Transforming IT
Showing results for 
Search instead for 
Do you mean 

Increasing Data Center Energy Efficiency: DCIM Is No Panacea, But You Can Make It Work

bimialmi on ‎04-12-2013 09:22 AM

I love technology. I love data. Put the two together and you have a potent solution that enables analyzing, reporting and taking action based on a constant stream of real-time information. This information can come from a number of diverse sources, used in many different applications and industries – healthcare, manufacturing, technology and even space exploration (remember “Houston we have a problem?”).


But after reading an online Data Center Journal article by Sev Onyshkevych on the elegance of data center infrastructure management (DCIM) tools, I’m reminded of the perils of becoming too reliant on technology to operate and manage our data centers. This informative article on DCIM, called “The 5 Pillars of DCIM – Improving Data Center Efficiencies”, lays out a framework for a robust yet complicated assemblage of all IT and facility systems accessible from a single dashboard.  While this sounds like a sexy solution to an otherwise messy situation, care must be taken in aggregating otherwise disparate domains.  


Don’t forget - I am a card-carrying evangelist of DCIM tools and the use of data acquisition and analysis to help us move forward as an industry. I am also the first person in line to expound on the importance of assimilating the IT and facilities domain together to form a more effective and efficient uber-domain, in which we will find the greatest energy savings and operational efficacy. However, before we can reach this data center nirvana, I think we have some homework to do first:


1. People need to lead the charge. It is incumbent on the CIO and COO to set the tone and provide marching orders to move toward an integrated IT and facilities model. This will require some outside, “non-interested parties” to facilitate this process and develop a roadmap. Without third-party consultants, internal politics could spoil the process.


2. Baseline data must be acquired. A Buddhist saying comes to mind here: “If you want to know your past, look into your present conditions. If you want to know your future, look into your present actions.” This applies to both IT and facilities. Knowing baseline performance data will set realistic expectations and help develop metrics for future benchmarking. A specific example here is measuring energy consumption in different areas in the data center and for different pieces of equipment. This will help in understanding the ROI that might come out of the investment in upgrades identified by the DCIM tools that can monitor data center energy use.


3. Examine the current IT hardware and networking plan. Any major virtualization, data center consolidation and equipment refresh ideally would occur concurrently or prior to a DCIM program. This eliminates redundancy or unnecessary work in the development of DCIM tools.


4. Performance validation. After the DCIM system is installed, thorough validation has to happen to make sure the system is monitoring and reporting correctly and that the claims made by the vendor are actually true!


Certainly these aren’t the only four steps, but they are indicative of the types of activities required to make a DCIM system a success. Ultimately the system must be user-friendly, not overly complicated, nor promise too much. This is especially critical in emergency situations where access to data cannot be slowed down by a convoluted interface and data that is buried too deep to access at a moment’s notice. These and other potential pitfalls can be avoided by developing a thoughtful, detailed roadmap containing all necessary steps for a DCIM deployment in planning, execution and on-going operations.


Learn how HP Critical Facilities Services can help you increase efficiencies in your data center.


To learn more about me and how I can help your organization rein in power consumption and curb energy costs, visit my HP Technology Experts profile.

0 Kudos
About the Author


on ‎10-11-2013 12:51 AM

Data Center Infrastructure Management (DCIM) is these days (and some of my nights as well) a hot subject for me and the team here in Asia Pacific.
I agree with the 4 steps you proposed here but also wanted to add some elements related to the risk of DCIM for people going too fast in the technology selection and implementation.
In many of the cases we did study we realized that number of time DCIM start with great expectations but run in major implementation troubles due to poor planning or low  product functionality understanding.
So to me the key recommendation is one more time to go in a "maniac" requirements and scoping activities when it come to DCIM because it is a complex domain that many vendors try to over simplify generating a fair number of white elephants in the market these days.
 Data Center Infrastructure Management is more a system integration process than just a software basic implementation. People sometime want to relate DCIM as the ERP of the Data Center it is true if you look at what was the ERP 15 years ago not what it is now.

27 Feb - 2 March 2017
Barcelona | Fira Gran Via
Mobile World Congress 2017
Hewlett Packard Enterprise at Mobile World Congress 2017, Barcelona | Fira Gran Via Location: Hall 3, Booth 3E11
Read more
Each Month in 2017
Software Expert Days - 2017
Join us online to talk directly with our Software experts during online Expert Days. Find information here about past, current, and upcoming Expert Da...
Read more
View all