– Green Datacenter –

1- Efficiency Best Practice

This assessment provides a baseline on how to improve data center energy efficiency and reliability. The resulting
report suggests areas for improvement, identifies best practice opportunities, and provides guidance for
prioritization of efficiency projects, which our professional service team can help you undertake.

Air Conditioning System Design and Capacity

• Air Flow – Evaluate volume, velocity, cold air supply and hot air return

• Plenums – Assess volume and pressure in plenum spaces to ensure sufficient server intake airflow

• Relative Humidity Findings – Evaluate relative humidity for ASHRAE compliance and issues

• Temperature Findings – Gauge the thermal conditions for ASHRAE compliance and situations where
temperatures could impact equipment reliability or availability

Electrical System Design and Capacity

• Standby Generators – Evaluate generator capacity relative to current and planned equipment loads

• UPS – Examine maintenance and testing strategy as well as assess redundancy attributes

Infrastructure Observations and Measurements

• Cable Management – Evaluate cable sub-system for airflow blockage situations

 • Fire Protection – Assess viability of fire protection system and impact on hardware if activated

• Overhead Ducting – Evaluate the airflow in ducting or opportunity for conversion of false ceiling space as a
return airflow plenum

• Raised Floor – Inspect the floor structure for strength, seismic design, test for electrical noise levels and
grounding best practices

General Environmental Conditions

• Cleanliness – Physical inspection of the data center that identifies situations and practices that can
contribute to excessive particulate matter and other issues

• Electromagnetic Interference (EMI) – Evaluate the environment for magnetic fields that can effect
equipment performance

• External Environmental Considerations – A more broad review of the site and installation that identifies
external conditions that can cause data center performance issues

• Noise Levels – Measure ambient noise levels to ensure OSHA compliance

• Radio Frequency (RF) Emissions – Determine presence and level of high frequency interference

• Security – Evaluate site security relative to industry standards

Efficiency Assessment

• Power Usage Effectiveness (PUE) – Establish a value for and describe the implications of this widely
recognized efficiency benchmark from The Green Grid that enables benchmarking for tracking efficiency
improvements over time

• Data Center Infrastructure Efficiency (DCiE) – Establish a value for and describe the implications of this
widely recognized efficiency benchmark from The Green Grid that enables benchmarking for tracking
efficiency improvements over time

2 - Data Center Energy Efficiency

Air Conditioning System Design and Capacity

• Air Flow – Evaluate air volume, velocity, cold air supply and hot air return

• Capacity Planning – Evaluate total cooling capacity relative to current and planned usage

• Computational Fluid Dynamics (CFD) Modeling

• High Density Cabinets – Examine high density cabinets for sufficient airflow and temperature levels

• Hot Spots – Identify cabinets where there is insufficient airflow

• Plenums – Assess volume and pressure in plenum spaces to ensure sufficient server intake airflow

• Relative Humidity Findings – Evaluate relative humidity for ASHRAE compliance and issues
 • Stranded Capacity – Identify situations where the sub-system design creates cooling capacity that is
impossible or difficult to deploy

• Temperature Findings – Gauge the thermal conditions for ASHRAE compliance and situations where
temperatures could impact equipment reliability or availability

Electrical System Design and Capacity

• Capacity Planning – Evaluate total power capacity relative to current and planned usage

• Standby Generators – Evaluate generator capacity relative to current and planned equipment loads

• Stranded Capacity – Identify situations where the sub-system design creates power capacity that is
impossible or difficult to deploy

• UPS – Examine maintenance and testing strategy as well as assess redundancy attributes

Infrastructure Observations and Measurements

• Cable Management – Evaluate cable sub-system for best practice installation

• Fire Protection – Assess viability of fire protection system and impact on hardware if activated

• Overhead Plenum – Evaluate the space above the drop ceiling for conversion to a return airflow plenum

• Raised Floor – Inspect the floor structure for strength, seismic design, measure electrical noise levels and
evaluate grounding best practices

Efficiency Assessment

• Power Usage Effectiveness (PUE) – Establish a value for and describe the implications of this widely
recognized efficiency benchmark from The Green Grid that enables benchmarking to measure
improvement when changes are implemented

• Data Center Infrastructure Efficiency (DCiE) – Establish a value for and describe the implications of this
widely recognized efficiency benchmark from The Green Grid that enables benchmarking to measure
improvement when changes are implemented

General Environmental Conditions

• Cleanliness – Physical inspection of the data center that identifies situations and practices that can
contribute to excessive particulate matter and other issues

• Electromagnetic Interference (EMI) – Evaluate the environment for magnetic fields that can effect
equipment performance

• External Environmental Considerations – A more broad review of the site and installation that identifies
external conditions that can cause data center performance issues

• Noise Levels – Measure ambient noise levels to ensure OSHA compliance

• Radio Frequency (RF) Emissions – Determine presence and level of high frequency interference

• Security – Evaluate site security relative to industry standards

3- Real-Time Monitoring & Efficiency Benchmarking

Phase 1: Survey
The first phase is a non-invasive inspection and data collection effort, performed with the client's data center and
facilities staff to ensure security and operational guidelines are met as well as to help the staff to gain a deeper
understanding of energy efficiency best practices. The data gathered includes floor plans, sub-system design
documents, equipment configurations, operational settings and data center management objectives.

Phase 2: Planning
The 42U team then develops an instrumentation plan. This includes a measurement point design that determines
the data and locations that will be monitored. The data collected in the survey phase is used to populate the
monitoring system database. A timeline is developed that ensures the appropriate technical staff is available as
required for the installation phase. Security and other considerations are included in the plan.

Phase 3: Installation
Once the design and timeline are approved the 42U team then configures and tests the sensor and network
designs. The equipment is then shipped to the client for installation in the data center. The 42U team deploys the
sensor, network and monitoring technologies. Finally, the 42U team tests and verifies that the monitoring system is
operational and accurately collecting data.

Phase 4: Measure
This is the real-time data collection phase. 42U works with the client to ensure that data is collected accurately and
the client becomes familiar with the capabilities of the monitoring system and management software. At the end of
the agreed-upon measurement period the client provides 42U with a copy of the monitoring database. The 42U
team then analyzes the data and produces a report based on the site inspection, staff interviews and sensor data.
The report includes the following sections as appropriate:

Executive Summary

Project Overview

• Project Scope – Define the coverage of the project to ensure expectations are correctly
established

• Sensor Solution – Describe the configuration of the sensor solution

Thermal Data
 • Thermal Map – Provide a color-coded image showing temperature ranges throughout the data
center

• Overcooling – Use thermal map and cabinet data to identify areas where cooling levels are below
ASHRAE standards

• Overheating – Use thermal map and cabinet data to identify areas where cooling levels are above
ASHRAE standards

• Air Mixing – Identify areas where heated air is infiltrating the chilled airflow

• Thermal Anomalies – Describe anomalous thermal findings

Air Conditioning System

• Air Flow Assessment – Evaluate air volume, velocity, cold air supply and hot air return

• Bypass Airflow – Identify conditions where chilled airflow is returning to cooling equipment without
cycling through IT equipment

• Cooling Infrastructure Capacity Analysis – Evaluate total cooling capacity relative to current and
planned usage

• Cooling Infrastructure Operational Behaviors – Identify inefficient operational behaviors

• Overhead Ducting Analysis – Evaluate sub-system for inefficiencies

• Plenum Pressure Analysis – Identify issues that can impact equipment availability or performance

• Plenum Pressure Map – Provide a color-coded image showing pressure ranges throughout the
plenum

• Raised Floor Analysis – Evaluate raised floor system for strength, seismic design, electrical noise
levels and grounding best practices

• Relative Humidity Data – Identify situations where relative humidity is not compliant with ASHRAE
standards

• Relative Humidity Map – Provide a color-coded image showing relative humidity ranges
throughout the data center

Electrical System

• Power Infrastructure Capacity Analysis – Evaluate total power capacity relative to current and
planned usage

• Power Infrastructure Operational Behaviors

• Standby Generators – Evaluate generator capacity relative to current and planned equipment
loads

• UPS – Examine maintenance and testing strategy as well as assess redundancy attributes

Infrastructure Observations and Measurements

• Cable Management – Evaluate cable sub-system for airflow blockage situations

• Fire Protection – Assess viability of fire protection system and impact on hardware if activated

• Overhead Ducting – Evaluate the airflow in ducting or opportunity for conversion of false ceiling
space as a return airflow plenum
 • Raised Floor – Test the floor structure for strength, seismic design, electrical noise levels and
grounding best practices

General Environmental Conditions

• Cleanliness – Physical inspection of the data center that identifies situations and practices that
can contribute to excessive particulate matter and other issues

• Electromagnetic Interference (EMI) – Evaluate the environment for magnetic fields that can effect
equipment performance

• External Environmental Considerations – A more broad review of the site and installation that
identifies external conditions that can cause data center performance issues, such as microwave
transmission towers generating interference

• Noise Levels – Measure ambient noise levels to ensure OSHA compliance

• Radio Frequency (RF) Emissions – Determine presence and level of high frequency interference
from wireless devices

• Security – Evaluate site security relative to industry standards

Energy Efficiency Assessments

• PUE – Establish a value for and describe the implications of this widely recognized efficiency
benchmark from The Green Grid that enables comparison across data centers

• Data Center Infrastructure Efficiency (DCiE) – Establish a value for and describe the implications
of this widely recognized efficiency benchmark from The Green Grid that enables comparison
across data centers

Efficiency Improvement Recommendations

• PUE/DCiE Improvements – Identification of specific steps that can be taken to improve these
measures

• Sub-Systems – Sub-System-specific recommendations that will improve efficiency and reliability

3- Troubleshooting Investigation and Assessment

Air Conditioning System Design and Capacity

• Air Flow – Evaluate air volume, velocity, cold air supply and hot air return

• Hot Spots – Diagnose hot spots and evaluate possible adjustments

• Plenums – Assess volume and pressure in plenum spaces to ensure sufficient server intake airflow

• Relative Humidity Findings – Evaluate relative humidity for ASHRAE compliance and issues

• Temperature Findings – Gauge the thermal conditions for ASHRAE compliance and situations where
temperatures could impact equipment reliability or availability

Electrical System Design and Capacity

• Standby Generators – Evaluate generator capacity relative to current and planned equipment loads

• UPS – Examine maintenance and testing strategy as well as assess redundancy attributes
Infrastructure Observations and Measurements

• Cable Management – Evaluate cable sub-system

• Fire Protection – Assess viability of fire protection system and impact on hardware if activated

• Floor Layout – Evaluate layout for energy efficiency best practices

• Overhead Ducting – Evaluate the airflow in ducting or opportunity for conversion of false ceiling space as a
return airflow plenum

• Raised Floor – Inspect the floor structure for strength, seismic design, test for electrical noise levels and
evaluate grounding practices

General Environmental Conditions

• Cleanliness – Physical inspection of the data center that identifies situations and practices that can
contribute to excessive particulate matter and other issues

• Electromagnetic Interference (EMI) – Evaluate the environment for magnetic fields that can effect
equipment performance

• External Environmental Considerations – A more broad review of the site and installation that identifies
external conditions that can cause data center performance issues

• Noise Levels – Measure ambient noise levels to ensure OSHA compliance

• Radio Frequency (RF) Emissions – Determine presence and level of high frequency interference

• Security – Evaluate site security relative to industry standards

Data Center Economizer Solutions

One of the most exciting and talked about best practices for data center efficiency is “free cooling”—using the
ambient environment to augment or replace mechanical air conditioning. Free cooling is achieved through
economizers, classified as either air-side or water-side depending on the medium used for cooling.

While these solutions are not entirely “free,” they can be extremely effective in the right geography.

Air Side Economizers

One of the most exciting and talked about best practices for data center efficiency is often labeled "free cooling"—
the use of Air-side Economizers to dramatically reduce your HVAC related energy consumption and costs. Air-Side
economizers have become a talking point, it seems, in any current data center cooling discussion. Facilities
professionals will point out that economizers have been around for a long time, and some data centers have indeed
used them for years. In the wake of exorbitant energy bills, the concept of economizers—and "free cooling"—is
gaining traction throughout the industry. End users have become interested in the economics (CapEX, ROI),
installation, and operation of these devices. The Green Grid recently announced the introduction of an interactive
free cooling map to help data center managers determine the amount of days they may be able to take advantage
of Air-Side Economizers in their geographical location. With this Green Grid tool, you will be able to enter in your zip
code and see the number of days your location falls within the ASHRAE recommended temperature and humidity
ranges.

What are Air-Side Economizers?

Mechanical cooling, depending on the source, is estimated to consume anywhere from 33% to 40% of a facility's
incoming electricity. Designed to accompany or circumvent this process, Air-Side economizers can bring Mother
Nature into the data center whenever the ambient conditions are favorable.

The outside air is brought into building and distributed via a series of dampers and fans. The servers ingest the cool
air, transfer heat, and expel hot air to the room. Instead of being recirculated and cooled, the exhaust is simply
directed outside. If the outside air is particularly cold, the economizer may mix the inlet and exhaust air, ensuring
that the resulting air temperature falls within the desired range for the equipment.
The economizer design is typically integrated into a central air handling system with ducting for both intake and
exhaust. The equipment includes filters to reduce the amount of particulate matter or contaminants that are
brought into the data center space.

Air-Side Economizer Configuration

A quick internet search shows a lot of discussion on the concept but very little practical information on installing an
Air-Side economizer within an existing data center. Instead, most articles emphasize proper design and alignment
within the larger cooling architecture of a new building. Data center authority Pacific Gas and Electric recommends
that the economizer and its controls are engineered "at the schematic design stage, where the required
architectural accommodations can be made at little or no additional cost."

Software giant Oracle addressed the "architectural accommodations" in its existing facility during the EPA's 2006
Conference on Enterprise Servers and Data Centers. To retrofit Air-Side economizers in its Tier IV, 45,000 sq ft
facility, Oracle would require a 3400 sq ft opening to move 1,710,000 cfm of air. Within their current layout, they
found no convenient way to distribute the outside air from the CRAC to the raised floor. For them, the economizer
was not a practical retrofit.

It was, however, for Red Rocks Data Center in Morrison, CO. The company in 2008 added an Air-Side economizer to
their portfolio, significantly reducing their dependency on mechanical air conditioning. The facility, nestled in the
foothills of the Rocky Mountains, estimates that they'll be able to utilize the mountain air for 80% of the year
(Reuters, 2008).

Though the approach will vary by the facility, the goal for all economizers is the same: "for all data center air
handlers to have access to 100 percent outside air as well as return air" (Rumsey, 2007). This air, properly
controlled and distributed, can significantly cut into the energy bill.

Air-Side Economizer Environmental Considerations

The data center has traditionally been a very controlled environment. Though this ecosystem has come at a price—
economically and ecologically—users are not going to compromise uptime, performance, or equipment longevity.
Introducing outside air and its by-products —higher temperatures, humidity fluctuations, dust/dirt—to this space
with thousands of dollars of servers, storage, and networking gear may seem illogical to some. Before we review
these concerns, point by point, here's an overview of ASHRAE TC 9.9's temperature and humidity ranges:

ASHRAE TC 9.9 Recommended Allowable

Temperature 18-27 deg C (64.4- 80.6 Degrees F) 15-32 deg C (59-90 Degrees F)

Humidity 5.5-15 deg C Dew Point (41.9-59 Degrees F ) 20%-80% RH

** Conditions reflect air ENTERING the IT equipment**

Temperature
Rackmount hardware requires a continuous supply of cool air, as indicated by the ASHRAE range. The more
temperate his environment, it would seem, the more a particular user could utilize ambient air for this supply. But,
as engineer Vali Sorell notes, virtually all environments can take advantage of the technology: "Common sense
would normally dictate that an outside air economizer in the hottest climate would not have a good payback. That
logic may be applicable to a typical office building where there are approximately 2,500 hours of use in a year.
However a data center must run continuously, 24 hours a day, for a total 8,760 hours per year."Hotter climates can
leverage cooler overnight temperatures or seasonal changes to incorporate economization.

To assuage concerns over varying temperatures, the economizer works in tandem with the central air handling
system to maintain a palatable supply temperature. Through its controls and measurement devices, the
economizer will compare the relevant temperature settings (outside air, server inlet, server exhaust) to determine
the most economical cooling approach, whether full economization (100% outside air), partial economization, or
mechanical cooling.

Humidity
At extremes, humidity presents challenges for computing equipment. Too high, users worry about condensate
forming on the equipment. Too low, users caution about electrostatic discharge. ASHRAE's guidelines, for these
reasons, seek a happy medium.
If facilities choose not to measure humidity, the economizer can operate strictly based on temperature differentials.
With humidity included, the economizer must perform additional measurements. If the outside air is cool but too
dry, the system may expend extra energy to humidify the air. Thus, proper management and controls, as discussed
within the temperature section, are imperative to ensure the right volume of air at the right temperature and
humidity are introduced into the space.

Particle Contamination
Pacific Gas and Electric, along with Lawrence Berkley National Labs, compared particle concentrations in data
centers with economizers against those who introduce minimal, if any, outside air. The results were expected.
Facilities with economizers have higher particle concentrations inside the data center. However, an improved filter
design can mitigate any contamination concerns (Tschudi, 2007). Using ASHRAE Class II, 85% filters, the authors of
the report envisioned that the particle counts in economizer facilities would rival the small amount found in non-
economizer data centers (In their study, data center CRACs and air handlers were equipped with ASHRAE Class I,
40% filters).

Maximizing Economizer Efficiency

The efficiency gain for Air-Side economization is obvious. Running a data center on outside air or as some have said
"opening the windows in the data center," significantly reduces or eliminates the chiller's energy consumption.
Parts common in mechanical cooling, like compressors and pumps, are inactive, as the fans (both inlet and
exhaust) become the primary mechanism in removing heat. As a result, most of the incoming electricity can be
devoted to the mission critical hardware and applications.

Yet this process isn't as simple as opening the windows in your home. There are several considerations for
maximizing the efficiency of the economizer.