That there is a lot of hype in the technology industry about “the cloud” is pretty obvious these days. You will be hard pressed to read an article about technology (this one included) without some reference to the beauty, ease and affordability of cloud services.
The argument is that IT infrastructures have become too complex and fragile for the pace and dynamism of modern day business. Champions of everything to the cloud are quick to point out that over 70 percent of current IT investment remains focused on maintenance. Worse yet, it is argued, users are clamoring for faster response times and of course management wants all the good stuff but are unwilling to pay for it. So, cloud computing to the rescue.
It is true that cloud computing holds real promises of agile, efficient IT service delivery that could dramatically reduce the complexity surrounding modern computing infrastructure, significantly reduce costs, and lead to greater business responsiveness. The concern, especially for the small business owner is that this promise of “ease” will be carried to an extreme and due diligence will be a victim. There seem to be a tacit assumption that once it is in the cloud, it must work and there is no need to do anything else.
Vendors who are pushing these offers of “a new model that cuts through IT complexity by leveraging the efficient pooling of on-demand, self-managed virtual infrastructure, consumed as a service” are not doing enough, in my opinion, to educate prospective candidates of cloud services about what lies ahead and what could happen should the cloud generate some thunderstorm and lightening. The “just put in the cloud” marketing blitz is using an overly simplistic approach that tend to trivialize the complex workings of cloud services.
For example, when a vendor convinces a merchant to turn payroll over to the cloud, are there fall-backs in place so that employees can still get paid even if the cloud service is unavailable. How about payment processing? Do you stop collecting payments when your online credit card system stops working or can you still collect payments and sync with the provider when service resumes? Is there a possibility that you (the subscriber) could lose your data, encounter massive disruptions and incur additional costs?
Take a look at the big guns of cloud services who have suffered some form of outage in recent times – Amazon, Yahoo!, Google, Microsoft, Skype, The Planet, Intuit, Netsuite, Hostin.com, Twitter, EMC Atmos, Salesforce.com, etc. Obviously these giants of cloud services did everything in the books to make sure they achieved the enviable five nines of uptime. But as we have seen, when stuff happens, it happens. The question is how prepared are subscriber to cloud services when there is an outage. When the Amazon outage came down, only a few companies had the foresight to pay extra money for fail-over locations and were thus able to avoid significant down time. The key word there is “extra money”.
We have read reports of businesses that lost lots of data because they put all their trust on a cloud service and got burned. The Amazon outage which was sparked by an error made during a network configuration change led to disruptions and service outages for its Elastic Compute Cloud (EC2) and Relational Database Service (RDS) customers and lasted for about four days. That is a lot of time to close shop if you had no alternative. The main reason most people were caught by surprise by the Amazon outage was because “it (Amazon’s Cloud Services) had worked well for so long and we did not think this would ever happen”.
How prepared are you to handle the disruption to business, and services to customers when lightening strikes; and how patient are you willing to be when the cloud service is offline and the provider is “working feverishly” to resolve the problem?
For example, when Intuit’s online accounting and tax services went down due to a power failure that occurred during a routine maintenance procedure, the incident affected the company’s primary and backup systems, and took down a number of the company’s Websites and services. The result was that as many as 300,000 small and midsize businesses were left in limbo until services were restored – several days worth of limbo.
So what now? The key to realizing the advantages of cloud computing is tailoring the cloud model to work for your business and your approach to IT. The bottom line is that we have to go back to basics – step back if you will, and take another look at the plan for cloud services if there is one. If there is none, grab your service provider or IT staff and hash one out.
What exactly are you purchasing and what is plan B? Do you understand the technology you are buying? How long can you afford to be closed for business when something goes wrong? Does it make sense for you to do a 50/50 implementation where you have your critical services on a fail-over server on premise or at a different geographic location (from the current host)? With virtualization technology and the falling cost of high-end hardware, one server can do a lot these days, does it make sense to have a box sitting in your office that you can fire up and at least get some work done when the cloud goes dark?
The IT world is used to outages and disruptions and can handle it very well. The mind-set of a business owner is different. It is no fun when you watch thousands of dollars slip away because your website is offline or the payment processing cloud service is down. At the end of the day, disaster recovery and fail-over strategies are key in dealing with outages in cloud services.
If you are a small business owner considering moving your critical applications and services to the cloud, the best path will be to have a subject matter expert assist you in the process.