Six Reasons A CIO Should Care More About DCIM

Add Your Comments

Matt Bushell leads Nlyte’s Product and Corporate Marketing efforts as their Sr. Product and Corporate Marketing Manager. Prior to joining Nlyte, Matt worked at IBM for more than ten years, helping to launch multiple products in their Information Management and SMB groups.

mark-bushell-tnMATT BUSHELL
Nlyte

I had the benefit of attending October’s Gartner Symposium, having also attended their Infrastructure & Operations show back in June as well as VMworld at the end of August. The symposium is rightfully positioned as catering to the CIO and their direct reports, and as a vendor in the Data Center Infrastructure Management space, I was wanted to see if and how DCIM would be a part of the conversation. Interestingly, while Infrastructure and Operations (where DCIM plays heaviest) was only one track of many, after meeting with Gartner representatives, attending multiple sessions, and walking the Expo floor, it began abundantly clear why a CIO should care about DCIM. Here’s some of my thoughts on the topic.

1. CIOs are moving from reactive to proactive planning and DCIM is proactive in spades with predictive planning. As Gartner analyst Dave Aron put it, the era of “Taming the Digital Dragon” is entering and “Riding the IT Horse” is ending. Sure, CIOs still have to run the business, but they are looking to capitalize on the huge advancements in big data and analytics and sensors to transform their business, proactively. DCIM fits into this perfectly given all of the information it collects, repeatedly, for the purpose of providing proactive solutions before problems manifest.

2. No longer a support organization, IT is commonly considered integral with the business – which happens to run on (well-managed?) infrastructure in the data center. CIOs are becoming more influential in the C-suite, and no longer are they just serving customers, rather, they are serving the business, with business-measured outcomes. But the business forgets that their infrastructure supporting all of this in their data center is frequently managed with non-best-of-breed solutions: Visio, AutoCAD and Microsoft Project (this is how one CIO report indicated they managed their recent data center consolidation.) These types of “solutions” are point solutions without controls nor governance.

3. You wouldn’t drive your car without gauges for speed or fuel (so why do you drive your data center without them)? I owe this elegant analogy to Gartner’s Bruce Robertson and his session on the Metrics Continuum. DCIM certainly measures what is going on in your data center, and you cannot improve what you cannot measure.

4. It is important not to measure, but to change behavior. Another maxim made by Bruce Robertson also discussed the habits of a workforce, and how to break the cycle. Data centers are notorious for having a high percentage of ghost servers consuming energy, or worse, older servers that are comparatively inefficient and thus wasting energy and money. DCIM can help track servers, what is running on them, how old they are, when their warranty and maintenance is up, and so forth. So DCIM fits perfectly here.

5. Data Center recommendation: think small, think dense and compute performance per kilowatt. This recommendation came from Dave Cappuccio in the Infrastructure and Operations overview session. As data center racks get more dense with more equipment in them, running at higher power, the need to know what is going in and where becomes imperative. And to measure compute performance per unit of energy, you are going to have to attribute power to each server (and each subsequent Virtual Machine) and which organization owns that server / VM. DCIM helps track all of this ownership and measurement which will help get you to the ideal of performance per kilowatt.

6. Move to an agility focus and change the game through IT as a Service (ITaaS). VMware’s session, “Becoming a Broker of IT Services – Management for the Cloud Era and the Software Defined Data Center” made this point. The idea was to segue from having an Infrastructure focus for CapEx savings to an applications Focus for Opex savings through automation to having an Agility focus and innovating. But even here, everything runs on infrastructure, and what if you move too many virtual machines to one physical part of your data center, even though you have the compute capacity? You might overload that rack (which has fixed power capacity) as compute cycles fluctuate, or you might overheat and run out of cooling (heating effects are non-linear with power in closed in locations.) Again, a DCIM solution integrated with a Virtualization solution can track your VMs, show you where they are and update them in real-time.

Industry Perspectives is a content channel at Data Center Knowledge highlighting thought leadership in the data center arena. See our guidelines and submission process for information on participating. View previously published Industry Perspectives in our Knowledge Library.

Add Your Comments

  • (will not be published)