73
centers is that a failure in the power
infrastructure will likely result in a
catastrophic, total unplanned
outage. This means that in addition
to any direct costs incurred to
remedy the cause of the outage,
indirect and opportunity costs also
will be significant due to the fact
that all stakeholders will be affected
by the outage.
eFFICIenCy
Data center managers today are
feeling increasing pressure to
operate efficiently—in terms of
both energy and management
resources.
This is reflected in results from the
Data Center Users Group (DCUG)
study sponsored by Emerson
Network Power, where for the last
few years, availability and efficiency
have been top concerns of
respondents. For the spring 2014
DCUG survey, when asked to
identify their top facility/network
concerns, 51 percent cited adequate
monitoring while 39 percent named
energy efficiency and availability as
top fears.
Power systems in use today in many
data centers still remain below the
efficiencies currently available.
While server power supply
efficiencies have improved
significantly since the days of 79
percent power supply efficiencies,
they continue to consume more
energy than is necessary. Most
estimates put the average power
supply efficiency at 86.6 percent,
well below the 93 percent that is
available.
Emerson Network Power
conducted a systematic analysis of
data center energy use and the
various approaches to reducing it to
determine which were most
effective. Using this analysis,
Emerson developed Energy Logic,
a vendor-neutral roadmap of 10
strategies that can reduce a data
center’s energy consumption by up
to 74 percent. One of the findings
of the analysis was that increasing
server power supply efficiency from
86.6 percent to 93 percent reduces
total data center power
consumption by 7.1 percent.
As with other data center systems,
server power supply efficiency
varies depending on load (Figure 5).
Figure 5 shows power supply
efficiencies at different loads for
two power supply models. At 20
percent load, model A has an
efficiency of approximately 88
percent while model B has
efficiency closer to 82 percent.
Enabling them to provide higher
efficiency, some new UPS modules
now have flatter efficiency curves
than traditional units even while
running at partial loading.
SCAlABIlITy
Increasing flexibility in how IT
systems and data centers are
operated and managed is now the
best way to achieve availability
while maximizing opportunities to
improve energy and operational
efficiencies. It also allows data
center managers to meet current
requirements, while ensuring the
ability to quickly and cost-
effectively meet growing and
fluctuating performance and
capacity needs.
Mathematically, Availability “A,” is expressed:
A = (MTBF / (MTBF+ MTTR)) x 100%, where
MTBF = Mean Time Between UPS Output Failures
MTTR = Mean Time To Repair (the total time from the power
failure to power restoration)
To improve availability, increase MTBF and/or reduce MTTR.
Figure 5. Power supply efficiency at various loads.
