Creating energy-efficient data centers
for sustainable business outcomes
White paper
Table of contents
Table of contents.................................................................................................1
Executive summary .............................................................................................2
Energy efficiency in the data center: why it matters to your company .........................2
Technologies that address energy efficiency ...........................................................3
Best practices for improving energy efficiency in your data center .............................4
Steps to increasing data center energy efficiency ....................................................6
How HP can help you meet your energy and environmental challenges .....................6
The road ahead ..................................................................................................7
Creating an energy-efficient data center ................................................................8
To l earn more.....................................................................................................8
pg_0002
Reduce energy costs and
environmental impact—while
increasing computing performance
Executive summary
In data centers around the world, power and cooling
costs are rising rapidly and consuming an ever-larger
portion of IT budgets. This trend is taking place in a time
in which companies are feeling increasing pressure to
operate in a more environmentally sustainable manner.
This paper explores technologies and best practice that
can help your organization respond to today’s power
and cooling challenges in a manner that reduces both
energy expenditures and environmental impacts—while
increasing computing performance.
Energy efficiency in the data center:
why it matters to your company
To further accelerate costs, servers and storage devices
are consuming ever-larger amounts of electricity,
and demanding ever-larger amounts of power and
cooling capacity.
First, take the case of servers. The U.S. Environmental
Protection Agency estimates that the amount of power
consumed by servers will double in the next five years.
Worldwide, servers now account for as much as one
percent of power consumption.
1
Two key dynamics
are driving the steep server power-consumption
curve: increasing server densities and rapid growth
in applications.
Server power density is growing by an annual rate
of about four percent.
2
Across a wide range of server
types, the heat load per product footprint is on the
rise. But this alone isn’t causing the doubling of power
consumption every five years. What’s really driving
the problem is increased growth in applications.
There is a huge and increasing demand for software
applications. Application growth is now outpacing
gains in server efficiency and performance.
So what about bigger, faster processors? While new
servers use ever faster processors, the power crunch
doesn’t stem from increasing server performance. That’s
because server efficiency is improving. Since the late
1990s, server performance per watt of power used has
been doubling every two years.
3
So in simple terms,
we’re all getting more computational performance for
the money we’re spending to power computers.
Storage, too, is proving to be a major consumer of
electricity. While servers account for an estimated 48
percent of the power consumed by data center
hardware devices (excluding air conditioning systems),
storage systems account for 37 to 40 percent.
4
What’s
driving up the power bills for storage? Companies are
saving more data, and storage is doubling about every
18 months. To compound the problem, companies have
to retain information for longer periods of time. This all
leads to more demand for storage capacity, and more
demand for electricity to power the storage hardware.
The rising power consumption is driving up data center
costs. At today’s energy prices, for example, by 2008 the
energy costs for running a server will exceed the costs of
buying the server.
5
And when you couple server energy
costs with associated infrastructure costs—including
the costs of data center generators, power supplies and
air conditioning systems—the overall cost of power
and cooling a server now doubles the cost of buying
the server.
This is a dramatic change for companies. As recently
as the 1990s, many companies considered power and
cooling costs as a negligible expense relative to the cost
of buying IT equipment. Today, with power and cooling
costs running out of control, this view has changed. And
as energy costs continue to rise, the numbers will only
get worse for status quo environments. Clearly, strategic
changes are in order.
The business case for improving energy costs and
reducing consumption
Anymore, it’s not hard to build a business case for
improving energy costs. Cutting energy consumption
and costs is now a strategic focus of many companies.
In the drive to lower energy costs, some companies are
locating data centers near power-generation sources
that offer plentiful supplies of lower-cost energy. That’s
the case with some Internet companies, for example,
that are building massive data centers near the Grand
Coulee Dam in the northwestern United States.
While searching for lower-cost energy, companies are
driving toward better efficiency. Gains in efficiency
improve not only energy costs but also infrastructure
costs. This lower cost structure makes your company
more competitive.
This focus isn’t just for budgetary reasons, however.
Companies are feeling increasing public and regulatory
pressure to operate in a more environmentally sustainable
manner. Cutting energy consumption is one of the keys to
reducing greenhouse emissions and a company’s carbon
footprint—and to creating a greener, more sustainable
business. This sort of environmental responsibility is
emerging as an important differentiator for companies
across a wide range of industries.
2
1
“Report to Congress on Server and Data Center Energy Efficiency Public Law
109- 431,” U.S. Environmental Protection Agency, 23 April 2007
2
Datacom Equipment Power Trends and Cooling Applications (ASHRAE)
3
Belady, C., “In the Data Center, Power and Cooling Costs More than IT
Equipment it Supports,” Electronics Cooling Magazine (February 2007)
4
“Green storage means money saved on power,” Network World,
May 21, 2007
5
Belady, C., “In the Data Center, Power and Cooling Costs More than IT
Equipment it Supports,” Electronics Cooling Magazine (February 2007)
pg_0003
Technologies that address energy
efficiency
To address today’s challenges, data center and
facilities managers have begun to think in terms of
overall power and cooling resource management. The
goal is to eliminate the over-provisioning of power and
cooling resources and to increase utilization of all the
equipment in a data center. In these efforts, forward-
looking companies are leveraging a wide range of
power and cooling technologies that can help cut
power consumption.
Server-level technologies
One way to cut power consumption is to throttle down
processors that don’t need to be running at full force.
Throttling allows an underutilized processor to use less
power. Typically a processor will have a maximum
frequency. When this processor is being underutilized,
you can use throttling technologies to put it into a
lower state, which means it runs at a lower frequency.
When it runs at a lower frequency, the processor uses
less power. In some cases, just by enabling the use of
simple component throttling, you can cut processor
power consumption by as much as 10 percent and
cooling requirements by up to 10 percent.
6
Virtualization is another server-level technology that
can help cut your energy bills. Virtualization allows
you to combine the workloads of several underutilized
servers onto a single server that is utilized more fully.
This approach allows you to reduce your power
footprint because you need fewer servers to do the
same amount of work. This can help you achieve
power savings of up to 50 percent.
Storage-level technologies
There are many emerging technologies that can help
cut the energy bills associated with data storage.
These include:
• Thin provisioning and dynamic capacity management—
These complementary technologies enable storage
systems to make more efficient use of capacity. With
these technologies, storage systems require as much
as 50 percent less storage than systems without the
same technologies. Fewer disk drives means less
power required.
• Logical snapshots and writeable snapshots—
Organizations often have many copies of data
that they use for many different purposes. Logical
snapshots are space efficient and save significant
disk space compared to full copies. A writeable
snapshot shares the source data with the original
primary data and only uses additional capacity
to track changes that are made to the writeable
snapshot. This technology can lead to significant
capacity savings, and associated energy savings.
• Data compression—This widely used technology
allows you to store more data within the same
amount of space. Depending on the type of data
and the compression schemes, you might be able
to compress some files to half their original size.
• Data de-duplication—The process of examining and
eliminating duplicate data can help you optimize
storage capacity. Data de-duplication can reduce
capacity requirements by more than 20 times. It is
mostly found in disk-to-disk backup and disk-based
archiving solutions today but is expected to soon
move up to online storage systems.
• Tiered storage—60 to 80 percent of all data is
dormant 90 days after it is created. This gives you
the opportunity to save disk storage capacity by
moving this less-used data to tape storage.
7
3
Figure 1. Costly trends
When you couple server energy
costs with associated infrastructure
costs, the overall cost of power
and cooling a server now doubles
the cost of buying the server.
Infrastructure costs and energy costs have exceeded server costs
Annualized data center costs
I&E costs are no longer negligible
(as they were in the 90s)
I&E costs are 2X that of a 1U server today
(more tomorrow)
Energy costs are rising and will make things worse
Typical 50,000 sq ft and 5 megawatt data center
can cost $61–$133 million
4000
3000
2000
10 0 0
0
1990 1995 2000 2005 2010 2015 2020
2001
2008
2004
Server cost
Infrastructure cost
Energy cost
Annual I&E
6
HP Insight Control, Power Regulator 3rd Edition Tech Brief February 2007,
ISS Performance Testing Engineering assuming 320 watt server
7
“Power, Cooling, Space Efficient Storage,” by Tony Asaro, Senior Analyst,
Enterprise Strategy Group (July 2007)
Source: Belady, C., “In the Data Center, Power and Cooling Costs More than IT Equipment it Supports”, Electronics
Cooling Magazine (Feb 2007). Calculated with IT and cooling power and electricity cost of $0.1/kW-hr
pg_0004
Rack-level technologies
Another area of focus is integrated power management
at the rack or chassis level. That’s the case with blade
technology. Technology is now available to pool the
power and management of an entire blade chassis, and
all the server blades within it. This pooled approach
allows better management of power and cooling, for
energy savings of up to 25 percent.
8
Data center-level technologies
At a broader level, companies are focusing on overall
data center power and cooling management. The focus
here is on managing power and cooling resources based
on specific needs, as opposed to simply over-provisioning
resources so you’re sure all your needs are covered.
For example, without the right management tools in
place, you might find yourself putting in a lot of extra
cooling capacity to address the problem of some hot,
high-power racks you’ve added at one end of your
data center. If you aren’t set up to provision cooling
selectively, you end up over-provisioning cooling to
the entire data center.
But with the right technologies in place, you could
instead target extra cooling capacity to the spots where
you really need it. With technologies that enable
selective cooling, you might achieve data center power
and cooling savings of 15 to 45 percent.
Closely coupled cooling
Another important data center trend is closely coupled
cooling. Closely coupled cooling makes your cooling
solutions more sensitive to individual pieces of IT
equipment and their cooling requirements. With this
approach, the cool air doesn’t have to travel a long
distance and mix with warmer air along the way. That
mixing equates to lost efficiency and higher energy bills.
One approach to closely coupled cooling moves
computer room air conditioners, or CRACs, off the
walls of the data center and positions them closer to
heat loads. This is sometimes referred to as “CRAC
by the rack.” With this approach, air is more localized.
There is less opportunity for the mixing of cool and
hot air. Less mixing equals more efficiency.
We’re also seeing the emergence of liquid-cooled racks.
This approach improves efficiency and power density
even further by bringing the cooling right into the rack.
You essentially create an enclosed box, or rack, with
built-in cooling capabilities. Servers pull in cool air
from one side of the rack and send hot air out of the
other side of the rack. The air then passes through a
heat exchanger, gets cooled and flows back to the
other side of the rack. With liquid-cooled racks, there
is very little opportunity for hot and cool air to mix, so
you have greater efficiency.
Looking ahead a bit, we may eventually see the advent
of direct liquid cooling, which brings cooling down to the
chip level. There is currently a lot of industry discussion
around this approach.
So the trend is clear: Data center cooling capabilities
are marching closer and closer to heat sources. And
in general, the closer you get, the greater the efficiency.
Many new technologies are emerging to enable
this trend.
Best practices for improving energy
efficiency in your data center
There are many best practices that can help improve
the efficiency of your data center. Some of the more
important best practices follow. This list is based on
our experiences in data centers around the world,
our research efforts and our close involvement with
organizations such as the Green Grid, an industry
consortium focused on energy efficiency in the
data center.
You don’t necessarily have to have budget in place to
pay for power and cooling best practices in your data
center. Many of these approaches are virtually free.
Use a hot aisle/cold aisle design.
One important best practice in data center design is
to alternate hot and cold aisles. With this more energy-
efficient layout, server racks are placed front-to-front in
the cold aisle and back-to-back in the hot aisle. In the
cold aisle, cool air comes up out of floor vents to cool
the server racks. In the hot aisle, the hot exhaust comes
out of the back of the servers and flows up into the
data center room. This layout creates greater segre
-
gation of hot air and cool air, which equates to gains
in cooling efficiency.
Match server airflows.
It’s important to match server airflows with the airflows
of your air conditioners or air handlers. If you can match
these airflows, you can better prevent hot air from the
hot aisle coming over the top of a server rack and
then rolling into the uppermost servers in the rack. This
situation can cause servers to overheat. When you match
your airflows, you do a better job of keeping the cool
air where it ought to be, and the hot air where it
ought to be.
Use long rows without gaps.
Server cabinets should be positioned in long rows
with no gaps in the rows. This design helps you better
separate hot air from cold air. Without gaps in your
rows, more of the cold air stays in the cold aisle and
more of the hot air stays in the hot aisle.
4
8
IDC white paper sponsored by HP, Forecasting Total Cost of Ownership for
Initial Deployments of Server Blades, Doc #202092, June 2006
pg_0005
Use cabinet blanking panels.
When you have an empty slot in a server rack, block
the hole with a cabinet blanking panel. Without a
blanking panel, hot air from the back side of the server
can flow through the opening and get ingested by the
server. You then have to provide additional cooling to
counteract the hot air that is coming through the rack.
A simple blanking panel will solve this problem.
Orient AC units perpendicular to hot aisles.
Air conditioning units should be positioned so they are
perpendicular to hot aisles. By putting the air conditioners
in hot aisles, you essentially use the hot aisle as a duct
that causes hot air to flow toward the air conditioner.
Seal cable cutouts.
Any time cold air leaks out of the back of a server into
a hot aisle, you’re mixing hot and cold air—and losing
efficiency. When you seal cable cutouts, you block
another pathway for the mixing of air.
Use 0.8 meter to 1.0 meter high floors.
Using higher raised floors can increase efficiency. It’s
much harder to distribute air evenly in a data center if
you don’t have enough air space under your floors. A
low floor tends to cause higher air velocities. You can
mitigate this problem by raising your floors to create
more air space and more uniform air pressure beneath
your floors.
Maintain high- and low-density areas.
When you maintain high-density and low-density areas
within your data center, you can focus your high-density
cooling technologies where they are most needed.
Thermal zone mapping capabilities help you identify the
areas of the data center that are influenced by each air
conditioning unit and to what extent the areas overlap.
This gives you the ability to turn down the cooling in one
area if you are over-provisioning, change the zones of
influence by adjusting the flow settings or direction,
and position high-density or mission-critical equipment
in zones that overlap to better protect the equipment
should a CRAC unit fail.
Consider economizers.
You can use power-economize modes on servers and
air conditioners to make equipment operate more
efficiently. In many cases, economizers are already in
place; they simply aren’t enabled.
Use metrics.
To truly understand the results you are getting, you need
to use metrics. The right metrics should give you a clear
indication of the progress you are making and the
energy you are saving.
You can measure your progress in terms of your Power
Usage Effectiveness (PUE) ratio.
9
This measure allows
you to track your improvements in terms of quantifiable
numbers. The PUE is a simple ratio that divides the
total amount of power coming into your data center by
your IT equipment power, or your critical load.
The PUE ratio enables your efforts to measure the right
thing, compare your results with benchmarks for your
industry and validate your suppliers’ claims. When
you adopt new, more efficient power and cooling
technologies, your PUE ratio should fall.
Multiple industry groups have adopted or are laying
the groundwork to adopt the PUE. These include the
American Society of Heating, Refrigerating and Air-
Conditioning Engineers (ASHRAE) and the Green
G rid industry consortium.
5
Figure 2. Closely coupled cooling
Closely coupled cooling
approaches move the cooling
solution closer to the heat source.
This increases energy efficiency.
Before Lots of mixing of hot and cold air stream
Better (CRAC by the rack) Slight mixing
Best (liquid cooled rack) No mixin g
T
CRAC
CRAC
T
CRAC
Ventilation tiles
Floor plenum
Equipment racks
Recirculation Short-circuit
AC
units
Room chilled water supply
Cool fluid
Hot fluid
Servers
Source: Christian Belady, Presented at the Uptime Conference on April 24, 2006
9
Malone, C., Belady, C., “Metrics to Characterize Data Center and IT
Equipment Energy Use,” Proceedings of 2006 Digital Power Forum,
Richardson, Texas (September 2006)
pg_0006
Business-driven benefits
Collectively, the best practices outlined here can yield very
tangible, business-driven benefits. They can help you:
• Reduce and control server temperatures to increase
the reliability of your systems
• Increase system and application uptime
• Increase server densities and open up space for
additional equipment, which can extend the life of
your data center
• Reduce energy usage and your total cost of
ownership (TCO)
• Reduce your carbon footprint and create a more
sustainable computing environment
Steps to increasing data center
energy efficiency
Creating a more energy-efficient data center is a stepwise
process. You can begin with some fundamental changes,
and work your way up from there. It may help to view
this progression in terms of four key steps.
1 0
Step 1. Use power management, virtualization and
consolidation technologies.
In many cases, a power management feature is already
on servers. Often turning on this feature is as simple as
pushing a button in an administrative console. The use of
power management can bring immediate power savings.
Virtualization is another approach that can help you hold
the line on rising power and cooling costs. Virtualization
allows you to pool and share servers, storage devices
and other IT assets to increase utilization. This allows
you to pack more computing capacity into the same
footprint. You can then consolidate hardware devices
to reduce the number of physical servers and storage
systems in your environment, which helps reduce your
power and cooling costs.
Step 2. Apply best practices.
The best practices outlined above can help you optimize
power and cooling efficiency.
Even if you can’t do them all, various combinations
of these best practices can help you increase energy
efficiency, reduce your energy bills and operate in a
more environmentally friendly manner.
Step 3. Tune your environment.
In this step, the focus shifts to tuning and optimizing
your overall data center. The use of computational fluid
dynamics (CFD) and other engineering tools can give
you a better understanding of your power and cooling
issues and opportunities. CFD, for instance, can help
you determine the best placement for equipment and
floor tiles to enable efficiency gains and power savings.
Step 4. Closely couple cooling solutions with individual
heat loads.
The fourth step in this process is to closely couple your
cooling solutions with the heat loads of individual pieces
of IT equipment. By narrowly targeting your cooling
capacity you can greatly increase the efficiency of your
cooling solutions.
These steps can help you cut costs and save energy, as
shown in Figure 3. In a typical data center, by simply
doing best practices, we project that you could save as
much as 20 percent of the power you are currently using.
If you tune further, using computational fluid dynamics,
your potential savings increase to 33 percent. And if you
look at new technologies for closely coupled cooling,
you might push your savings to as high as 47 percent.
11
These same steps might make it possible to put more
equipment into your data center. With these approaches,
you could conceivably nearly double the amount of IT
equipment in your data center.
How HP can help you meet your
energy and environmental challenges
HP offers a wide range of services and technologies
to help you address the energy and environmental
challenges that come with running a modern data center.
These offerings build on years of HP research, which
has yielded more than 1,000 energy-efficiency patents.
HP Dynamic Smart Cooling
HP Dynamic Smart Cooling technology sends cool air
to where it is most needed, when it is needed. With
this HP innovation, sensors deployed on racks feed
real-time thermal data to your management software.
This allows your computing equipment to interact with
the power and cooling facilities that support your data
center. This smarter, more holistic approach to cooling
is one of the keys to cutting your power bills.
HP IT Consolidation Solutions
IT complexity can impact the growth of your business and
drive up costs. Our IT Consolidation solutions respond
to this challenge with offerings that help you simplify,
optimize and automate your IT environment. From
the data center to the workplace, HP IT Consolidation
solutions transform your IT environment, shifting IT budget
from operations to innovation.
6
10
Belady, C., “How to Minimize Data Center Utility Bills,” Line 56 (September
5, 2006)
11
Malone, C., Belady, C., “Metrics to characterize Data Center and IT Equipment
Energy Use,” Proceedings of 2006 Digital Power Forum, Richardson, Texas
(September 2006)
pg_0007
HP Data Center Thermal Assessment Services
These services leverage sophisticated modeling tools and
techniques to determine the unique thermal conditions in
your data center. HP Services professionals can then
recommend changes to optimize climactic conditions and
reclaim capacity. These changes can help you postpone
or greatly reduce the need for costly mechanical
upgrades. The information generated by thermodynamic
modeling helps you understand what you need to do to
improve the efficiency of your data center and to gain
more value from your energy expenditures.
HP Thermal Logic technology
This groundbreaking HP technology helps control power
consumption and thermal output in HP BladeSystem,
ProLiant and Integrity servers. HP Thermal Logic
technology takes a holistic approach to power and
cooling management. It helps you control the balance
between power and cooling to boost data center energy
efficiency at the component, enclosure and rack levels.
Ultimately, this technology can make it possible to pack
more computing power into your data center while
holding the line on power consumption.
HP Insight Power Manager
HP Insight Power Manager is an integrated power
monitoring and management application that provides
centralized control of server power consumption and
thermal output at the data center level. It provides a
single pane of glass that you can use to measure,
regulate and cap power usage to make more efficient
use of your infrastructure. It extends the capacity of your
data centers by enabling you to reduce the amount of
power and cooling required for HP BladeSystem, ProLiant
and Integrity servers.
HP Modular Cooling System
The HP Modular Cooling System uses chilled water
and closed racks to supply uniformly distributed cool
air across the entire front of the server rack. This spot-
cooling solution allows hardware densities and power
consumption levels that in the past would have been
difficult, if not impossible, to support. With this cooling
technology, you can run up to 35 kilowatts per rack—
triple the standard cooling capacity of a single rack.
This means you can increase server densities within
the same footprint.
HP StorageWorks XP Thin Provisioning sof tware
HP StorageWorks XP Thin Provisioning software enables
you to allocate your anticipated future storage capacity
needs from a virtual storage pool. It reduces the amount
of physical disk capacity required. By reducing allocated
but unused storage you can eliminate stranded storage.
HP StorageWorks Dynamic Capacity Management
EVA software
HP StorageWorks Dynamic Capacity Management EVA
software continuously monitors user-set thresholds and
automatically grows or shrinks the storage/host volumes
to map data resources to actual data needs within the
HP StorageWorks EVA. You can also manually grow
and shrink volumes if needed.
The road ahead
So where do go from here, and what lies ahead on
the power and cooling horizon? Based on current
issues and trends, here is some of what we expect
in the road ahead.
7
Figure 3. Cutting costs, saving
energy
Here is a look at the potential
impacts of the use of best
practices, tuning and closely
coupled cooling solutions in a
typical data center.
10
Potential impact of best practices, CFD and closely coupl ed solutions on a typical data center
Energy savings potential for many data centers
10 0 %
80%
60%
40%
20%
0%
PUE=3.0
typical
PUE=2.4
best
practices
PUE=2.0
CFD
PUE=1.6
closely
coupling
Energy savings
Cooling
Power
IT
Potential increase in IT equipment with no additional energy use
10 0 %
80%
60%
40%
20%
0%
PUE=3.0
Typical
PUE=2.4
best
practices
PUE=2.0
CFD
PUE=1.6
closely
coupling
Cooling
Additional IT
Power
IT
Optimization can delay building
another costly data center
Optimization can make significant
impact to profit margin
25% 50% 88%
21 % 3 3 % 47 %
Malone, C., C. Belady, “Metrics to characterize Data Center and IT Equipment Energy Use,” Proceedings of 2006 Digital Power Forum, Richardson, Texas (September 2006)
pg_0008
Integrated power and cooling management
Companies are going to work actively to achieve the
seamless integration and management of power, cooling
and compute resources. HP Labs is currently doing a
substantial amount of work in this area. The ultimate goal
is to optimize the complete data center ecosystem in a
manner that allows diverse technology components to
work together in a more energy efficient manner.
The widespread use of metrics
We also expect to see broad adoption of industry metrics
for data center power and cooling efficiency and for
server efficiency. In the years ahead, metrics will likely be
widely used for benchmarking purposes. These will
include new metrics from the Standard Performance
Evaluation Corporation (SPEC) benchmarking body.
Increasing commoditization
The years ahead will bring increasing commoditization
of the data center. The data center of the future is likely
to be filled with plug-and-play equipment. Companies
will strive to deploy servers that are designed to the
same environmental specifications. This trend is already
under way, in the form of emerging industry guidelines
from organizations such as ASHRAE. These wide-ranging
guidelines attempt to define standard data center
environments and configurations, all the way down
to target environmental temperatures and humidity
specifications for servers and facilities.
The rise of new cooling technologies
We expect to see the emergence of new technologies
that allow cooling capacity to be more closely targeted
at heat sources. These will include liquid cooling
technologies that allow cooling solutions to be placed
inside server racks for higher levels of efficiency.
An emphasis on sustainable computing
For companies around the world, the years ahead
will require an increasing emphasis on sustainable
computing practices that help reduce power consumption
and cut operational costs. This focus on energy conser
-
vation isn’t just about saving money and stemming the
dramatic increases in power and cooling costs. This is
just as much about reducing the consumption of natural
resources and the release of C0
2
emissions.
With the use of natural resources and greenhouse
emissions increasing every year, it’s no surprise that
enterprises around the world face a growing range of
environmental regulations—and a growing realization
that the status quo has to change. In enterprise environ
-
ments, the future is likely to bring a constant focus on
energy conservation and sustainable computing practices.
Creating an energy-efficient data center
Power and cooling isn’t just your problem. It’s an
industry problem, a technology problem and a global
environmental problem.
The technology industry is responding. Across the
industry, new technologies and new initiatives are
emerging to address power and cooling issues. These
efforts will allow your organization to take greater
advantage of the next generation of server and
storage technologies.
At HP, we are taking a leadership role in helping
organizations like yours address your power and
cooling challenges. We’re driving energy-efficiency
initiatives. We’re working with industry partners to
leverage innovative energy-aware technologies that
put control of energy usage back into the hands of
data center managers. We’re researching, designing
and developing new solutions to address future IT needs.
Ultimately, these diverse efforts come down to a narrow
point of focus: At HP, we want to help your organization
create an energy-efficient data center that allows you to
control costs, conserve energy, cut your carbon footprint
and achieve sustainable business outcomes.
To learn more
To learn more about data center power and cooling
solutions, visit: www.hp.com/go/energyefficiency.
To learn more, visit www.hp.com/go/energyefficiency
© Copyright 2007 Hewlett-Packard Development Company, L.P. The information contained herein is subject to
change without notice. The only warranties for HP products and services are set forth in the express warranty
statements accompanying such products and services. Nothing herein should be construed as constituting an
additional warranty. HP shall not be liable for technical or editorial errors or omissions contained herein.
4AA1-4756ENW, September 2007