OUTAGES/UPS


FOCUS on POWER


continuity approaches and policy.Where power equipment and flowis in question, a first responsewill be to obtain all supplier/ support contractor or partner support,where key electrical hardware such as UPS, static switches, site-specific standby or generator configurations are concerned.


But hardware is only part of the storywhen it comes to copingwith the dangers of power outage disruption to a data center.Human error iswithout question an issue. Poor operational procedures, caused essentially by inadequate training, can often be a contributing factor to a mini-drama becomingmore of a crisis.“Many systems havewraparound andmaintenance by-pass, but local site staffmay not knowhow to operate these in emergency conditions, and instanceswhere load has been dropped could have been avoided if robust and understood operational procedures had been implemented,”points outHewitt.


O’Keeffe agrees that procedure is vital here, as the understandable rush to get things back up and runningmay impact the longer term lesson-learning opportunity.“It is important at the initial stages of the fault that not only is the problemcontained but that the gathering


of data to enable the investigation into the event is undertaken in a controlledmanner.On many occasions, specialist repairs have been undertaken prior to the establishment of the root cause of the fault, orwhen vital status data of the equipment hasn’t been captured at the time of the incident and immediately post the event.This has resulted in repeated faults and continuing disruption to operations,”hewarns.


But one should be able to rely on the expertise of external support engineers as part of the process ofmoving on fromthe immediate power outage crisis. Specialist support engineers are trained not only inmaintenance and repair of equipment, but also fault-finding and investigation, and their reportswill provide vital information into the fault investigation.


At the same time, sensor-based or other forms of remotemonitoring should also play a role here and can provide a backup to any site- based control or recording system.


Comprehensive remotemonitoring systemdata can significantly reduce the time involved in the decision-making process by providing accurate and rapid information on your power outage event.The supplier should be able to provide


The UPS is changing


Once seen as the byword for stability in IT, drivers for more efficient architecture and green issues are revolutionising the standard UPS.


Key to all power contingency plans in the data center – the UPS complements such auxiliary or emergency power systems, or on-site back-up generators, as it will provide instantaneous or near-instantaneous protection from input power interruptions.


As iswell known, it will either be powered by one or more attached (typically lead acid) batteries, or by means of diesel generators or rotary (flywheel) quick-spurt power generation, until main or standby power kicks in. A UPS, in any case, has to convert incoming AC power into DC, then back into AC for the actual equipment’s use.


UPS is a very established category of data center equipment, but that doesn’t mean it doesn’t require careful evaluation to make sure the best systems for your specific use get used. Vendors argue that there are palpable differences in performance depending on which of the two major topologies in use today parallel online (also known as line interactive) and double- conversion, are chosen.


This is especially important when it comes to evaluating efficiency claims. Some UPS vendors are far too optimistic, shallwe say, on this point. Benchmarks may be impossible to replicate in a properly designed mission- critical infrastructure at the end-user site. As Active Power spokesman Lee Higgins notes, “There is a need to be cautious and critical of the data provided since it can change dramatically as options (such as filters and


for a full forensic analysis data on equipment and supply status, fault type and timing information on any supply interruptions; pre-fault loadings; battery status; temperature conditions in the equipment locality and switching actions (taken pre- and post-event).


Clearly, key suppliers should be part of the incidentmanagement process, and the establishment of clear contractual support arrangements required to achieve all these actions is not something that should be assumed or left until an incident takes place.


So, in combinationwith a detailedmaintenance and service plan for all parts of the power- using tree in your facility, prepare a response plan for any incidents that do fault containment first, followed by as full and accurate a data- gathering exercise as possible to enable a full investigation process to be launched, and deliver proper fault identification and repair lessons, proper fault analysis, and be the basis for a‘lessons learned’post power outage plan.


You knowit’ll happen sooner or later, sowhy not be preparedwhen it does?


“The problem with both fuel cells and flywheels, and indeed all non-voltaic power storage solutions, is that they are still very expensive and I’d say unproven in comparison to lead acid.”


www.datacenterdynamics.com


7


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25