The EC2 Outage of 2010: Amazon Makes It Rain … Doubt Over Public Cloud #amazonoutage
I was left pretty speechless after reading this email below, which was sent out to Amazon EC2 customers in the aftermath of the recent outage of their cloud platform. Recently, I had been pretty conservative and cautious when speaking to customers about Cloud adoption.
Amazon’s outage and consequent response to their own mistake pretty much sums up most of my fears about Cloud: the lack of accountability cloud providers will assume for their customers data, as well as basically treating yours and customers/clients data and resources like a utility.
Hello,
A few days ago we sent you an email letting you know that we were working on recovering an inconsistent data snapshot of one or more of your Amazon EBS volumes. We are very sorry, but ultimately our efforts to manually recover your volume were unsuccessful. The hardware failed in such a way that we could not forensically restore the data.
What we were able to recover has been made available via a snapshot, although the data is in such a state that it may have little to no utility. If you have no need for this snapshot, please delete it to avoid incurring storage charges.
We apologize for this volume loss and any impact to your business.
Sincerely,
Amazon Web Services, EBS Support
The above email looks and smells like the response I get when my cable provider goes down for a day and a half: the company apologizes for the inconvenience of no cable service, and I get thrown a modest credit for the day of no service. In no way, shape, or form is Amazon saying that they are going to reimburse their customers for the loss of data or productivity. Maybe instead of an apology email, Amazon should send a form to their customers to quantify the amount of data that was lost and reimburse their customers based on their information.
Amazon’s outage, coupled with other recent outages, definitely confirms the fears I have had with the concept of Cloud in general. In the “Dotcom” era, I previously worked in both hosting and managed services at the beginning of my IT career. During that period, I watched the likes of Exodus, Worldcom, and other organizations I worked with and for dissolve due to poor business models, lack of accountability, and loose SLA’s. The Cloud is not much different than what was offered back then, outside of the flashy acronym and possibly more scalability to automate the use of resources that are needed on demand.
My top three biggest concerns currently when it comes to Cloud are:
- What is the accountability of the Cloud provider that I have partnered with?
Notice that I used the word “partner”. Keep this in mind as well: when I start to run my mission critical applications, do I want to send that to a utility company or someone that is going to take the same sense of ownership of my customers data? - My cloud service is cheap for now … ?
What prevents my cloud provider from getting me to sign in at a low rate, and then after the contract is signed, sticking me with an increased rate? Much like the Comcast Cable “Triple Play”, which offers to decrease all of your phone, internet, and cable down to $99 bucks/year, only to find that a year later your bill will be double the advertised rate. Also, when I want to choose another provider, am I going to have to pay outrageous amounts of money to migrate from the Cloud provider’s platform? - What is the stability of the Cloud provider?
By stability, I mean what is the stability of the platform which I am assigning my critical business data on. In the Cloud model, users really will not have visibility into what redundancy is in their server environment. As a customer, for all you know your data is sitting on a USB hard drive connected to a server sitting on a coffee table. This is why SLA’s are extremely important when deciding on a cloud provider. Stability also means where the Cloud provider will be in three to ten years from now, as moving to the Cloud is not a short term project. Switching to Cloud is a long term business commitment, and the value of this system is depreciated vastly if I have to bounce around from provider to provider as they go out of business. This issue would cost organizations more money in the long run rather than being cost effective to the data center.
When I look at Cloud adoption overall, I would say that Amazon has been one of the bigger players in reaffirming the fears of many potential adopters. I believe that this event will make people take a step back and also slow adoption exponentially. Time will tell if Amazon’s debacle was a speed bump or road block; but at a minimum, I believe the caution flag is out for this lap at least.
Photo Credit: Ewon Thot

