Wednesday, September 14, 2011

A tale of two outages

 

Let me tell you something you already know and is bleedingly obvious anyway – Computers fails, IT systems go down. With this in mind I’d like to compare two recent examples.

 

Example 1.

 

Office 365 recent had an outage of a few hours (technically it was DNS not the Office 365 service but unavailable is as good as down). During that time let’s have a look at the impact. I certainly couldn’t receive any emails, I also couldn’t send any emails but I could still compose them and have them queued in my mailbox. I couldn’t have gotten to my SharePoint data and Lync would have also been offline.

 

So it was certainly preventing me from potentially doing some work but I could still access my calendar, contacts and previous emails.

 

Example 2.

 

The server on which this blog runs blew a power supply. So again system down but this time no access to any information on that box. Tools down time.

 

Wanna known the difference? In example 2 with my blog server, I had to pack up the machine. I had to drive it to repair shop. I had to wait until the power supply was changed. I had to drive the machine back. I had to connect it all back up and make sure it was working. I lost over 3 hours from start to finish getting the server back online.

 

In example 1, I kept an eye on Twitter to see when the system was back online for others and until then I went on with OTHER WORK.

 

So in which scenario was I more productive? For me personally it was example 1 as I could get on with other things because I knew someone (a.k.a. Microsoft) was working on the problem. I could still use some of my systems that had local copies (i.e Outlook) and could have with SharePoint if I had chosen to use SharePoint Workspace.

 

With example 2, nothing was going to get fixed until I fixed it.

 

Moral of the story? Computer systems go down, whether they are in the cloud or whether they are machines at the end of your fingers. It therefore follows that no matter where the computer are you use, you need to have some plans for when they fail (just like you need a plan to backup them up).

 

So Office 365 was unavailable. In this case I was more productive than when my own site server failed. I also content that would be a similar experience for most businesses.

 

Computers fail, deal with it. Develop a contingency plan to stay productive. What would I have done if the electricity failed? With Office 365 I would have worked off my laptop battery and wireless Internet connection until the battery ran out and then I would have relocated elsewhere to where the power was working. If I had all on site equipment I’d have no choice but to wait in the dark until the power came on.

 

It would be nice to see people actually discussing solutions to contingency problems rather than playing chicken little and blaming the sky falling on the evils of cloud computing. Come and see me when you are ready to have a BUSINESS conversation rather than a hysterical rant.

 

Computers fail, deal with it. A smart business would.