Monday, July 06, 2009
Cloudy PR Thinking
I read a couple of articles today about cloud computing outages (penned by Carl Brooks of SearchCloudComputing.com), and I was struck by the different approaches that Amazon and Rackspace chose in their handling of customer notifications when their cloud computing services went down. With viral social media at the tips of every customer's fingertips, rapid customer response, or lack there of, will invariably have ramifications for PR.
When one of Amazon's data centers was hit by lightening, disrupting services for five hours last week, one customer reportedly grumbled that "he always felt left in the dark when outages occur." An Amazon Web Services spokesperson pointed out in the article that AWS runs a status update page, but must be mindful of exposing vulnerabilities in its infrastructure that could lead to security risks. Still, for the most part AWS' customers took it all in stride, recognizing that some of the inconvenience is attributable to their own configurations or cost-cutting measures that affect service levels.
When Rackspace's hosted services recently went down for about 48 minutes, Rackspace turned to Twitter to update its customers (and by the nature of this medium the greater public). Brooks wrote: "Rackspace has possibly set a new record for transparency and accountability, if not customer satisfaction, by tirelessly tweeting the entire episode." On the flipside, however, at press time the reporter was still awaiting a reply to his request for a comment from Rackspace.
As cloud computing grows in popularity with the promise of cost-effective and secure data storage and virtualized enterprise-class infrastructure, each service provider's responsibility for an effective crisis or incident communications strategy will grow exponentially. The way that each publicly handles service interuptions or security breaches reflects on the reputation and viability of the cloud computing industry as a whole. This year to date, high profile hosting services such as Google, Salesforce and Microsoft's cloud development platform have seen their blackouts cataloged by the media.
When one of Amazon's data centers was hit by lightening, disrupting services for five hours last week, one customer reportedly grumbled that "he always felt left in the dark when outages occur." An Amazon Web Services spokesperson pointed out in the article that AWS runs a status update page, but must be mindful of exposing vulnerabilities in its infrastructure that could lead to security risks. Still, for the most part AWS' customers took it all in stride, recognizing that some of the inconvenience is attributable to their own configurations or cost-cutting measures that affect service levels.
When Rackspace's hosted services recently went down for about 48 minutes, Rackspace turned to Twitter to update its customers (and by the nature of this medium the greater public). Brooks wrote: "Rackspace has possibly set a new record for transparency and accountability, if not customer satisfaction, by tirelessly tweeting the entire episode." On the flipside, however, at press time the reporter was still awaiting a reply to his request for a comment from Rackspace.
As cloud computing grows in popularity with the promise of cost-effective and secure data storage and virtualized enterprise-class infrastructure, each service provider's responsibility for an effective crisis or incident communications strategy will grow exponentially. The way that each publicly handles service interuptions or security breaches reflects on the reputation and viability of the cloud computing industry as a whole. This year to date, high profile hosting services such as Google, Salesforce and Microsoft's cloud development platform have seen their blackouts cataloged by the media.
Comments:
Links to this post:
<< Home
Just a clarification. We have never suffered a 24 hour outage - the outage from last week lasted about 48 minutes, although some customers took much longer to come online, depending on the complexity of their environments - but nothing close to 24 hours.
We love being able to update customers in near-real-time, and see that as an extension of "Fanatical Support".
Thanks for the article.
Rob La Gesse
Director of Customer Development
Rackspace
210-845-4440
We love being able to update customers in near-real-time, and see that as an extension of "Fanatical Support".
Thanks for the article.
Rob La Gesse
Director of Customer Development
Rackspace
210-845-4440
Rob, thank you for your clarification. I have updated the post to reflect this new information. Apologies for the inconvenience.
Interesting analysis on the respective responses from the co's that had the outages. ITDatabase's host has had a couple of hiccups (very brief outages). They are not acknowledged by the host, which I find alarming. If the host doesn't own it and point out that something went wrong and they are fixing it, it makes me feel uncomfortable (and wonder if it's likely to be a recurring problem).
BTW: One of the worst things about using a host (cloud or otherwise) is that when something goes wrong you are completely at their mercy. You can't troubleshoot or remedy on your end, and it's just an utterly helpless feeling.
Post a Comment
BTW: One of the worst things about using a host (cloud or otherwise) is that when something goes wrong you are completely at their mercy. You can't troubleshoot or remedy on your end, and it's just an utterly helpless feeling.
Links to this post:
<< Home
