7x24 2015 Fall Magazine

15 / 88

15
basic thermal management, and networking
capabilities. Simple, right? But taken collectively,
these are complex networks with considerable
management challenges. It might not be what the
earliest DCIM engineers had in mind, but the
distributed data center is a challenge today’s DCIM
solutions are built to address.
Case in point: Cambridge University
Cambridge University in England is one of the
world’s leading research institutions and has been
for more than 800 years. The institution’s
dedication to learning and innovation extends
beyond academics and into its operations,
including IT. The university values unified
technology support that enables and drives
achievement, but maintaining an integrated IT
infrastructure has been a challenge. As the campus
and various departments grew and evolved,
separate micro networks popped up. Each
department had the option to integrate its servers
with the campus’s main data center, but it wasn’t
required. As a result, the campus ended up with
more than 200 server rooms serving 120
departments, and most of them operated
independently.
None of this is all that unusual, but it’s an inefficient
way to operate—and the IT personnel at Cambridge
knew it. All of these separate, independent IT
nodes eliminate the benefits of consolidated
management, standardized service delivery and
improved security and availability. On top of that,
each department was free to build its computing
capabilities to its own specifications—and using
whichever equipment manufacturers it liked. That’s
200 loosely connected server rooms with little or
no consistency in terms of server or infrastructure
vendors—again, not a model of efficiency.
The university’s IT staff understood the need to
improve visibility and management of this
computing quilt, believing it was the key to
optimizing computing performance and efficiency
and reducing skyrocketing operational costs. Their
solution? Consolidating some of the far-flung IT
rooms in a single modern data center and uniting
the new facility and remaining distributed
computing sites under a single DCIM system.
DCIM in a Distributed Environment
While the distributed model is relatively new,
applying DCIM to improve visibility and control of a
web of hard-to-see IT assets is more or less exactly
what the technology was developed to do. The
physical location and heterogeneous nature of the
Cambridge facilities just added another layer of
complexity. Ultimately, the desire was the same as
it is for any DCIM customer: to see and control
multiple assets—not just servers, but every
component across the network—from a single
location.
Cambridge settled on Emerson Network Power’s
Trellis™ platform, a DCIM solution that enabled
management of multi-vendor IT, power and cooling
resources through a single pane of glass. The
consolidation, data migration and DCIM
implementation is an ongoing process, but the
early returns are overwhelmingly positive.
The system works with output from the various
pieces of equipment including critical
infrastructure, configuring and organizing data
from multiple sources and translating it into a set of
unified actions. It allows the university to reduce
both capital and operational expenses by
virtualizing platforms and refreshing equipment,
using the central data center space intelligently,
and implementing evaporative cooling systems. It
has made everything the university does around its
IT systems smarter, and we would expect nothing
less from a place like Cambridge.
Enzo Greco is Vice President and General Manager, Software, Data Center Solutions at Emerson Network Power.
He can be reached at