A Dark Sky for Cloud Computing


The message I got when I logged in to Twhirl today – another cloud computing outage. With all the hype around Cloud Computing right now, the darling of the genre is having issues, and not for the first time. As ever, I don’t point this out to have a pop at Amazon as frankly I love the pro-bono work they’re doing to raise awareness and use of this new computing approach. Being top of the pile comes with downsides though and one of them is when this happens, people notice and you make it to the top of Techmeme pretty quickly.

GigaOm talks about the S3 Outage Highlights Fragility of Web Services and Loic who owns Twhirl weighs in too. As always, Richard McManus and the RWW guys have some of the most insightful coverage.

People talk of this stuff as Utility Computing but right now it’s not – imagine how pissed off you’d be if your electricity or water supply went down for 6 hours or more. The big shift is ahead for the industry but not without the following

  • some big names will lose along the way – or get eaten up
  • lawsuits will happen as SLA’s are promised and not met
  • the backlash as cloud computing goes over the hype curve is inevitable
  • a few small players will make big, big investments that are needed to make the cloud work
Comments (4)
  1. Michael Entwhistle says:

    I think its been overblown. I work in London and there was a power outage in most of Soho about 18 months ago so its not like there isn’t precedent in utilities.

  2. Don’t hosting companies only offer a guaranteed uptime of 99% for a reason?

    Cloud computing is no different. Systems are put in place to minimise the disruption of a site going down so why should the applications we use for worl be any different?

  3. By Dan D. Gutierrez

    CEO of HostedDatabase.com

    We launched the web’s first web-hosted database offering in the SaaS space back in 1999 under the name eCriteria.net (recently re-launched as HostedDatabase.com). In the nearly 10 years we’ve been serving customers, our SLA has never been breached. Back in 1999, SaaS (or ASP as it was called then) was pre-mature because most products didn’t take their SLAs seriously. We knew the SLA would make or break our business, so it was number one on our priority list and it remains top of the heap today.

Comments are closed.

Skip to main content