Posted by Bob Landstrom
Enterprises have embraced the data center tier classification system developed by the Uptime Institute for evaluating their own data center facilities and those of potential colocation and hosting providers.
The subject of facility availability has matured over recent years. The mind set has matured from recognition that existing IT facilities were a problem that needed to be improved to questions of how to improve. How much should we improve them? What do we improve? How far can we grow? Traditionally, these questions were answered without much guidance beyond the level of capital funds available for improvements.
The tier classification system developed by the Uptime Institute is an academic framework that can be used as guidance for determining the type of data center facility appropriate for the Business. It’s a seminal body of work, and has become part of the daily lexicon of those working in the data center world. We’ve written about it several times in this forum, covering what it is, where it came from, and what it’s not. Indeed we’ve dedicated time in this blog talking about how the tier classifications are (very) frequently misused.
Even professionals in the data center industry abuse the Uptime Institute tier classifications. Most often, this comes in the form of claiming fractional tiers. One will hear mention of “Tier 2+” or “Tier 2 and a half” or “near Tier 3” descriptions of data center facilities. While, in a formal sense, this is not possible because facilities can only be classified as whole number tiers, this terminology is being used casually by even very seasoned data center pros. Colocation and hosting providers often claim Tier 3 or Tier 4 status, even without certification.
Whether it’s right or wrong to talk about fractional tiers and tiers without certification, many are clearly comfortable doing so. You can’t really blame them. The Uptime Institute’s work on this topic makes good sense to people (at least in the ways they’ve interpreted it for the time being), and as a conceptual model it’s sufficiently similar in all our heads to use as a basis for conversation.
What does this mean though, for the integrity of the tier classifications in general? Growing consensus seems to point to a desire to see tier classification definitions expanded to consider those cases which force us to misuse the terminology. This is especially true between Tier 2 and Tier 3. There’s a big step between Tier 2 and the concurrently maintainable Tier 3. In some cases, a site will have part of the MEP infrastructure complying with the spirit of concurrent maintainability and other components that do not. If a change is not made in this area, then we can expect continued misuse of the tier classifications by professionals in the industry.
What about referring to data center sites as a certain tier level when they’re not certified? Well let’s talk about tier certification for a moment. To have your facility certified as meeting one of the Uptime Institute tier levels, there is only one company that can be hired to do that. The Uptime Institute, which retains all rights to certification, uses ComputerSite Engineering to audit, assess, and ultimately grant an official Uptime Institute tier certification. It’s a very involved audit, and very expensive. Oh and by the way, you should know that there is only one colocation facility that has been granted Uptime Institute certification. It’s a sub-10,000 square foot Tier3 facility operated by OnePartner, located in western Virginia. Besides that, no colocation or hosting provider has achieved Tier 3 certification. So if you’re comparing colocation providers and they’re telling you that the site is of a certain tier classification, understand that’s how they believe (or hope) they’d score if an audit was actually done. This point alone is pretty clear evidence that the value of certification is not aligned with the cost of achieving certification.
Let’s pause on that point and elaborate a bit further. We often hear claims of “Tier 2+” or claims of “Tier 3” without certification to back up the claim. Clearly suppliers and consumers are dealing with this situation rather than demanding or seeking tier certification. The cost and complexity of certification are certainly a barrier. There’s more to it than that though. The community is increasingly aware that a tier level designation is a predictive classification, and that many facilities that would classify as a lower tier are demonstrating historical availability performance greater than the highest tier classification. In other words, if you can show me you have enjoyed exceptional uptime then I’m not as concerned how we agree on your facility tier level.
For example, there are colocation providers that operate facilities with demonstrated availability performance exceeding Tier 4 guidelines, but choose not to pursue certification of its sites. How does one achieve availability excellence without high tier certification? Regardless of what tier classification a site would be certified, the true availability performance of that site is dependent upon the excellence of people , processes, and procedures by which the site is managed. The strength of the Standard Operating Procedures (SOPs) and Methods of Procedures (MOPs) are fundamental to delivering consistent availability performance.
The data center tier classifications from the Uptime Institute have brought monumental contributions to the data center industry. As with all new developments, we’re now well along on the maturity curve and signals are emerging demanding new guidelines.