7x24 Fall Magazine 2014 - page 15

15
1
2
At the source of this growth is an insatiable
demand for more capacity and with it, a desire
for greater efficiency in design, construction,
and costs. A number of industry stakeholders
have since taken notice and put forth new
methods and technologies to better integrate
modularity and support today’s system needs.
At the recent 7x24 Metro New York Chapter
Meeting that took place in May 2014, a number
of ideas were shared by some of the top leaders
in the data center industry. Based on the
knowledge shared from that meeting along with
an assessment of industry trends, a few key
assumptions about the status of the modular
data center were made:
ASSUMPTION #1: THERE IS NOW
WIDESPREAD SOCIAL ACCEPTANCE
OF MODULARITY
A few years ago, the idea of building a data
center off-site in a factory was met with
hesitation and skepticism. Questions had yet to
be answered about the difference in cost,
quality, design and engineering versus
traditional field-built construction. Whereas
owners might have had an interest inmodularity,
a lack of understanding and capability led
contractors and engineers to argue against it.
Today, much of the industry has a solid
understanding of the benefits of modularity, but
questions still remain about how to buy it and
who to partner with.
This confusion may stem from the fact that
everyone seems to be selling modularity in one
form or another, but with various purposes in
mind, either to sell their own products and
equipment along with it, or to capitalize on the
opportunity to expand their portfolios into this
new growing market. Therefore, staying
competitive in marketable bids now means
championing this concept and being able to
offer an array of modular solutions to your
customers.
With the rise of interest in modularity, it’s
important to be able to differentiate between
building for the product and building for the
solution. Whether you’re a manufacturer,
designer, contractor, builder, colocation
provider or end-user, if modularity is your goal,
then all product, equipment, and service
considerations should gravitate around the
intended solution and not the other way around.
For example, to build a rocket ship, you wouldn’t
start by evaluating what electrical system you’re
going to use. You’d start by identifying what the
purpose of the rocket ship is and what goals you
wish to accomplish with it in order to guide the
design. Is your goal to repair a damaged satellite
or simply change the flag on the moon? The
same principles should apply when determining
if a modular data center is right for your needs.
Start with the solution and let all other matters
be dictated by it.
ASSUMPTION #2: THE INDUSTRY IS
CONSOLIDATING FROM COMPETITORS
TO PARTNERSHIPS
Building a pre-fabricated modular data center
involves a number of moving parts and
collaboration begets quality. The IT industry is
well accustomed to solving problems via
collaboration as demonstrated by open source
software and bitcoin mining. Therefore, it’s no
surprise that as modularity demand has grown,
the benefits of collaborating with other industry
players has made for a natural progression
either towards partnering up or acquiring
operational
capabilities
to
keep
them
competitive across an actively shifting
landscape.
The results of this trend have bred a number of
successful projects. While each company has
their own way of approaching modularity,
knowing how to effectively engage with other
players to maximize solutions is key. For
example, bitcoin mining, the process of using
software to solve math problems and receive
bitcoins as a reward, is all about server power
and putting capacity on the ground. The need to
drive these supercomputers is off the charts and
demand for data center capacity has surged in
recent years as a result. Partnering with other
suppliers not only streamlines the design and
engineering process, but also ensures
expectations of quality and cost predictability.
And when customers don’t have to worry about
unexpected cost differences from quote to
billing, everyone is happy.
1...,5,6,7,8,9,10,11,12,13,14 16,17,18,19,20,21,22,23,24,25,...92
Powered by FlippingBook