In my last post we took a look at the costs of owning and operating your own equipment. We looked at the labor cost of people dedicated to just supporting your computer systems, operating them, and the systems software, and security that just creates and maintains the environment where your business systems run. We touched on buying large systems that accommodate your peak period processing even if that period only lasts a month or two on a seasonal basis. We looked at IT’s habit of over buying hardware to insure that the equipment meets its 2 to 3 year life expectancy and any unanticipated application requirements.
Now consider an environment where you employ no one to manage your computer hardware, system security, etc. Will that alone save you hundreds of thousands of dollars? Millions? Now consider paying only for the computer resources you use when you use them. You pay a low rate for your normal processing and then pay more when peak demand hits your system. If your hit with an unanticipated peak period, the system can be set to adapt automatically or at worst case you can increase capacity via a single phone call. In nearly all cases the price reverts to the lowest price when the peak period demand ends.
Consider disaster recovery. Reputable cloud providers have distributed systems that span multiple locations in geographic areas that protect you from natural disaster. Your application runs on all of these systems and both your applications and data are automatically distributed across multiple locations. When a user accesses your system they are routed to the data center and machines that provide the best response for them based on their location. If a cloud vendor were to lose a site the others would immediately pick up the workload with zero down time for you.
Do you have adequate skills on your staff to handle system security? Most cloud vendors provide both security experts that you could never afford to hire as well as sophisticated monitoring software that protects you and your applications. It is critical to the success of a cloud vendor that they protect their clients.
Middleware (software) that you often have to pay thousands of dollars to acquire and thousands more each year in support fees are often provided by the cloud vendor at no additional cost to you.
The downside of cloud computing is that in addition to a number of large, financially stable, and highly reputable vendors, there is an increasing number of vendors that do not provide the quality that the major providers do.
You need to know:
- Where are the machines located? Are applications and data distributed across multiple locations — note many small vendors have a single machine in a single location. Be sure you understand what the disaster recovery implications are.
- What do you pay for? Are there any hidden charges? The major vendors including IBM, Microsoft, and others have on-line tools that calculate your precise costs.
- What software is provided and included in your fee?
- What are the conditions governing seasonal or peak period processing? Will the price revert to the lower level at the end of the peak?
- What are the security resources available to protect you and your systems? Find out about hardware, software, and people.
This is a quick overview, read the book for a more in-depth discussion.