Outsourcing infrasturucture – the S3 flipside…

In somewhat unfortunate timing, ZDnet blogger Phil Waineright posted saying that SaaS vendors that don’t have an outsourced perspective of infrastructure are something of an anachronism. Phil quoted the following;

many established SaaS application providers are applying much more of their precious focus and capital to infrastructure issues than newer competitors that are aggressively utilizing service-based infrastructure

The very next day Amazon’s S3 cloud based storage service had some significant outages. Bear in mind that S3 is a service that SaaS vendors use to outsource their storage needs rather than having to build their own data centres. So does the S3 outage bring a hard reality check for cloud computing in general and SaaS vendors in particular?

Zoli says no,  reminding us that;

this is the first time Amazon S3 went down, and it’s already back up.  Salesforce.com had its fair share of outages, so did other SaaS providers, and so did just about any in-house systems companies run their own installed software on

And this I guess is the bottom line – servers crash, eat that. Power has outages, so what. The fact of the matter is that a professionally run data centre whose primary role is just that – running a data centre – will have higher levels of redundancy than if the data centre is an adjunct to a business’ core function. The S3 outage will be a good reminder to Amazon about the importance of SLAs and the like, a reminder to SaaS vendors to plan for multiple redundancies and a reminder to  SaaS users that no system if totally infallible. What it will not be however is a major barrier to the ascendancy of either the cloud in general or SaaS in particular.