RapidCMDB, an open source federated CMDB solution for the enterprise

The problem is well known and stated over and over again throughout the last decade. Most IT organizations are structured around functional silos of expertise and too focused on technology, hence there is a gap between IT organizations and the business. It is mostly agreed that IT organizations need to move toward service management, but there are different schools of thought on how to get there.

My personal journey in this realm started over a decade ago when I worked for a very large system integrator/outsourcing company that was organized in very distinct, geographically dispersed silos.

At least some people in senior management was well aware of the problems inherent to this organizational model, and tasked the tools team to alleviate this problem and enable horizontal communications among the silos, slowly triggering organizational change. We had Unicenter TNG available in the organization and had to built a solution around it.

The solution was very imaginatively named "Information Window" and marketed internally to various departments. The core idea was to provide a unified view of the entire IT infrastructure, the services and their dependencies to all the users in the organization facilitating better understanding of the environment and more effective collaboration among number of silos. How is that for grand ambitions :)

Over the years, I've been part of number of projects where the buzzwords, methodologies, and tool sets have changed but more or less had the same objectives the Information Window project had. Over a decade later, we're still searching for better ways to solve the same problems, moving towards service management, better collaboration, more efficient processes, aligning IT with business, etc.

I've seen the same patterns from that first project in many of the projects I worked later on. The lessons I've learned stayed with me through the years.

  • Change is hard, organizational politics is fact of life, stop complaining about it and learn how to work with it.
  • There is more than one "true" way. Some level of skepticism (not cynicism) is healthy.
  • The more disruption there is to existing tools and systems, the more collaboration and participation is required, more likely you are to fail.
  • If the number of people/entities that needs to collaborate/participate is high, strong senior management buy-in and active sponsorship is a must
  • The more you have to rip and replace existing tools, the more opponents you will gain.
  • The path of least resistance is implementing solutions that can augment the existing solutions and show value with minimal disruption to existing systems (at least at first)

The purpose of this long narrative is to give some context to the choices we've made when designing RapidCMDB, an open source federated CMDB solution for the enterprise. There are two schools of thought about ITIL/ITSM/BSM/CMDB implementations. Idealists vs pragmatists.

Idealists emphasize how life altering these projects are for the organizations. Lots of preparation is needed, everybody needs to be educated (converted?), processes need to be changed etc. In short, full commitment of the organization and more importantly senior management is needed. Their conviction is driven from experience. There are too many examples of failures when full organizational commitment is not there.

Pragmatists on the other hand advocate more of a sneak attack. Forget about the ideal solution they say and look for the low hanging fruit that may have high impact. If possible, leverage existing systems to add value even if they are not exact fit, or implement the solution for couple of critical highly visible services. Fight the battles you can win so to speak.

As I stated above, I don't think there is a single true way of doing things. Different organizations have different needs and capabilities. If an organization is ready to make fundamental moves and senior management is onboard and ready to lead, the return from full ITSM/CMDB/BSM implementations would be higher for the organizations. And if not, no need for despair, pragmatist approach is a valid one.

RapidOSS (built on RapidCMDB technology) is a pragmatic solution. It does not compete with traditional CMDB products. It is designed as a federated solution from the ground up (not an afterthought) to leverage existing systems and data sources easily. We believe ITSM/CMDB/BSM projects don't have to be multi-year, multi-millon dollar projects, and it is feasible and may be even advisable to begin with an approach that leverages existing management tools and data sources, and what better way to give it a try but with an open source project?

RapidOSS v3.1 has just been released along with the website to support open development.

Looking forward to hearing your thougths.