Google Blogger off-line. Cloud computing has a storm.

At approximately 4pm EDT (Eastern Daylight Time), I lost access to my blog on Google’s Blogger. Okay, it’s probably just a small glitch. I try at 5pm, no go. I try again at 6pm, 7pm. I occupy myself with other stuff. The evening wears on but still no go. I start doing some lookups on Google to see if I can find anything about the problem. Finally, I find some entries about an update to the Blogger system that went haywire and the Google engineers were trying to put the system back to the way it was just before the update. Whoa! This sounds major.

The idea of cloud computing sounds attractive. You put things onto the Net and it goes… somewhere. Go to any other computer on the planet and get your stuff. Sounds like a great idea… except when it doesn’t work.

On April 21, 2011, Amazon which offers cloud computing services to a whole whack of Internet services had their North Virginia centre go down and with it went some major players on the Web like Reddit, Foursquare, and Quora. Even some European companies lost their window to the world. Things were eventually sorted out but it certainly raised a lot of questions about whether cloud computing, which looks so good on paper, actually lives up to its promise.

Like anything which offers redundancy or a backup, the idea is that if one computer centre goes down, the other centres will pick up the slack. How well does that work? On February 27, 2011, Gmail went off-line for a few hours, but when it did come back up it was reported that 150,000 users had empty Inboxes. Where did the emails go to? This got sorted out within a week, but the publishing of the how made me shake my head. Google had to restore from backup tapes. Backup tapes? Excuse me but that strikes me as a little “primitive” in these days of fancy pants cloud computing.

As a user, I am sitting here wondering about reliability; I like to turn on the lights and actually see them light up. In defense of Google, I realise that in computer circles a small error by itself is no big deal but unfortunately, that small error times a million all of a sudden seems really, really big. Having worked in the computer field for the past 25 years, I have had my own times where I’ve hit the Enter key then suddenly realised what I wanted to do is not working correctly… all as I watch my mistake propagating throughout the network onto every machine in the company. Bingo, my small error times a hundred, a thousand, a hundred thousand, a million. It’s amazing to feel that cold chill come over you as you stare at the computer screen and the slow realisation sets in that what you’ve just done is so wrong. Picture parachuting out of an airplane and after your freefall, you go to pull your ripcord and find out you didn’t put on your chute, you put on your backpack containing your lunch. That’s the cold chill I’m talking about. You are in big trouble spelled with a capital F.

Good luck Google engineers. It’s Friday the 13^th.

References

ZDNet – May 13/2011

Google’s Blogger outage makes the case against a cloud-only strategy by Ed Bott

ZDNet – May 13/2011

Google promises Blogger outage will be over ‘soon’ by David Meyer

CloudTweaks – March 7/2011

Gmail Outage – Is Cloud Computing To Blame? by sourya

Postscript – Fri May 13/2011 at 1:30pm EDT

I’m back on-line. I was shut out for twenty-one and a half hours.

Click HERE to read more from William Belle

Article viewed at: Oye! Times at www.oyetimes.com