My Thoughts on Requirements Gathering nothing so simple sounding should be so involved.
Lets face it, requirements gathering is never the highlight of any project; it’s just that, a requirement, not a desire. It is boring, methodical and very “unsexy” with little to get exhilarated about. Why can’t we just jump right into the fun stuff. Of course, By fun stuff I mean building the solution or developing the magic button. Well if you have had success in our industry you know that requirements are the pivotal piece of any good project. When done properly they should provide a list of attainable items that the project will be able to cover. Many times it is prudent to also create a list of what it will not do. How else will you know if the button is actually performing magic?
Many different tools and methods exist to perform the “gathering requirements” phase. Some are very formal in nature like developing a series of use case models for each function. Others informal like a quick session to review what a product can do and selecting what is needed based on the delivered product functionality. The latter works quite well for COTS applications that are not meant to be heavily customized. The more you customize the longer the conversation needs to be.
I believe that the best way to perform this portion of any project is with a hybrid approach. The best tool(s) will depend on the size and scope of the overall task at hand. If you are replacing a sink you do not need to discuss the heating system and how it makes hot water. Correct? Well the same mindset works for Projects. Whenever possible, Keep it simple.
When archiving In the Data Management world using the IBM Optim suite the requirements generally revolve around identifying what data is acceptable to remove from the Day to day operation of a given system or environment. Once an organization has identified the need to manage the growth of its data then with Optim we can provide a proven solution to help them accomplish this requirement.
An organization will have a need like : Help us to Effectively Manage Data Growth
The Requirements could be pretty straight forward:
- Improve application performance
- Control data growth to save hardware & storage costs
- Delay CPU/MIPS upgrades
- Streamline application upgrade process
These are vague and pretty high level requirements but with the right toolset they can all be accomplished and we can effectively corral that data.