This is Part 5 of our five-part series on the countless number of decisions an organization needs to make as it embarks on the DCIM purchase, implementation, and operation journey. The series is produced for the Data Center Knowledge DCIM InfoCenter.
In Part 1 we gave an overview of the promises, the challenges, and the politics of DCIM. Read Part 1 here.
In Part 2 we described the key considerations an organization should keep in mind before starting the process of selecting a DCIM solution. Read Part 2 here.
In Part 3 we weighed DCIM benefits versus its costs, direct, indirect, and hidden. Read Part 3 here.
In Part 4 we pondered the challenges of implementing a DCIM solution. Read Part 4 here.
The preceding four parts of this series examined vendor promises, purchasing guidelines, potential benefits, as well as the implementation challenges of DCIM. The long-term promise is gathering data across all areas and creating a centralized database used to deliver unified visibility for all management and technical domains. If this broad-scale vision is finally delivered, DCIM should be the solution that finally allows IT and facilities managers to work together to make more informed and presumably better decisions about overall energy usage, operational optimization, and workflow optimization. And now for the most elusive task, which is identifying which of the myriad of DCIM functions can produce the most quantifiable financial payback and provide clearly definable cost justifications.
Simple Payback - Facility Energy Savings
Let’s start with the simplest cost-justification example: the payback based on energy savings. This depends directly on several primary factors: the present facility energy efficiency, the potential projected efficiency improvement, and last, but certainly not least, the cost of energy. For example, in a large older facility with relatively poor energy efficiency (i.e. a PUE of 2.5), the largest energy waste will typically be the cooling system. While a DCIM system by itself will not deliver as much of an improvement as replacing an old inefficient chiller, it can identify, analyze, and help tune the operating parameters of facility equipment as well as optimize the airflow issues in the whitespace.
As a hypothetical case, if we assume an IT-equipment draw of 1,000 kW, at a PUE of 2.5, the facility overhead requires an additional 1,500 kW, for a total site draw of 2,500 kW. This results in an annual energy consumption of 21,900,000 kWh (based on 8,760 hours per year). Assuming an energy cost of $0.10 per kWh, the annualized energy cost would be $2.19 million. Assuming a DCIM system helped improve cooling system efficiency though optimization, thus reducing facility overhead by 20% to 1,200 kW, the PUE would then be 2.2 (a conservative projection). Therefore, the annualized energy cost would be reduced to $1.927 million – a direct saving of $263,000 per year.
This offers a well-defined basis for an ROI payback calculation. However, if your cost of energy is lower or higher, it will obviously change the ROI. In the US, the cost of energy can range from $0.025 to $0.25 per kWh. So, as can be seen from the above case, the cost of energy will directly impact the annualized cost saving ($131,500 at $0.05 per kWh to $526,000 at $0.20), either diminishing or enhancing the ROI.
Conversely, if the site is a newer, more efficient facility and already has a relatively good PUE (i.e. an initial PUE of 1.5), it is less likely that DCIM will be able to provide anywhere as much in energy saving. So, while any energy efficiency improvements are always welcome, in this case it would make it very difficult to make energy saving the primary basis for cost justification.
Facility Capacity Recovery and Resource Management
Capacity management is another function that offers a solid basis for cost-justification purposes. Virtually all data centers have a mismatch between available facility design capacity and actual usable capacity. Many data centers have stranded capacity because of the divergence of space, power, and cooling capacity, primarily related to IT provisioning at the rack level. This represents a waste of the large capital investment in the data center. DCIM systems can substantially improve resource utilization, resulting in a significant cost justification.
According to a 2008 white paper by the Uptime Institute, it can cost as much as $25,000 for each kilowatt of critical load capacity (for the infrastructure equipment, in addition to the base cost of the building shell), to build a Tier IV data center. For this example, we assume that the data center can effectively utilize only 70 percent of its rated design capacity (which is not uncommon). This clearly represents a significant and costly waste of capital investment. If we use the same 1,000 kW critical IT load, this would represent a potential capacity loss of 300kW, which represents a wasted investment value of $7.5 million (300 kW x $25,000). From a practical point of view, DCIM will not be able to recover all of this lost capacity. However, it is reasonable to expect DCIM to increase the usable capacity and potentially reclaim 100 kW of stranded capacity, which would result in a capital investment recovery value of $2.5 million – a very tangible basis for cost justification.
This is not just a one-time fix, since this functional disparity can occur continuously over the operational life of the facility. The primary reason is because the facility design is based on a fixed ratio of space, power, and cooling, while IT hardware is being constantly upgraded, and their power and cooling requirements change. This results in less than optimal, often reactive, IT equipment deployment strategies. In effect, DCIM would continue to provide continuous capacity optimization by modeling of placement of IT hardware, which would help minimize stranded capacity over the life of the data center.
Of course, these are simplified examples. Your organization’s cost basis for the facility will be different and you may need to engage your financial department to make a more accurate case, using Net present Value (NPV), depreciation, tax benefits, and any other more sophisticated financial analysis to create a solid ROI case for senior management.
The Value of “Priceless”
While the above examples of improved energy efficiency and usable capacity have a clear and direct economic payback, other benefits, such as labor saving and faster deployment, that result from better workflow management are obviously good. Nevertheless, they are harder to quantify on a spreadsheet. There are other important benefits, such as early detection of facility systems equipment failures by continuous monitoring of energy usage and the performance of cooling subsystems (chillers, pumps, fans, etc.), that can improve overall availability. Also, DCIM’s ability to do real-time branch-circuit and rack-level power monitoring can avoid tripped circuit breakers, which can directly help avoid an outage. This also negates the exposure to human error caused by workers performing manual electrical load surveys on energized panels, which could lead to an outage. Moreover, since energized panels remain closed, it reduces exposure to worker injury, which is now subject to more restrictive OSHA limitations. Of course, besides risk reduction, there is the direct benefit of reducing the recurring labor cost of these repetitive manual branch circuit surveys.
When calculating the value of improving overall availability and avoiding an outage, the word “priceless” comes to mind (to borrow a popular advertising tag-line). Again, while these types of benefits are unmistakably “priceless,” it is hard to define their direct financial value on the ROI spreadsheet. So, why are they so difficult to value? Especially since we ultimately judge a data center by its uptime, typically expressed by the number of “9”s. This was the original underlying basis used to justify to senior management why they should initially invest significant capital in highly redundant systems because of widespread acceptance that redundant systems are a “must have” for ensuring 7x24 operations. Yet, it remains more difficult to make a clear financial case for the value DCIM provides to monitor and enhance the reliability of its operational status.
In addition to the above examples of improved availability, there are many other quantifiable facility-based cost justifications, such saving from proactive and predictive maintenance. However, if energy reduction or capacity optimization does not represent a substantial portion of the short-to-medium-term cost justification, the payback of any other potential facility benefits will be that much tougher to use as the primary basis to make the facility-side case for DCIM.
IT and Computing Resource Management
The facility has some direct cost justifications that can be identified. However, facilities systems change at a near glacial pace compared to the ongoing IT hardware changes and rapid metamorphosis that “open everything” brings to computing architectures. To one degree or another, some of today’s DCIM systems have the ability to measure IT-device-level utilization, correlating the impact of software on system throughput and its relationship to power consumed. This obviously requires close cooperation across and between facilities and IT (an invaluable mandate with or without DCIM).
In the long term, DCIM capabilities will increasingly need to focus on monitoring and managing virtualized assets and the coming of age of the Software Defined Data Center. This means that DCIM systems will not just need to collect and integrate facility-based information (assuming that the organization still operates its own data center), in addition to IT hardware, virtualization software, and application information. Moreover, DCIM will also have to be capable of querying multiple external cloud platforms (IaaS, PaaS, SaaS, etc.). These may be spread across diverse service providers (which currently all have proprietary consoles and tools).
Eventually, if promise of DCIM as a truly holistic management system is to come to fruition, it must be able to provide an integrated comprehensive view of any and all internal, external, virtual, and physical IT assets (even if they are geographically distributed, which is becoming more common), to create a global view of the status and capacity of all computing resources. From a TCO view, this means DCIM should be able to provide price-based modeling (e.g., localized cost of power, site PUE, IT resource efficiency, cloud service costs), capable of analyzing and optimizing the entire operational cost model. Only then will DCIM become not just a tactical tool to optimize resources and manage operational workload allocations, it will also be a truly financially strategic, high-level TCO modeling tool for adapting to dynamic, ever-evolving circumstances.
The Bottom Line
We have now identified some of the simple facility-based examples for cost justifications, as well as other less direct, yet valuable benefits. Nonetheless, should DCIM now be viewed as a long-term strategic investment or still as just another hyped up product? In 2014 there were approximately 75 vendors, consisting of both small startups and major well-entrenched players. Since DCIM as a category originally burst on the scene over six years ago, Gartner’s Hype Cycle seems to have perfectly depicted the way DCIM has been perceived by potential buyers. Starting with the “Initial Peak of Inflated Expectations,” followed by the depressing “Trough of Disillusionment,” and then moving upward on the “Slope of Enlightenment.” DCIM vendor offerings may finally be entering the last stage of the cycle: the “Plateau of Productivity.” DCIM has now moved far beyond just a glorified PUE console, and offerings are now in the second- and even third-generation stages of development. DCIM can now function as a useful solution set beyond physical aspects of the facility and begin to meet the promises of a holistic overarching toolset to align the physical and logical elements of the data center.
Finally, it is important to consider that the data center is somewhat like a widget factory, processing and delivering information rather than widgets. Like any other manufacturing process, management expects it to be cost-effective and deliver a measureable Total Cost of Ownership, based on CapEx and OpEx values in one form or another.
When evaluating DCIM’s potential cost justifications, start by considering your organization’s culture and its strategic vision for computing, not just the physical data center. The maximum benefits of a DCIM investment will be only realized by what your own organization does with the information it can provide, and which management domains will own, operate, and derive from the DCIM system. In the end, DCIM’s long-term value and cost justification will be decided by the degree of convergence among stakeholders and their agendas and politics. This is particularly true in today’s paradigm shift to divest the organization of the traditional facility-centric enterprise IT architecture, as the move to colocation, cloud, and other hosted services gain momentum. Therefore, managing and optimizing diverse computing resources may represent the greatest opportunity for economic justification of DCIM.
This is the final part of our series on DCIM decisions. Explore other areas of the Data Center Knowledge DCIM InfoCenter to learn about leading vendors and solutions in the space and to keep abreast of the latest news in this space.
