During our weekly sysadmin call this morning several of our experienced sysadmins quickly pointed out that clients seeking very high up time should not necessarily look for it on the cloud. I couldn’t believe it, but almost as an omen, this story came on my RSS feed from Webware: Google apologizes for email outage
Outages pose problems for Google as it tries to persuade companies to buy into its cloud-computing vision, in which applications are hosted on the Internet rather than on corporate computers. But Google argues its service availability is competitive with most organizations’ abilities to run their own e-mail servers.
Clearly the google cloud isn’t going to be the only one having an outage from time to time. It seems to me that while still in the infancy, these services are vulnerable to unexpected problems, kind of like the famous first internet worm - the Morris worm - that brought the internet to its knees back in the 1980s.
While cloud services offer a lot of promise, and overall should offer a better level of redundancy and up time, this shows that the cloud is also not immune to some down time.
Here is the link to the original Google cloud problem blog post.