Total Cost of Ownership: An Executive Overview
June 19th, 2012 By: Julius Neudorfer
Total Cost of Ownership “TCO” is basic and easy to calculate; just ask any first year economics or business major. Nonetheless, the true TCO of the data center is somewhat more elusive to project accurately. There are many subtleties which can be overlooked or are simply unaccounted for (or perhaps underestimated), over the operational life of a data center. This Data Center Knowledge Executive Guide Series of articles will examine the many aspects of TCO, such as energy costs and operating efficiency, as well as other data center specific issues.
Like any other TCO analysis, there are the more obvi¬ous directly identifiable line items, such as initial capital investment and depreciation; however, the TCO of data center has many more elements, some of which are far more complex. Let’s start by defining and understanding the basic factors of TCO calculations.
Whether you decide to build, lease or buy a data center, before you can compare and calculate TCO, it is important to understand and clarify the specific terms of the data center industry:
- Total Space (Building Shell)
- Whitespace (Raised Floor)
- Effective Usable Space (Rack Space)
- Adjacent Total Lot Size (raw lot size)
- Critical Load Power Capacity
- Power Density
Unlike other commercial real estate purchases or leases, the raw square footage of the data center Building Shell is a relatively minor factor in the overall TCO calculation. The more frequently used basis is “whitespace” which is a common industry term for the raised floor area where the computing equipment resides. However, be sure to not limit your TCO space evaluations just on “whitespace” alone. It is the net usable footprint that is available for the computer equipment, the majority of which is housed in racks or cabinets. The effective usable rack space is less than the whitespace and significantly less than the gross building shell space.
The other key factor is the Critical Load Power Capacity. Effectively, and most importantly, this is the net power available for your computing payload, primarily based on the available power capacity of the UPS systems.
It should not be confused with gross utility power to the site, which while it is important to the overall power required to operate the site’s other systems such as cool¬ing equipment, is the Critical Power that represents the conditioned and protected power that is available for your computers.
Beside the Critical Power capacity, the power density will impact how much computing equipment can be placed in each rack. A data center with a lower power density would mean that you may need to use more racks (and whitespace) to house the same amount of comput¬ing equipment than at a higher density site. Power den¬sity is typically expressed in two ways; watts per square foot or Kilowatts (kW) per rack, or sometimes both. This is primarily based on the design and type of the data center cooling system. Many older data centers cannot effectively or efficiently cool more than 5 kW per rack (some even less), and in some cases their efficiency goes down beyond 3 kW per rack. Even today, not all newer data centers can accommodate medium (5-10kW per rack) or “high-density” racks which require 10 kW or more per rack.
In addition, the overall land area required around the building is a secondary, but necessary consideration. This is used for infrastructure related support systems, such as utility transformers, generators and fuel storage, cool¬ing equipment, as well as security systems such as entry and exit point gates and security kiosks. There is also ex¬terior space required for staff and visitor parking, as well as separate receiving zones for equipment deliveries and cargo security inspection of materials entering and being removed from the data center.
The TCO initial capital costs are directly related to the levels of redundancy (Tier 2, 3 or 4) of the power and cooling infrastructure equipment. The higher the level of redundancy the more equipment is needed, in order to avoid infrastructure equipment failure from impacting the op¬eration of the computing equipment (reader note: please refer to the Uptime article in the first Executive Guide Series for a detailed explanation of redundancy levels). The maintenance of the additional equipment obviously also impacts the ongoing costs as well.
Over the next few weeks I will publish the following articles in the DCK Executive Guide Series on TCO:
To download entire DCK Executive Guide on Data Center Knowledge TCO, courtesy of Digital Realty, click here.