I’ve been watching as my colleagues over at the Uptime Institute (Note: the Institute is another business unit of my employer, the 451 Group) worked to put together a reasonable set of guidelines designed to help datacenter owners and operators align management behaviors with the business objectives for the site. As I’m sure you’d agree, improper management processes and procedures can reduce the effectiveness of even the best designed physical site.
Here’s what the Institute has to say about the new standard
Uptime Institute (Institute) today introduces a new standard for data center owners, operators, and managers: Tier Standard: Operational Sustainability. This Standard addresses the facility’s management behaviors and risks that affect long-term data center uptime by aligning the infrastructure’s capabilities, as defined by the Tier Classification System (I-IV), with its ongoing management.
The Institute created Tier Standard: Operational Sustainability in response to data center operators’ needs for an industry standard to minimize operational risk and maximize uptime. The Institute’s Owners Advisory Committee (OAC), which is composed of 29 global data center owners, has reviewed the Standard and will adjudicate any changes.
The Operational Sustainability Standard, which works in tandem with the Institute’s international Tier Classification System, is structured around the three most influential Elements of a data center’s ongoing performance in the following prioritized order: Management & Operations, Building Characteristics, and Site Location.
Each of the three Operational Sustainability Elements has multiple categories and components pertinent to a data center’s Tier level. Detailed tables are provided in Tier Standard: Operational Sustainability. Following are summaries of these Elements:
- Management & Operations: Staffing and organization; maintenance processes and procedures; training; and planning, coordination, and management.
- Building Characteristics: Building features that include the condition, age, and type of structure; design principles; and the operational conditions of the equipment.
- Site Location: Potential local or regional natural and/or man-made disasters and corresponding mitigation plans. This includes everything from the likelihood of a tornado or flood to a disruptive accident at a nearby railroad system.
The weighting of the potential impact of the Elements is based on analysis of the Institute’s Abnormal Incident Reports (AIRs) database. AIRs are ‘lessons learned’ from thousands of reported data center incidents at live, high-performance facilities around the world. AIRs analysis demonstrates that 70% of the reported outages are directly attributable to human error, which is addressed by the Management & Operations Element of the Operational Sustainability Standard.
Snapshot analysis
I know that this appears to be pretty boring reading. Facilities management and site operational planning documents, while critical, often are pretty dry reading material. It is also foreign territory for most IT professionals. This sort of planning, however, is an absolute necessity if the site is going to be successful in meeting the business objectives set by the organization
You can just imagine how difficult it must have been to sift through the best practices of major datacenter operators, spelunking into the error reporting database to determine what processes or procedures led to an outage and then codify that into a set of guidelines, training sessions and certification testing. I’d glad they did that. I don’t think I could put something together of this depth and complexity.
If you’re part of your organization’s datacenter management team, you are likely to find “Tier Standard: Operational Sustainability” very useful. The document is freely available from the Institute’s Publication page.
