In this blog, the fourth part in our ROI series, we’ll do a return on investment (ROI) analysis for Data Center Infrastructure Management (DCIM) software. If […]

ROI for Network Documentation & DCIM Software: DCIM Analysis

In this blog, the fourth part in our ROI series, we’ll do a return on investment (ROI) analysis for Data Center Infrastructure Management (DCIM) software. If […]

In this blog, the fourth part in our ROI series, we’ll do a return on investment (ROI) analysis for Data Center Infrastructure Management (DCIM) software.

If your main pain point is a lack of control over your Data Center spending, inefficient use of your space, power and cooling — and/or overall deficient asset and cabling management — and you’re reading this because you’re looking for a potential solution, then this blog and the next one (in which we go over a numerical example) are for you. Many of the cost savings associated with a network documentation system (which we analyzed in the previous 2 blogs in this series) also apply to DCIM — so, if you’ve read those blogs, you’ll notice similarities.

As we’ve written about before, DCIM can solve a myriad of different pain points in the data center. Two key areas in which DCIM can help is through better planning and reduced downtime:

  • DCIM can help you identify stranded assets, unutilized network ports, zombie cables, underutilized switches and routers, extraneous licenses, applications and more
  • Downtime is reduced with a DCIM tool (and, in this sense, it doesn’t differ that much from a network documentation system)

ROI based on saved rack space

The main area in which a DCIM solution can really help you save money may be through improving your capacity planning. Because netTerrain DCIM is able to do active, real-time power monitoring (as we’ll dive into below), you’re able to improve rack capacity allocation based on actual power usage instead of nominal, or derated power values, which leads to wasted rack units.

The more rack units you need for the same amount of assets, the more racks you consume and the more space you need (which, of course, has a higher cost).

The Uptime Institute estimates that the cost-per-rack-unit for a Tier 3 data center is $1200 — without the equipment in it. The industry average rack utilization hovers around 60% (if you’re not doing any proper power monitoring and rack optimization). If you’re able to monitor your racks in real-time, you may be able to oversubscribe them and improve your rack utilization by as much as 15 to 25%.

Using SNMP, IPMI, and other protocols netTerrain can monitor up to one year of temperature and power figures and create time series and profiles with granularity down to 30 seconds. As I mentioned above, if you use vendor-independent and out-of-band management, you can put most of the rack devices under monitoring and obtain actual real-time power and temperature figures.

Instead of planning server additions to your racks using derated, or nameplate, power figures you can use real power consumption profiles and oversubscribe your racks. This leads to more devices being added per rack which ultimately means that you can use less racks for the same amount of devices and, in turn, leads to savings in space allocation.

How much savings are we talking about?

We’ll do an actual calculation in our last blog in the series but, using Uptime’s estimate of $1200 per rack unit saved, you start to get the idea…

One common concern is: what happens if there’s a peak in power consumption that goes over the power threshold?

With netTerrain’s Environmental Module, this can be prevented by power capping your racks. Power capping means that you are putting a limit on the max power drawn from certain devices inside the rack. With the IPMI protocol, this can be done on a granular (server-by-server) basis by controlling the CPU cycles based on a priority schedule on your servers. If you want to know more about this, check out our EM module on our website.

ROI based on reduced downtimes and improved IT efficiencies

Just as we saw in part 2 and part 3 with the analysis of a network mapping software, DCIM software can also help reduce mean-time-to-repair (MTTR), mitigate the amount of downtime, and improve IT efficiencies by improving your overall IT visibility.

As we discussed in part 2, bad IT visibility comes in many forms but it’s usually related to stranded or under-utilized IT assets like routers, switches, servers, power equipment, and more. With the advent of hybrid networks, you can add virtual instances and cloud instances to that mix. DCIM software brings it all together under one visual umbrella to provide you with enough context to take adequate corrective actions and seize opportunities for optimization. The figures and process we are going to use for this part of the ROI are similar to parts 2 and 3 in this blog series.

Finally, just like with network mapping software, netTerrain DCIM gives you insights into the data center that help improve operations and reduce mean-time-to-repair, including:

  • Easy visualization cues and filters to identify single points of failure in racks, cabinets and devices
  • End-to-end circuit layout records, from the site and building level down to rack device and port
  • Dependency mapping between devices, racks, and other assets affected by an outage
  • Physical to logical network mapping
  • Quick contact information and key detail lookup

In this series’ final blog (part 5), we’ll work on an actual calculation for your DCIM software’s ROI. Stay tuned!

Jan Durnhofer
Jan Durnhofer
As CEO / Product and Engineering Manager, Jan joined Graphical Networks with the purpose of creating the most advanced DCIM and IT visualization company in the market.

Leave a Reply

Your email address will not be published.

See what netTerrain can do for you

Start My Free Trial Schedule My Demo Get My Pricing Quote Find My Product

Stay Informed!