources. For that matter, this advice applies to all games, regardless of popularity.
So before you release your mobile video game, here are a two critical rules you should follow:
The cloud is the most sensible platform for running any application, but especially widely used applications. This is because the cloud is more flexible than physical infrastructures and can scale according to your needs. Most cloud service providers allow you to increase your existing resources to accommodate increased usage. Conversely, you can downscale when needed, paying only for the services/storage/compute you are using.
Physical infrastructure and data centers do not allow for the scalability found in the cloud. If you decide to use physical infrastructure and invest in hundreds of on-site servers, you’ll eventually regret the decision. We hate to say it, but another hot new game will come along after yours, and the demand for your offering will decrease. You’ll be stuck with the cost of maintaining those unused servers. By using the cloud, you can easily support business growth without expensive changes to your existing IT systems.
Whether you’re in the cloud or relying on a Pokemon army’s worth of physical infrastructure, you’ll inevitably need to recover data. No solution is perfect; there will be outages. Even the best companies have them, and they happen more frequently than you might imagine.
Having a disaster recovery (DR) strategy is paramount to maintaining business continuity. IT downtime can cost anywhere from $1 million annually for a mid-sized company to more than $60 million for a large enterprise. Moreover, companies face public backlash and diminished reputations if outages strike often. Niantic hasn’t taken a hit, but if service outages continue, Pokemon will probably start falling on the app download charts.
When planning your DR strategy, keep in mind that just backing up data to the cloud is not enough for DR. Entire systems and applications must be included in order to successfully be back online quickly after a disaster. A DR service such as CloudEndure not only can have you back up in minutes, but thanks to our continuous, block-level replication technology, you won’t lose any data in the process.
So long before you start creating your mind-blowing video game, get yourself set up in the cloud – and make sure you have a strong DR strategy in place. You certainly should aim to have a hit as big as Pokemon Go but you definitely don’t want to suffer the slings and arrows of downtime.