One thing I find increasingly bizarre is the way everyone accepts without question the myth that most enterprises own and operate their own high-specification data centers. In reality, the term 'on-premise' means nothing of the sort. The vast majority of data centers are in co-location facilities operated by telecoms providers, outsourcers or managed hosting providers. This is what enterprises choose as supposedly preferable to relying to cloud providers.
Even more bizarre is what happens when a cluster of disasters strike at co-location centers where all those on-premise servers are housed, as happened last week. For the news coverage reports it as a problem for cloud computing, even though very few of the affected servers belonged to providers of cloud services. It wasn't Amazon Web Services or Salesforce.com that went down. It wasn't even the Rackspace Cloud, even though one of the data centers affected was Rackspace's biggest managed hosting and co-location facility. Look at Data Center Knowledge's list of facilities that went down and it's crystal clear that it was a hundreds or thousands of so-called on-premise servers that were the main victims:
- "On Monday June 29, Rackspace Hosting (RAX) experienced a power outage at its Dallas data center that left several areas of the facility without power for about 45 minutes, knocking many popular customer web sites offline.
- "Early Thursday Equinix Inc. (EQIX) data centers in Sydney, Australia and Paris each experienced power failures. While the power outages were brief Equinix said the Sydney event lasted 12 minutes while power was restored in Paris in just one minute many key customer sites took considerably longer to recover their systems ...
- "A fire at Fisher Plaza in Seattle late Thursday night left many of the building's data centers without power ... The biggest impact was at payment gateway Authorize.net, which was offline for more than 12 hours, leaving its merchant customers unable to process credit card sales. Other sites experiencing lengthy downtime included AdHost, GeoCaching and Microsoft's Bing Travel."
But of course you don't hear about all the enterprise IT systems that were out of action because of those colocation center failures, because the companies concerned are too embarrassed to talk about it publicly. Just as you never hear about the daily outages and data losses suffered by all those other servers that really are on-premise tucked under desks in offices, rack-mounted in a dusty corner of the IT department or hidden away from sight in a switching cupboard somewhere.
Here's the uncompromising view of Jesse Robbins, who until recently was responsible for keeping all of Amazon.com up-and-running, quoted in Seattle business journalist John Cook's blog:
"The people that I hold responsible for outages are not the data centers but the people who built systems that rely on a single data centers or that depend on disaster recovery plans that they write once and never put into practice and end up getting caught with their pants down."
The truth about on-premise computing is that there are barely a handful of organizations in the world that operate data centers to the same standards of security, staff vetting, disaster recovery and performance management as those operated by leading cloud vendors. The best way to guarantee your data and applications stay safe and reliable is to run them on cloud services.