The task of a data center operator is to continuously deliver service availability, compute capacity and energy efficiency. As data centers have become more complex and more mission critical, operators have sought greater quantity and detail in their information in an attempt to manage their facility.
Gathering data had traditionally been a slow and labour intensive process, with the information stored in a variety of formats and tools making it difficult to assess in a holistic manner. DCIM offerings are used to connect up or replace ad-hoc and incomplete monitoring systems across IT and facilities.
DCIM will typically monitor elements across both IT and Infrastructure including: ITAM Inventory Management, Auto-Discovery, Data collection, Cable Management, Monitoring & Alerting, Control, Trend analysis, Dashboards and increasingly important Integration (because to implement a complete DCIM system it is vital that the individual DCIM solutions share the same data).
Whatever the list of functionality drawn up by any one owner/operator the one thing in common with all of these listed above is that they are REACTIVE. While it is clearly true that “You can’t manage what you don’t measure” this doesn’t mean that measuring is all that is required. Data Center Management requires Measurement, Simulation and Predictive Modeling.
Where #DCIM fits into the #DataCenter landscape
DCIM is a positive part of the evolution of the management of data centers but it is not an end point in itself. DCIM offerings CANNOT PREDICT how the changes you plan to make tomorrow will affect your availability, capacity and efficiency. The only way to check your future plans before you make a mistake is to pass the data collected from DCIM needs to be to a solution that can PREDICT cause and effect.
The solution, in other words, is Simulation and that is another blog post…