Cloud server hosting has been around for several years, and has definitely grown past the “early adopter” stage into a mature and thriving market. Having watched the market reaction to IaaS, and having spoken to many businesses which have moved their systems to the cloud, I’d like to share my impression of the public sentiment which I’ve seen from my end.
The classic arguments in favor of Infrastructure-as-a-Service revolve around cost-savings and convenience. Essentially, the argument is that you can save precious IT time by eliminating manual labour associated with hardware management, upgrades and maintenance. You also supposedly save money by eliminating over-spending on under-used hardware and resources.
There are definitely some truths when it comes to this. However, I’ve found that the reality of IaaS can change depending on who the target audience is.
Within smaller businesses, cloud hosting definitely makes financial sense from the simple fact that these businesses no longer have to worry about setting up their own server rooms.
Especially for smaller companies which handle sensitive information, the security measures implemented by cloud host data centers are superior to what could be financially feasible with an on-premises server room. And the fact that it’s available on a pay-as-you-go basis means that there are no capital expenditures required to implement or modify the data center environment.
Once a company has grown and established themselves, the benefits of cloud become slightly more complicated. Although the marketers would like to have you think that you’re simply moving servers to another location, it’s actually somewhat more difficult than that.
I’ve often heard others in the industry equate IaaS with virtualization. The rationale is usually something along the lines of “You already virtualize your servers. Why not let someone else worry about the hardware?”
Cloud computing is an entirely different way of managing IT… and it requires a completely new skill set. For time-pressed, overworked IT administrators, the idea of having to learn a whole new way of working in order to modify something which already works fine is not always compelling.
And of course, complex implementations rarely ever work flawlessly the first time. There is always a period of tweaking and adjustment. For a new system, this is acceptable. However, this kind of risk is unacceptable when dealing with established systems which support business-critical processes.
Although cloud hosting may have been seen as a security improvement for smaller companies, they might be seen as a potential security risk when it comes to larger organizations. When sensitive data is stored on premises, it provides a sense of control and safety which is hard to replicate when the responsibility for protecting this data is placed into the hands of a third party. This is particularly true of companies which have already invested in the physical and technological security of their IT environment.
Resistance to change is further compounded by tough economic times which make it difficult for perceived cost-centers such as IT to obtain budget increases. When belts need to be tightened, companies prefer to spend just enough to keep the engines running… and invest more heavily in revenue-generating areas of the business. (This is why it’s important to maintain focus on IT projects which support strategy and drive growth. But that’s a topic for another article)
Another potential sticking point would be the feat of vendor lock-in. If servers are moved to a cloud provider, there may be a high switching cost if the cloud host offers unsatisfactory service. Efforts to eliminate lock-in have been addressed through the creation of various industry standards in recent years. But this is still a point of concern amongst potential IaaS customers at larger or established companies.
However, there is one area where cloud hosting is particularly well-suited for use by larger companies. When it comes to temporary or emergency capacity requirements, cloud servers are an excellent alternate It simply wouldn’t be feasible to purchase and set up new hardware for a resource-intensive temporary project. Also, cloud servers make excellent emergency facilities for hosting critical applications in the event that the primary site is destroyed or becomes unavailable.
As mentioned before, this is an oversimplification of the pros and cons of IaaS which have been relayed to me through numerous discussions with other professionals in the field. A more detailed analysis of the realities of the cloud would require a much deeper and more granular look into each of the issues mentioned above. This may be an ideal topic for a future article.
In the meantime, please leave a comment and let us know your point of view on this issue.
Image courtesy of twobee at FreeDigitalPhotos.net