Dealing With A Variety of Data Center Rack Sizes and Models9 min read
Containment is not as ubiquitous as the EIA rack. While there are a variety of reasons for some remaining industry intransigence, cost and ineffectiveness are not defensible. Airflow containment will always provide a data center environment that can support higher power densities and lower energy costs and will still pay for itself quickly. Sometimes these reasons may just be excuses camouflaging a general antipathy toward change, effort or knowledge acquisition. More often, however, I suspect the citations occur when confronted with cold, hard reality which may not be as tidy as textbook installations illustrated in conference presentations, data center certification courses, magazine articles, or blogger pontifications. Reality often includes apparent barriers to optimum airflow management such as physical, mechanical or hardware complexities or obstacles. The fact is, unless we are talking about a new design that has had active involvement from IT, facilities, architectural engineering and strategic planning from the very beginning, there are going to be complexities and obstacles not accounted for in the idealized vision of containment.
In recent blogs, I have addressed obstacles to effective airflow management presented by fire suppression systems and associated codes as well as by installed mechanical infrastructure not associated with reaping the economic and performance benefits of containment – i.e., single speed fans and DX coolers. I have explored the confrontation with overhead physical obstacles such as ductwork, ladder rack, power busway, basket tray, or any variety of hanging Unistrut-type configurations. I have delved into data centers without dedicated supply and return air paths and reviewed problems associated with equipment that does not conveniently breathe front-to-rear. In all these cases, I showed how data center airflow containment is deployed efficiently and provided some guidance on how to estimate financial paybacks from set point management and airflow volume control. I have been reviewing (and de-bunking) a list of obstacles to airflow containment for over two months now; seems like there shouldn’t be anything left to discuss. Wrong.
Way back at the beginning of the introduction of data center airflow containment, the most frequently mentioned obstacle to retrofitting existing spaces was the ubiquity of different data center rack sizes or shapes and to outfitting new designs. The often cited barrier was the omnipresent danger of someone needing to bring in a cabinet in the future that did not precisely conform to the external physical specifications of the containment-compatible installed base. Since these obstacles were, in fact, sales objections to a nascent technology, they were the first addressed by solution providers and third-party bandwagon hoppers and the variety and efficacy of ways to overcome differences in cabinet heights, shapes or vendor idiosyncrasies are mature and widely available.
Data centers achieved early retrofits to containment by adding aisle containment plastic curtains. These could be as sophisticated as engineered-to-order systems with overhead support infrastructure and fire-suppression activation release mechanisms or as simple as home-grown DIY plastic strips bridging gaps from overhead supply ducts to cabinet tops. In these cases, lengths of plastic strips or curtains are merely trimmed to fit. It cannot get much simpler than that. Some of the shine on this means of overcoming the inconsistent cabinet issues dissipated with videos of curtains flapping in the breeze and NFPA crackdowns on heat-sensitive mechanical release topologies; nevertheless, early adopters got some degree of before-and-after efficiency and effectiveness improvement. Since then, containment vendors have strengthened their offerings with all variety of approaches to accommodating cabinet differences with sliding adjustable partitions, trimmable partitions and various flexible plugs for filling all the holes.
The prospect of having an efficient containment data center undermined by the future introduction of a rogue non-compatible cabinet can be daunting. However, the designs of retrofit cabinet chimneys which allow for the segregation of rogue cabinets addresses this issue. Such separation is no longer required as most all containment vendors offer partition options and hole- plugging accessories that provide natural paths for inclusive integration. Add to the containment barrier pieces the airflow re-directing cabinets for side-breathing switches reviewed in my last blog, and rogue cabinets no longer represent insurmountable obstacles to airflow containment.
Finally, whether facing a retrofit dilemma or a new design with real or imagined present or future impediments, partial containment can be an efficient way to realizing most of the benefits of containment with cabinets from multiple vendors. No matter the solution, such as the end of row doors or cabinet top-mounted partial partitions, solutions to these problems had existed for some time. This post intended to serve as a quick reminder for any friends who may have been distracted at the moment.
Data Center Consultant
Let's keep in touch!
Airflow Management Awareness Month
Free Informative webinars every Tuesday in June.