Start to create infrastructure visibility

Of course, every organization has a formalized process that kept record what was purchased. This process is known as configuration management, with associated Configuration Management Database (CMDB) as central register. A system which worked fine in those days of monolithic systems, but now starting to show cracks.  

Not only due to change of  technology but also as result of mergers between companies. As a result, data from the configuration management became slowly distributed and an overall view at the infrastructure is complicated. 9 out of 10 organizations have a greater or lesser degree of burden and make decisions that are not entirely founded at the real state and value of their infrastructure.

To excel or not

An old saying is: “You can not manage what you don’t know”, so at various levels within IT organizations information is recorded to carry out daily activities. These well-known and  ever-present Excel lists are maintained due to the fact the central repository is insufficient. If these lists are shared and reliable it wouldn’t be such a problem. Unfortunately they are often incomplete, out-dated or irrelevant just like the configuration database.But above all have Excel has limited opportunities to capture the dependencies and relations between various components and services.

It's all about a relational database, which allows to consolidate data from different sources, which can also be external. So some kind of 'data mining' can be used at the metadata of the infrastructure. Through correlation of data patterns can be recognized that help in decision making.  Enrichment of data is all about a shift in priorities from what and where to why and how. Addressing questions to do it your self or outsource require knowledge about the net worth of IT to the business. This is not only about economics but also the environmental awereness, which become slowly more important by stakeholders and customers. 

Intelectual property or proprietary
Easiest way is to choose off-the-shelf commercial products with regular modulair updates instead of spending some time to understand the needs. Unfortunately, this complicates establishing a single repository which can be used as an oracle now and tomorrow. Actually it will result in a plurality of configuration management databases. Different databases, focussed to many domains or silos are available but all contain pieces of the puzzle. Unfortunately these databases or the dataschemes are often proprietary and make it difficult to export data. with most often proprietary database schemes. A relational database with ease to import and export allows to find relations and perform 'data mining' on the metadata of the infrastructure. Such repository is advantageous in several aspects:

1. Today technology flexibility to support tomorrow processes
2. Facilitates enrichment of data with data from other processes
3. Prevent a sprawl of overlapping functions

Service knowledge management

Those familiar with ITILv3 know above is exactly what Service Knowledge Management is all about. The boundaries of the datacenter are becoming blurred and it's all about the service. Nevertheless, the need for information on the infrastructure remains necessary. Information can come from various sources which are not always self-managed. To incorporate this information within the processes it helps if data can be imported. That simplifies the problem to relate data and use queries to find patterns to be used in decision making.

Different suppliers of Configuration Management solutions do this by providing an extension that allows insight into application dependencies. This is done in an automatic manner by gathering communication patterns and relate these to configuration items. Or manually by mapping the organizational relationships, often stored in people's minds or those excel sheets. Unfortunally the combination of both is rare but who can combine the already present knowledge is with data obtained from the online infrastructure gets an oracle that is not only able to support today decisions but also for tomorrow. 

To improve visibility it’s time to consolidate the scattered but relevant management metadata into a central repository. So already available knowledge can be reused to support decisions about the infrastructure.