Google App Engine: Just Like Amazon; Now with More Outages!

First came the Tweets: is such-and-such down? Can anyone else get into Google App Engine? Then came the blog posts and the endless discussion of "giving up your app to the cloud."

The simple fact is that web sites will go down. All web sites will fail to work at some point, because 100% uptime just doesn't exist. Software will be buggy. Hardware will fail. And the best anyone can hope for is that the site is large enough and designed well enough to have enough fail-safes in place that the inconvenience is a tiny blip.

The Google App Engine fans , however, have taken great delight in pointing out Google's stellar uptime for its own sites in comparing GAE to Amazon Web Services. Amazon's recent site outages caused the poor dead horse to be beaten yet again about the AWS outage back in February that took down the precious Twitter. Now GAE is taking its turn in the Fail Whale spotlight, with its first major outage a mere two months after launch.

As it turns out, even the mighty Google is capable of coding errors, and a message posted today from the App Engine Team acknowledged both the bug and the plan to prevent such a bug from replicating today's outage in the future. In other words, they really are just like Amazon.

Today's outage was indicative of a bigger issue that a bug at Google (since I'm sure they face bugs all day long): we are quickly become consumers who are moving at such a ridiculous pace that we can't tolerate normal failure rates. It has nothing to do with cloud computing or hosted apps or even GAE, and has everything to do with our unrealistic expectations of perfection in all things.