We are watching the evolution of Data Centre Infrastructure Management (DCIM) systems and their adoption by data centres with great interest. It is similar to the evolution and adoption of Enterprise Resource Planning (ERP) systems in the 1990′s. The scope and complexity of some of these systems implementations nearly caused some large enterprises to go out of business. This was often prevented by the enterprise “upgrading” to a “lesser” system.
In the same way there are some large organisations with multiple global data centres who are currently upgrading from a major DCIM system to spreadsheets.
The reason is that they have often bought the “ideal” without looking at the reality. The cost in time and money of implementing the system can result in a negative ROI. A DCIM system is only as good as the data in it. So the population of the system is key. How long will this take? How accurate can this be when “go live” is reached? Can the accuracy of the data be maintained within the DCIM, in some cases globally at data centres of different sizes with different levels of management resource.
Automatic discovery and monitoring tools will, over time, improve the accuracy of data within DCIM and therefore the management information generated. However once again we are early in the evolution of these tools and should not be too influenced by the ideal. Unless we are implementing DCIM for brand new data centres with new IT and Infrastructure, possibly from limited vendors, we should be aware of the limitations of relying on automatic discovery and monitoring tools. For example, these tools cannot identify equipment which is switched off or not connected.
The DCIM system can become as mission critical in enterprises as the ERP system has become. Therefore the selection, implementation and progression are vital, and each organisation must look at their own priorities and resource, not aiming for the ideal, but good, continuing and improving ROI.
In starting this major project, an organisation must not overlook the basics; what is in the data centre, and to what level of detail and accuracy can the DCIM be populated and maintained?
Spending significant sums of money on DCIM for the really exceptional capacity planning does not help if the only budget/resource remaining means the data in the system can only be maintained at 70% accuracy. The business information from the system would probably be as accurate using thumb, pen, paper and calculator. How good is a dashboard if it is wrong?
On the other hand, if the initial budget is used to populate and maintain detailed and accurate infrastructure (asset) data in the management system, an organisation can work from here to evolve the system in manageable stages. Early ROI in the form of control on OPEX can lead to increased and better directed CAPEX both on the DCIM and data centre as a whole.
DCIM systems and tools are a necessity to business as the digital age progresses. By getting the basics right, implementing the DCIM can provide early business benefits without increasing or missing risks in the data centre.
Find out more about TRACKIT’s alternative approach to DCIM implementation at www.trackit-solutions.com