Current data center standards (such as Uptime Institute, ANSI/TIA942 or BICSI) can’t keep up with the latest innovations in data center engineering. Because these classifications are more than two decades old, they don’t recognize modern techniques that could bring much-needed visionary designs embracing energy efficiency and sustainability to the industry.
All of these existing classifications are based on four progressive classes that only cover data center designs powered by redundant diesel generators and UPSs. The classes (further called “the base”) are:
- Basic non redundant: capacity requirements for a dedicated data center site
- Basic redundant: capacity components that increase data center availability
- Concurrent maintainable: increased level of redundancy to enable subsystems to operate while parts of the power and cooling system are offline
- Fault-tolerant: data center site with fully redundant subsystems
Because of these prescribed redundancy measures, data centers that embrace new tech can’t be classified and may automatically fall — possibly even below — the lowest rating. As a result, data center operators are often forced to sacrifice innovation that could lower operating costs and increase efficiency in order to comply with existing standards and gain classification.
For data center operators to avoid being excluded from the search list by potential customers, while still maximizing innovation, a complementary set of standards need to be applied on top of these legacy classifications that help promote Resilience, Sustainability, and Efficiency. In months from now, each facility will be able to be measured and graded on these three criteria by assessing the data center’s five main operational layers: Energy Source, Electrical System, Mechanical System, Network Topology and IT. OSDA, or Open Standard for Datacenter Availability will start with the Electrical and Mechanical infrastructure first before continuing on to the other layers.
To measure for Resilience, data centers can be graded on a standard point scale, with higher scores representing greater resilience. For instance, on a scale of one to 10, a data center can accrue points for its design resilience; being identified against ‘the base” and separately for the type of Energy Source and energy efficiency if it embraces innovative power alternatives such as solar, wind, fuel cells, hydro, tidal and geothermal.
Scores can then be tallied after the assessment of each of these operational layers to give the data center a grade on a scale of — for instance — five to 50, with 50 being the optimal target for Resilience. This score will complement grades for SustainabilityandEfficiency that can be determined by a set industry benchmark.
Sustainability and Efficiency can be classified via a letter-grade system, for instance, or an existing classification similar to LEED standards. The beauty of this proposal is that it’s theoretical and still fluid, so operators across the industry can discuss the best methods for implementation.
These three criteria and the aforementioned grading system work as a starting point to help governments and data center owners to be able to work together in their quest to advance legacy industry standards and the industry overall. This scoring model gives more flexibility for players across the data center landscape to improve their operations without facing noncompliance and being able to show where their design is positioned vs. “the base” and in the sustainable and energy efficiency landscape. The challenge now is to get the conversation in motion, and for standard bearers to explore options that make data center operations meet their requirements while still being ultimately eco-friendly.