3 Steps to Calculate Total Enterprise

IT Energy Consumption Using DCIM

White Paper 100

Version 1

Energy Management Research Center

by Robert Moore and Patrick Donovan

Executive summary
Reducing the environmental impact of enterprise IT
begins with collecting data and reporting metrics to
establish a baseline. But with dozens of potential
metrics and complex accounting to consider along
with the highly distributed and diverse nature of hy-
brid IT portfolios, this can be a daunting task. For
many CIOs and their IT operations teams, this is also
a new task. This paper is a straightforward, “how to
get started” guide that identifies energy consump-
tion as the fundamental data point to start with and
gives a simple 3-step framework to start measuring
the environmental impact of the IT estate.

RATE THIS PAPER 

Schneider Electric – Energy Management Research Center White Paper 100 Version 1 2

Introduction Measuring, reporting, and managing a company’s sustainability performance is

starting to drive employee retention, corporate reputation, and contribute to busi-
ness success. Chief Sustainability Officers (CSOs) will report environmental sustain-
ability metrics for the entire enterprise. However, to track progress against the en-
terprise’s goals and determine where inefficiencies are and improvements can be
made, CSOs will need each department or function in the organization to provide
more granularity in their measurement of sustainability metrics. IT is one important
department where there is much interest in understanding energy consumption
and, ultimately, its greenhouse gas (GHG) emissions.

Schneider Electric White Paper 67, Guide to Environmental Sustainability Metrics for
Data Centers, helps address this challenge specifically for data centers by identify-
ing and explaining the key metrics to track based on their maturity. However, CIO’s
have an IT estate that goes well beyond just on-premise data centers to also in-
clude public cloud, colocation providers, distributed IT sites (i.e., edge computing),
and a work-from-home (WFH) end user population.

We recommend that CIOs who are beginning their sustainability journey start
with measuring just one metric: total energy consumption. This fundamental met-
ric shows the direct impact of IT operations and is used by CSOs to calculate GHG
emissions. Energy consumption is also good to start with because the data sources
and collection tools are readily available today, and in many cases, are likely al-
ready in use by your IT operations team.

This paper provides a simple framework to measure and track IT energy con-
sumption across the entire enterprise IT portfolio. There are three steps:

Step 1: Assess the coverage of your IT estate

Step 2: Determine metric data sources and deploy metering and software
Step 3: Use data center infrastructure management (DCIM) to aggregate data and
report total energy consumption

The paper will explain how to do this across the entire IT estate as summarized in
Figure 1.

OUTSOURCED ENTERPRISE-OWNED
INFRASTRUCTURE INFRASTRUCTURE
DCIM ON PREMISE
BMS DATA CENTERS
EPMS
PUBLIC CLOUD
Figure 1 Vendor
report DCIM
A view of a typical enterprise IT Dashboard
estate showing data flow for the DISTRIBUTED IT
SITES
total energy consumption met-
ric and the software manage- COLOCATION DCIM
ment tools that collect and send
that data to dashboards. Vendor
report End-user Devices

Enterprise-wide ITAM
Sustainability
Dashboard

3 Steps to Calculate Total Enterprise IT Energy Consumption Using DCIM

Schneider Electric – Energy Management Research Center White Paper 100 Version 1 3

Following the guidance in this paper creates a solid foundation to collect and report
additional metrics (e.g., water use, Scope 3 emissions like embodied carbon, and
so on) as the organization’s sustainability program matures.

Next, we will explain the total energy consumption metric.

The “total energy This is the total energy consumed, measured in kilo or mega-watt hours (kWh,
MWh), to operate outsourced and enterprise-owned infrastructure. This is typically
consumption” the electrical energy drawn from the utility grid but would also include any energy
metric production onsite from generators, solar, or wind. Energy imported in the form of
natural gas, steam, or chilled water should also be counted, if applicable.

Energy = Power (watts) * Time (hrs)

Ideally, energy is measured in real time using power meters. This could be the
utility meter (for core data centers), branch circuit power meters, or meters built-in
to an uninterruptable power supply (UPS), power distribution unit (PDU), or rack
PDU (rPDU). Power meter data would then be sent to infrastructure management
software which aggregates and presents the information for users and stakeholders
through its dashboard or report. The infrastructure management software tools re-
ferred to in this paper include:

• data center infrastructure management (DCIM)

• building management system (BMS)
• electrical power monitoring system (EPMS)
• IT asset management (ITAM)

Note, devices and meters may report energy and/or power in varying ways in terms
of the unit of measure and polling intervals. It is important that the infrastructure
management software can normalize this data into a consistent energy consump-
tion metric. This complication is a reason why the the use of simple spreadsheets
instead of purpose-built management software untenable.

In cases where there is no meter (e.g., end-user devices), energy consumption is

estimated based on the device power rating from the vendor and assumed time du-
ration of use. Note, we recommend taking 60% of the manufacturer’s power rating
for a given device to more accurately reflect what it actually draws in operation.

In situations where upstream power and cooling infrastructure is shared (e.g.,

distributed IT sites) or its energy data is not attainable (from a BMS/EPMS), the total
IT energy can be multiplied by the site’s power usage effectiveness (PUE) value to
get a total energy number that accounts for all the power losses of the supporting
power and cooling infrastructure. Schneider Electric provides an easy way to
quickly estimate PUE for a given location through this freely-available online tool
called the Data Center Efficiency & PUE Calculator. Note some DCIM software tools
will calculate PUE in real time for the monitored site assuming power data is being
collected from the infrastructure. For more information about calculating PUE for
data centers, see Schneider Electric White Paper 113, “Electrical Efficiency Model-
ing for Data Centers”.

Next, we will explain the 3-step framework for calculating and reporting the total en-
ergy consumption metric.

3 Steps to Calculate Total Enterprise IT Energy Consumption Using DCIM

Schneider Electric – Energy Management Research Center White Paper 100 Version 1 4

Step 1: With Figure 1 in mind, this step is about ensuring you’re accounting for everything
that consumes energy across your portfolio of sites and that you understand where
Assess the gaps exist. Having as full of an accounting as possible ensures the metric is mean-
coverage of your ingful and represents reality. This means that end user devices and outsourced in-

IT estate frastructure should be included, as well. Energy consumption, and sometimes even
the GHG emissions data, for public cloud and colocation services is typically pro-
vided by the vendor which makes this portion of your IT estate easier to account for.

For enterprise-owned infrastructure, however, you need to assess whether power

meters are in place and software management tools are in use. Power meters and
management software is what gives visibility to on premise data center and distrib-
uted IT site and device energy consumption. End user device energy consumption,
on the other hand, would ultimately be derived from data in the ITAM tool. Note, the
next section of this paper, “Determine metric data sources and deploy metering
and software”, will explain your energy data source options.

This inventorying and assessment process may expose gaps in visibility to your IT
and/or its supporting infrastructure. This is most common at smaller distributed,
edge IT sites. Asset management and monitoring software is always recommended
for complete accounting of your estate to ensure system reliability, security, and a
full accounting of the sustainability impact of IT operations. Your organization will
need to come up with a plan and budget to address the gaps. Enlisting the support
of corporate-level sustainability officers and teams, if they exist, would likely be
helpful.

The following sub-sections explain further what infrastructure should be included

and how energy consumption can be calculated.

On -p re m is e d a ta c e n te rs
An on-premise data center may be an entire standalone building dedicated to IT
operations, or it may be a data center inside a mixed-use building. For standalone,
dedicated data centers, 100% of the energy consumption for each site should be
accounted for. For data centers in mixed-use buildings, assessing energy con-
sumption may be more challenging. In a mixed-use facility, the data center share of
the overall energy consumption must be estimated or derived. One option in this
mixed-use scenario is to use the UPS output power and multiply by the PUE to ac-
count for upstream infrastructure power losses.

Systems to be accounted for in data centers include:

• Servers, storage, and networking devices (i.e., the IT equipment)

• Electrical & mechanical (HVAC) systems
• UPSs & generators
• Lighting, security systems

Dis trib u te d IT s ite s

These sites are typified by having a small number (<10) of IT rack enclosures and
distributed non-racked IT equipment. Edge compute sites often involve shared in-
frastructure systems. The simplest approach would be to only use the energy con-
sumption of the UPS supplying the IT load plus the estimated energy consumption
of any non-racked IT equipment dispersed at the site. A more accurate accounting
would involve taking the UPS output and multiplying by the estimated PUE and then
adding in the non-racked IT power consumption.

3 Steps to Calculate Total Enterprise IT Energy Consumption Using DCIM

Schneider Electric – Energy Management Research Center White Paper 100 Version 1 5

Systems to be accounted for in edge compute sites include:

• Racked IT equipment
• Non-racked IT equipment dispersed at site

En d -u s e r d e vic e s
For many enterprise companies a significant percentage of their workforce now
works from home (WFH) using enterprise-owned devices like laptops, monitors,
printers, and cell phones. The McKinsey Institute estimated that these enterprise
end-user devices can consume almost half as much energy as on-premise data
centers 1. Given their significance in terms of energy consumption and emissions,
they should be counted in the total energy consumption metric.

While we believe it is best to track actual energy consumption, this may not always
be feasible, particularly for end user devices and WFH IT equipment where meter-
ing is not practical. In those cases, the metric data source would be an estimate of
energy use based on device power draw and how much time it is in use to arrive at
an energy figure.

Note, there are small residential, cloud-connected UPSs available on the market
which can report energy data about the connected load that could be collected and
added into reported sustainability metrics. For enterprises who choose to help en-
sure power availability for WFH users with small UPSs, this could be an option for
collecting actual energy consumption data.

Ou ts o u rc e d in fra s tru c tu re
Outsourced infrastructure includes the company’s use of public cloud services and
colocation data centers. In the case of the public cloud, service providers should
issue energy consumption and emissions data accounting for the use of cloud ser-
vices and workloads. This may be in the form of a report, dashboard, or as part of
the billing invoice.

For colocation data centers, assuming you do not monitor and manage your own in-
frastructure via your own DCIM, the colocation provider should offer a vendor report
to show energy consumption and emissions data for your enterprise IT deployed in
their data center(s). Like the cloud provider, this data could also be given via a
dashboard or perhaps be attached to the billing invoice. Also, some colocation pro-
viders use commercial DCIM software that gives their customers a “tenant portal”.
This provides tenants real-time visibility into their asset status including their total
energy use.

Step 2: The next step is to identify your sources of energy data across your entire estate
and make sure that data is being reported to management software or, in the case
Determine metric of end user devices, is available in your ITAM tools. Gaps in visibility identified in
data sources and Step 1 should be filled by deploying power metering and/or management software

deploy metering where needed.

and software Our recommended data sources for energy consumption across your enterprise IT
estate are summarized in Table 1.

1
McKinsey, The green IT revolution: A blueprint for CIOs to combat climate change, 9/2022

3 Steps to Calculate Total Enterprise IT Energy Consumption Using DCIM

Schneider Electric – Energy Management Research Center White Paper 100 Version 1 6

Enterprise IT
Recommended Energy Consumption Data Sources *
infrastructure
On-premise Utility meter at site or utility energy invoices
Data Centers
Table 1 Built-in power meter of UPS**
Edge Computing Sites
Recommended Estimation based on power ratings and time used, either manually collected
data sources for End User Devices or provided by ITAM software
energy consump-
tion Outsourced Vendor provided via sustainability report, online dashboard, billing invoice,
Infrastructure or DCIM “tenant portal”

*There are more advanced, granular methods for measuring energy consumption in each area, but
for simplicity we have focused on the simplest method to get started.
**Note, some UPSs will report actual energy vs. just power.

De p lo y e le c tric p o we r m e te rs wh e re a b s e n t
For the purposes of measuring energy consumption for sustainability tracking, gen-
eral power meters – either permanently installed as a standalone meter or the ones
found built into UPSs – provide a cost-effective means for collecting the data for
calculating the total energy consumption metric. For power meters built into a UPS
or rack power distribution unit, there is no extra cost for that functionality.
Standalone power meters used at the service entrance mains or PDUs, for example,
are lower cost than power quality sensors and tend to be very simple to use.

Table 1 above gives the basic, minimum requirement to capture the data needed
for overall energy consumption related to IT operations. Users may decide to add
additional meters to increase the granularity of the energy data being collected to
understand, for example, the difference in energy consumption between various
subsystems including IT, power, and cooling to calculate a more accurate PUE, for
example. Meters could also be added to PDUs in the IT space to be able to track
energy consumption at a department level for charge backs within the organization.

De p lo y in fra s tru c tu re m a n a g e m e n t s o ftwa re wh e re a b s e n t

Power meter-sourced data needs to be collected by infrastructure management
software tools to calculate and ultimately present the total energy consumption met-
ric to stakeholders. So, wherever there is a gap in data collection, monitoring soft-
ware should be deployed. However, procuring and deploying software systems
may not be feasible. In these cases, energy consumption must be estimated and
recorded manually.

For example, in shared use facilities, energy data for electrical and mechanical sys-
tems upstream of the IT room may have to be estimated by multiplying the IT en-
ergy use (from UPS output power meter) times an estimated PUE. This would ac-
count for the IT’s share of the cooling and electrical distribution’s power losses
yielding a more complete accounting of energy use. The site’s Facilities team may
or may not have a BMS and if they do, it may not be practical for the IT Ops team to
get that data from Facilities.

The following describes the primary types of data each software tool collects/moni-
tors. Note, that tools that embrace standard open communication protocols and
APIs are able to share data with each other.

DCIM – for UPS & rPDU power meters and IT whitespace infrastructure visibility at
on-premise data centers and distributed IT sites

3 Steps to Calculate Total Enterprise IT Energy Consumption Using DCIM

Schneider Electric – Energy Management Research Center White Paper 100 Version 1 7

ITAM – for estimating energy use of non-racked IT, WFH, and end user devices

BMS – for HVAC system energy of larger buildings; otherwise rely on PUE or esti-
mation using data supplied by facilities team

EPMS – for more granular PDU and branch circuit power meter data collection

En s u re a s s e t p o we r d e p e n d e n c ie s a re a c c u ra te ly d e fin e d
For tools like DCIM to be able to aggregate power and energy consumption accu-
rately in data centers and distributed IT sites, it is important that equipment power
dependencies are understood and represented correctly in the software tool. You
want to avoid, for example, double counting power by including the power reading
from both a UPS and its downstream rack PDU which is feeding the servers.

With metric data sources identified and metering and infrastructure management
software in place – and/or plans in place to use estimations or billing data – the next
step is aggregating the data and reporting the metrics using DCIM.

Step 3: Ch o o s in g s o ftwa re to u s e fo r d a ta a g g re g a tio n & d a s h b o a rd in g

Use DCIM to Modern BMS, EPMS, DCIM and ITAM software platforms all tend to be designed for

aggregate data data sharing and system integration with other platforms. However, we believe that
modern DCIM tools are best positioned to provide CIOs and their IT operations
and report teams with environmental sustainability metric tracking and reporting as they are
total energy beginning to address these needs out of the box. There are DCIM offers available
today that can aggregate data and report PUE, total energy consumption with
consumption breakdowns by subsystem, and even GHG emissions.

The ease of data collection into one dashboard depends on the specific attributes
of the software tool used and the capabilities of the vendor. Some more traditional
DCIM tools, for example, are architected for single sites, lacking the out-of-box abil-
ity to give an aggregate view of resources and status across multiple, distributed
sites.

For environmental metrics to be meaningful, as said before, it is important the soft-

ware tool used to aggregate all the data can communicate with and normalize data
from all data sources. This ensures a more complete picture of environmental im-
pact. Software system attributes that help make this easier to accomplish include:

• Being designed for multi-site portfolios with one login giving view to all sites’
data at a site-by-site level or in aggregate
• Supporting standard, open communications protocols (e.g., SNMP, BACnet,
Modbus, etc.)
• Providing APIs/web services for data sharing and integration with other 3rd
party software tools, devices, databases, and files
• Offering customizable dashboards and reporting that accommodates data
sources from outside the software tool providing the dashboard
• Offering additional custom services to build bespoke dashboards, reports,
connections to meet a variety of customer-specific needs*

*Figures 2 and 3 show examples of custom dashboards made to show energy con-
sumption, cost, carbon emissions, PUE, power losses, and capacity utilization.

3 Steps to Calculate Total Enterprise IT Energy Consumption Using DCIM

Schneider Electric – Energy Management Research Center White Paper 100 Version 1 8

Figure 2
This screenshot shows an ex-
ample custom dashboard from
Schneider Electric’s IT Advisor
planning and modeling soft-
ware that shows energy con-
sumption, cost, and carbon
emissions for various infrastruc-
ture systems of a data center.

Figure 3
This screenshot shows an
example custom dashboard
from Schneider Electric’s IT
Advisor planning and mod-
eling software that shows
the current, real-time PUE,
power losses by subsystem
and capacity utilization.

This reporting gives CIOs and their operations teams an ideal starting point to track
and report the impact of their IT estate on environmental sustainability. It will raise
awareness, provide a basis for setting goals, and identify opportunities to enrich the
measurement data in line with the CSO’s priorities. Benchmarking becomes possi-
ble to help the organization know where to improve, what to prioritize, and how to
show progress year over year.

Going beyond the 3 steps described in this paper, the final step would be to incor-
porate the IT estate energy consumption data with enterprise wide ESG reporting.
The CSO and their team can take this data and calculate Scope 2 GHG emissions.
Since emissions are driven by the local energy mix of the utility grid, providing the
CSO with location data along with the energy data would be helpful in generating a
more accurate emissions metric. Otherwise averages have to be used perhaps at a
country or region basis. See the Appendix for more information about calculating
GHG emissions.

For those enterprises who are using an environmental sustainability management

(ESM) platform, energy data from your on-premise data centers and edge compu-
ting IT sites can be sent from your DCIM software to the ESM tool via an API. This
provides enterprise sustainability teams with a consolidated view of the impact of
your on-premise IT portfolio.

3 Steps to Calculate Total Enterprise IT Energy Consumption Using DCIM

Schneider Electric – Energy Management Research Center White Paper 100 Version 1 9

Conclusion Measuring, reporting, and managing a company’s sustainability performance is

starting to drive employee retention, corporate reputation, and contribute to busi-
ness success. We recommend that CIOs who are beginning their sustainability jour-
ney start with measuring just one metric: total energy consumption. This fundamen-
tal metric shows the direct impact of IT operations and is used by CSOs to calculate
GHG emissions. Energy consumption is also good to start with because the data
sources and collection tools are readily available today, and in many cases, are
likely already in use by your IT operations team.

This paper provides a simple framework to measure and track IT energy consump-
tion across the entire enterprise IT portfolio. There are three steps:

Step 1: Assess the coverage of your IT estate

Step 2: Determine metric data sources and measurement tools
Step 3: Use data center infrastructure management (DCIM) to aggregate data and
report total energy consumption

Following the guidance in this paper creates a solid foundation to collect and report
additional metrics (e.g., water use, Scope 3 emissions like embodied carbon, and
so on) as the organization’s sustainability program matures.

About the authors

Robert Moore is Head of Strategy & Sustainability for the EcoStruxure IT Line
of Business at Schneider Electric. In his 12 years at Schneider Electric, he has
held various positions in Strategy and Finance working with commercial, inno-
vation and product teams. Robert is based in Ireland, is a Chartered Account
(ACA), Chartered Tax Advisor (CTA) and holds a Bachelor of Commerce from
the National University of Ireland, Galway.

Patrick Donovan is a Senior Research Analyst for the Energy Management Re-
search Center at Schneider Electric. He has over 27 years of experience devel-
oping and supporting critical power and cooling systems for Schneider Elec-
tric’s Secure Power Business unit including several award-winning power pro-
tection, efficiency, and availability solutions. An author of numerous white pa-
pers, industry articles, and technology assessments, Patrick's research on data
center physical infrastructure technologies and markets offers guidance and
advice on best practices for planning, designing, and operation of data center
facilities.

RATE THIS PAPER 

3 Steps to Calculate Total Enterprise IT Energy Consumption Using DCIM
Schneider Electric – Energy Management Research Center White Paper 100 Version 1 10

Guide to Environmental Sustainability Metrics for Data Centers

Resources
White Paper 67

Electrical Efficiency Modeling for Data Centers

White Paper 113

Environmental Sustainability Management (ESM) Software for Colocation Data

Centers
White Paper 231

Recommended Inventory for Data Center Scope 3 GHG Emissions Reporting

White Paper 53

Browse all
white papers
whitepapers.apc.com

Browse all
TradeOff Tools™
tools.apc.com

Note: Internet links can become obsolete over time. The referenced links were available at the time this
paper was written but may no longer be available now.

Contact us © 2023 Schneider Electric. All rights reserved.

For feedback and comments about the content of this white paper:
Schneider Electric Energy Management Research Center
[email protected]

If you are a customer and have questions specific to your data center project:
Contact your Schneider Electric representative at
www.apc.com/support/contact/index.cfm

3 Steps to Calculate Total Enterprise IT Energy Consumption Using DCIM

Schneider Electric – Energy Management Research Center White Paper 100 Version 1 11

Appendix En e rg y-re la te d GHG e m is s io n s

Greenhouse gas emissions are calculated from the energy consumption data. For
enterprise-owned infrastructure, DCIM 2 and Environmental Sustainability Manage-
ment (ESM) platforms can do this calculation automatically. And cloud and coloca-
tion providers will typically give this data to their customers and tenants through a
report or dashboard. The energy-related GHG emissions calculation is the energy
consumed times an emission factor which yields the total amount of emissions, usu-
ally measured in metric tons, resulting from a given amount of energy consumption.
The emission factor (in units of tCO2/MWh) is a coefficient that represents the
amount of GHG that is released into the atmosphere per MWh of electricity gener-
ated. Since the energy fuel mix (e.g., coal, oil, solar) varies between power plants,
each local grid has its own unique emission factor. Those emission factors also vary
over time as the generation mix itself varies.

Therefore, to get a more accurate energy-related GHG emissions value, it is im-

portant to know the local emission factor. In the absence of that, national, regional,
or state/province averages can be used. Some DCIM and ESM platforms have built-
in emission factor libraries to automate and simplify GHG emissions calculations.
Using management software to aggregate the data is discussed later in the paper.

We recommend that CIOs and their teams who are just getting started focus on lo-
cation-based emissions. Market-based emissions, if tracked, would be done at the
enterprise-wide level.

Finally, the reader will have heard emissions being categorized as Scope 1, 2, and
3. While this paper is largely concerned with Scope 2, we provide a summary of the
3 scopes from Schneider Electric White Paper 53:

• Scope 1 - Direct GHG emissions: All direct emissions within the operational
control of an organization (e.g., operating diesel generators onsite).
• Scope 2 - Indirect energy GHG emissions: Indirect emissions generated from
purchased or acquired electricity, heat, steam, or cooling.
• Scope 3 - Other indirect GHG emissions: All other indirect emissions from
sources such as business travel, waste management, manufacture of the
products you buy across the value chain.

2
Note, many of today’s modern DCIM platforms have this capability. Sometimes it is a standard feature,
other times it is provided through a custom engineering service.

3 Steps to Calculate Total Enterprise IT Energy Consumption Using DCIM