* The very model of a modern management system
Get ready for the models – no, not the Sports Illustrated swimsuit models. As management vendors start to develop their next generation of management tools and strive to help customers develop more efficient IT processes, they are increasingly turning to model-based management.
The use of models in management is not new, but the increasing use and interest in them is. In the past, models have been used in capacity planning and in the development environment, as well as in other management products. But models fill the need to define in a structured way repeatable IT processes, configurations, workflows, data schema, best practices and more.
Microsoft, as an example, is using the System Definition Model as a core component of its Dynamic Systems Initiative (DSI) management strategy. And other management vendors are also turning to model-based management.
There are different kinds of models. Some are just workflows of processes or sequences of events. These models tend to be generally sequential in nature. And they lend themselves to process-oriented activities.
Then there are models that attempt to characterize more complex problems or situations. Models that accurately reflect reality can enable automation and predict behaviors (or performance) if changes are made to the environment. For these types of models that involve more complex techniques, the important issue here is how accurately the models reflect reality – because models are just representations of reality.
The issue with a model is its ability to replicate the many inter-related variables that can affect the results, particularly when it is used to predict future behaviors. So it is important to know how accurate models are, and if they account for the inter-relationships of the components and variables that are essential to accurately model the situation or environment.
It is important to understand what management vendors mean when they say their products are model-based. You need to understand what the models are depicting, how they are put together and how accurate they are. Be sure you’re comparing apples to apples – and not equating simple workflow capabilities to complex modeling techniques. Also, if a vendor uses more complex modeling, you need to know how accurately its model reflects real-world environments.
Finally, models are useful, but if they are too difficult to create, they won’t be used. So investigate the process of developing the model – is it a totally manual process or it is assisted with some automation? Dynamic models need to be updated, and it’s important to keep the models up to date. How easy is it to do that?
If you’re not already working with models in your management tools, prepare yourself – because you probably will be in the not-too-distant future.
