[PPT] - GHG Protocol Product Standard ICT Sector Guidance St akeholder PowerPoint Presentation

SLIDE 1

w w w .gh ghgpr gprot

t ocol
l.or
rg

1

GHG Protocol Product Standard ICT Sector Guidance St akeholder Advisory Group ( SAG) present at ion

Ma r c h 2 0 1 2

SLIDE 2

w w w .gh ghgpr gprot

t ocol
l.or
rg

2

I CT Sector Guidance Stakeholder Advisory Group Webinar – Seeking Public Comments on the Draft Guidance March 2012

GHG Protocol Product Accounting and Reporting Standard

SLIDE 3

w w w .gh ghgpr gprot

t ocol
l.or
rg

3

Agenda

I ntroduction to the I CT Sector Guidance

(15 mins) –

Background to ICT Sector guidance

–

Overview of the process (including timeline)

Summary of I ntroduction Chapter: General Guidance and infrastructure

summaries (hardware, software, networks and data centers) (15 mins) –

Summary of guidance, Any key points for discussion, Question and answer

Summary of Guidance for TNS and DMS chapters

(15 mins) –

Summary of guidance, Any key points for discussion, Question and answer

Key guidance for Cloud chapter

(15 mins) –

Summary of guidance, Any key points for discussion, Question and answer

Key guidance for Transportation Substitution chapter

(15 mins) –

Summary of guidance, Any key points for discussion, Question and answer

General Question & Answer

(15 mins) –

Any other questions, comments or discussion points

SLIDE 4

w w w .gh ghgpr gprot

t ocol
l.or
rg

4

Governance Structure

Convening Secretariat

WRI, WBCSD, GeSI, Carbon Trust

Steering Committee

Founding companies, Carbon Trust, Advocacy Groups , Academia, NGOs

Technical Working Group

(Practitioners)

Companies
OEMs
Service Providers
Consultancies
Academia

Stakeholder Advisory Group

(open to all)

Environmental advocacies
Industry analysts
Governments
ICT customers (corporates)
Developing Countries

subgroup subgroup subgroup

SLIDE 5

w w w .gh ghgpr gprot

t ocol
l.or
rg

5

Initiative jointly convened by:

– WRI (World Resources Institute) – WBCSD (World Business Council for Sustainable Development) – GeSI (Global e-Sustainability Initiative) – Carbon Trust

Steering Committee:

– EU Commission, MIT, ITU-T, CDP

, Gartner, ICT Companies

Companies participating in the TWG:

– Alcatel Lucent, BT, Capgemini, Cisco, Deutsche Telekom, EMC, Ericsson,

Fujitsu, HP , Microsoft, NetApp

TWG also has invited experts
Stakeholder Advisory Group (SAG)

– Over 200 participants, 50 companies and 45 countries

Carbon Trust are acting as facilitator and coordinator

Who is involved in the I CT Sector Guidance

SLIDE 6

w w w .gh ghgpr gprot

t ocol
l.or
rg

6

Chapter Structure

Transport Substitution Desktop Managed Services Telecoms Network Services Cloud and Data Center Services References Glossary Introduction & General Principles ‘Infrastructure summary’

Hardware
Software
Networks
Data Centers

Services Chapters Technical Support Chapters

Hardware Software (Energy Used by) Networks Data Center (Standalone)

I ntroduction Appendices

SLIDE 7

w w w .gh ghgpr gprot

t ocol
l.or
rg

7

I ntroduction:

–

Andie Stephens andie.stephens@carbontrust.co.uk

Telecommunications Network Services:

–

Glyn Stacey glyn.stacey@bt.com,

–

Tom Okrasinski tom.okrasinski@alcatel-lucent.com

Desktop Managed Services:

–

Andy Lewis Andrew.Lewis@uk.fujitsu.com

Transportation Substitution:

–

Darrel Stickler darrel.stickler@cisco.com

Cloud & Data Center Services:

–

Andrew Armstrong andrew.armstrong@WSPGroup.com

Technical Support Chapters:

–

Hardware: Elsa Olivetti elsao@mit.edu, Tom Okrasinski tom.okrasinski@alcatel-lucent.com

–

Software: Dan Williams v-dawill@microsoft.com

–

Data Centers: Amaya Sourarez amayaso2@microsoft.com

Contacts for Lead Authors

SLIDE 8

w w w .gh ghgpr gprot

t ocol
l.or
rg

8 Month Week 1 Week 2 Week 3 Week 4 Week 5

March

27 28 29 1 2 5 6 7 8 9 12 13 14 15 16 19 20 21 22 23 26 27 28 29 30

April

2 3 4 5 6 9 10 11 12 13 16 17 18 19 20 23 24 25 26 27

May

30 1 2 3 4 7 8 9 10 11 14 15 16 17 18 21 22 23 24 25 28 29 30 31 1

Public comment phase

India workshop

Public comment phase

Full draft* released for public comment Publication of finalized ICT Sector Guidance and launch event

Key:

Workshop Public comment phase * Full draft released to public domain and available for companies to begin using

Public comment phase

Draft released for public comment China workshop Webinars Webinar Webinar

Timeline

SLIDE 9

w w w .gh ghgpr gprot

t ocol
l.or
rg

9

Summary of I ntroduction Chapter

SLIDE 10

w w w .gh ghgpr gprot

t ocol
l.or
rg

10 10

Context for ICT

– long and complex global supply chains – Complex ICT services – Significant use phase – Current best practice

Overview of chapter structure
Relation to other standards
Key principles (relating to the Product Standard)
Boundary Setting (what to include and exclude)
Allocation
Assurance
Reporting
“Infrastructure Summaries”

– Hardware; Networks; Software; Data Centers

Summary of I ntroduction Chapter

SLIDE 11

w w w .gh ghgpr gprot

t ocol
l.or
rg

11 11

Refer to Hardware Chapter for detail of approaches

I nfrastructure summary: Hardware (embodied) – assessment approaches

Asse ssessm ssm ent Approach

LCA using primary data sources (PDS LCA) LCA using primary and secondary data sources (PSDS LCA) LCA estimation using component characterization LCA estimation using hardware parameterization LCA estimation using life cycle stage ratio profiles Estimation using environmentally extended input-output (EEI O)

SLIDE 12

w w w .gh ghgpr gprot

t ocol
l.or
rg

12 12

Power state

– E.g. full load, typical load, low power, stand-by

Power measurement and Allocation
Use profiles
Overhead energy allocation

– E.g. HVAC

Conversion from kWh to CO2e

I nfrastructure summary: Hardware – use stage emissions

SLIDE 13

w w w .gh ghgpr gprot

t ocol
l.or
rg

13 13

1. Divide the network into separate sub-networks by technology and region
2. Assess each sub-network to determine its energy intensity factor

– By provisioned bandwidth – By mean traffic – By data transferred – By number of ports used – Or by duration of voice call – [refer to TNS chapter]

3. Calculate the portion of the network energy used by the product being

assessed, using appropriate allocation metric

4. Calculate the GHG emissions using the appropriate grid emission factor(s)

I nfrastructure summary: Networks

SLIDE 14

w w w .gh ghgpr gprot

t ocol
l.or
rg

14 14

Life cycle assessment

– Material Acquisition and Pre-processing

Generally not relevant

– Production [= Software Development]

Heating, lighting etc. of buildings for software development
Energy used by equipment for development and testing
Consumables (e.g. office supplies)
Business travel

– Distribution and Storage

Electronic distribution
Physical media

– Use

[see also the Software Chapter]

– End of Life

Relevant to physical media

I nfrastructure summary: Software

SLIDE 15

w w w .gh ghgpr gprot

t ocol
l.or
rg

15 15

Data Center Chapter covers life cycle approach for Data Centers

– Aligned with The Green Grid approach

Cloud Chapter also covers Data Center Services
Overhead energy allocation

– Measurement – Use of PUE to allocate to individual hardware and services

I nfrastructure summary: Data Centers

SLIDE 16

w w w .gh ghgpr gprot

t ocol
l.or
rg

16 16

Summary of Guidance for TNS and DMS chapters

SLIDE 17

w w w .gh ghgpr gprot

t ocol
l.or
rg

17 17

Telecommunications Network Services (TNS) chapter

SLIDE 18

w w w .gh ghgpr gprot

t ocol
l.or
rg

18 18

Telecommunications Network Service (TNS) Elements in GHG Emissions Assessment

Customer Domain Service Platform Operational Activities

In-use GHG emissions associated with ICT end-user / customer premises equipment c1 b1 a1 b2 a2 c2

Embodied Use

GHG Emissions LCA Phase

In-use GHG emissions associated with ICT network and service platform supporting / connecting (but not in) customer domain In-use GHG emissions associated with labor and non-ICT infrastructure supporting Customer Domain and Service Platform equipment Embodied GHG emissions associated with Customer Domain equipment / infrastructure Embodied GHG emissions associated with Service Platform equipment / infrastructure Embodied GHG emissions associated with Operational / non-ICT capital infrastructure

Equipment / network use
Repairs / maintenance

“Use” Includes:

Raw materials acquisition &

pre-processing

Production
Product distribution / retail
Installation
End-of-life treatment

“Embodied” Includes:

Telecommunications Network Services (TNS) Guide

GH GHG G em em issions el elem em en ent s

SLIDE 19

w w w .gh ghgpr gprot

t ocol
l.or
rg

19 19

Approaches for assessing GHG emissions of Cust om er D

Dom ain

(Listed in decreasing order of accuracy)

TNS Element Use “a1” Embodied “a2”

Customer domain

1. Detailed use stage measurement:

directly measure power consumption of ICT equipment via physical power monitoring under specific operating conditions and usage profile

2. Use estimation: estimate typical energy

consumption of ICT equipment based on categorical equipment type, anticipated usage profile, and relevant country/region location of usage

3. LCA stage ratio profiles: estimation

based on percentage of use-stage GHG emissions for the total life cycle GHG emissions, based on historical LCA data

I CT Equipment:

1. Primary / Secondary Data Source LCA:

use combination of primary and secondary data sources to perform detailed LCA to determine the ICT equipment’s GHG emissions

2. LCA estimation: estimated GHG

assessment from techniques such as:

Components – estimation techniques

based on categorical component similarities

Equipment – estimation techniques

based on ICT equipment type parameterization

LCA stage ratio profiles: estimation

based on percentage of embodied-to- use stage life cycle GHG emissions

Environmentally extended input-
utput: estimation based on eco-

economic information

Telecommunications Network Services Guide

GHG Em issi ssions s Elem ent s s – Cust om er Dom ain

SLIDE 20

w w w .gh ghgpr gprot

t ocol
l.or
rg

20 20

Approaches for assessing GHG emissions of Service

ce P Plat fo form

(Listed in decreasing order of accuracy)

TNS Element Use “b1” Embodied “b2”

Service platform

1. Bottom-up model: total service platform

energy consumption can be calculated “bottom-up” from an inventory of equipment

Coarse/ fine-grain models: a

combination of data requirements and energy modeling parameters to yield a more refined level of information for the use phase

2. Top-down model: apportionment of energy

to individual telecom services calculated with ratio of capacity used by customer or service and mean traffic being carried by the network.

1. Follow approaches as per Customer

Domain – Embodied phase

2. Screening estimation: where significance

is low, use existing LCA studies as proxies.

Telecommunications Network Services Guide

GH GHG G em issi ssions s elem em en ent s – Service ce P Plat fo form

SLIDE 21

w w w .gh ghgpr gprot

t ocol
l.or
rg

21 21

Approaches for assessing GHG emissions of Operat io

ional l Act iv ivit it ie ies

(Listed in decreasing order of accuracy)

TNS Element Use “c1” Embodied “c2”

Operational activities and non- I CT infrastructure

1. Primary Data Assessment - assess GHG

emissions from resources (people and equipment) involved in delivering the TNS service, their specific work assignments, and their time allocations.

2. Secondary Data Assessment - assess

GHG emissions from resources involved in delivering the TNS service by using costs and conversion factors from economic input / output tables and apportionment factors based on the type and extent of operational activities.

1. Primary / Secondary Data Source LCA –

use primary and secondary data in a detailed life cycle assessment (typically carried out by the owner / maintainer of the operational / non-ICT infrastructure)

2. Screening estimation – via screening

evaluation these activities in the embodied phase may be represented as either a simplified percentage of the total LCA emissions, e.g. 1% , or it may be excluded due to the relatively small impact (less than 1% )

Telecommunications Network Services Guide

GHG Em issi ssions s Elem ent s s – Operat io ional l Act iv ivit it ie ies

See appendices for examples

SLIDE 22

w w w .gh ghgpr gprot

t ocol
l.or
rg

22 22

Desktop Managed Services (DMS) chapter

SLIDE 23

w w w .gh ghgpr gprot

t ocol
l.or
rg

23 23

Very broad scope - good test of methodology but…….scope varies with

supplier (Gartner definition is broadly adhered to)

DMS chapter a ‘recipe book’ - dependent on several other chapters, e.g.

datacentre, hardware, networks

Common outsourcing service in many countries. e.g. European DMS market

(13 million desktops) worth $6bn for top 15 vendors (Gartner 2010)

Increasing demand from customers for footprinting DMS, e.g. to compare

with the pre-outsourced estate, to allow accurate monitoring against C02 reduction targets

Many component parts, so subject to constant change (e.g. client device

refresh) so high footprinting overhead (esp for long term contracts)

Desktop Managed Services (DMS) Overview

SLIDE 24

w w w .gh ghgpr gprot

t ocol
l.or
rg

24 24

Desktop Managed Services (DMS) Overview

SLIDE 25

w w w .gh ghgpr gprot

t ocol
l.or
rg

25 25

Desktop Managed Services (DMS) Components

SLIDE 26

w w w .gh ghgpr gprot

t ocol
l.or
rg

26 26

Key Guidance for Cloud Services

SLIDE 27

w w w .gh ghgpr gprot

t ocol
l.or
rg

27 27

A range of business and consumer applications are increasingly provided from

cloud architecture:

E-mail, calendar, document and other business applications.
Consumer photo, video and music and other data storage applications.
Search, social networking and database applications.
Data center services are also increasingly leased by organizations to support

the delivery of their IT services.

This chapter provides guidance on how to quantify the energy and GHG

emissions associated with these services.

The guidance is written from the perspective of a “user” of cloud and data

center services and aims to provide standard and repeatable methods in order to facilitate a better understanding of the energy and GHG impacts of alternative ICT service delivery solutions.

Key Guidance for Cloud Services - Purpose

SLIDE 28

w w w .gh ghgpr gprot

t ocol
l.or
rg

28 28

Wholesale leases: lessee may lease the entire data center,
perate and maintain data center infrastructure and pay

utility bills

Colocation leases: lessee may buy rack-space; the owner
perates and maintains all critical data center infrastructure
I nfrastructure as a Service (I aaS): Access to a virtual

server and a storage pool with full access to the server’s

perating system and to the applications that are running on

it.

Platform as a Service (PaaS): Platforms for running web-

based applications. PaaS clouds do not provide access to the underlying operating system.

Software as a Service (SaaS): Enables commercial SaaS

applications in a hosted environment.

Key Guidance for Cloud - Applicable Services

Increasing User Control of Energy Using Equipment

SLIDE 29

w w w .gh ghgpr gprot

t ocol
l.or
rg

29 29

Cloud Services - Scope of Equipment Covered

Hosting and fulfilment: servers,

servers, switches and routers that store and transmit information, critical systems and associated data center infrastructure including HVAC systems required for server cooling.

I nternet transfer: includes wired or

wireless network infrastructure for transmitting information from cloud infrastructure to end users

User access: End-use devices (or

“client”); including computers, smart phones and other devices used for accessing cloud services

SLIDE 30

w w w .gh ghgpr gprot

t ocol
l.or
rg

30 30

Cloud Services - Functional Units

Unit Example Notes

Cloud Per-user licence (or user group) E-mail, calendar, document and

ther business applications

High data storage requirements and high user access Per-unit of storage capacity Consumer photo, video and music and other data storage applications High data storage requirements and low user access Per-transaction Search; social networking and database applications Low data storage requirements; high user access Data Center Allocated Power Kilowatts (kW) Common denominator of capacity and ease of calculation Physical capacity Number of racks or floor area (m2 or sqft) Easily converted back to total kilowatts

The method of procurement is useful starting point for defining the functional unit and provides users of cloud and data center services the ability to compare alternative service delivery options

SLIDE 31

w w w .gh ghgpr gprot

t ocol
l.or
rg

31 31

Cloud Services - Method for Allocating Equipment

1. Establish the application demand in

relation to physical resource requirements, such as Iops (input-

utput operations per second) or

WebAPIs (i.e. number of web request/responses) over a specified period of time

2. Match application demand with IT

hardware requirements for core hosting and fulfilment processes

3. Estimate data center, network

infrastructure and end use device energy consumption

Figure below: Steps for establishing an inventory of equipment associated with cloud services. Adapted from Accenture’s Cloud Reference Model Framework – April 2010

SLIDE 32

w w w .gh ghgpr gprot

t ocol
l.or
rg

32 32

Primary Data Requirements Secondary Data

Number of users and use profiles
Licensing or service level agreements
Transactions
Data center location
Server count
Network Link equipment count
Device utilization
Power consumption per type of IT

hardware

Data center Power Usage Effectiveness

(PUE) If primary data is not available, secondary data may be used to develop assumptions for processes that are not under the ownership or control of the cloud or data center service provider. These might include:

Internet transfer: the number of

internet hops and equipment required for data transfer to end users.

Embodied energy for hardware:

LCA estimates of embodied energy on a per-server basis.

SLIDE 33

w w w .gh ghgpr gprot

t ocol
l.or
rg

33 33

Calculating I nventory Results

Data Center Emissions Network Emissions End-Use-Device Use Life Cycle Emissions TOTAL EMISSIONS TOTAL EMISSIONS User License Count Transaction Count Storage Capacity (GB) Allocated Power (kW) Physical capacity (m2) SERVICE DELIVERY EFFICIENCY

+ + +

r
r
r
r

SLIDE 34

w w w .gh ghgpr gprot

t ocol
l.or
rg

34 34

Specific I ssues for SAG to Consider

The purpose of the guidance is to stimulate a transition to more sustainable data center and cloud infrastructure and we need the guidance to be useable and widely adopted by both providers and users of cloud and data center services:

Are the functional units proposed useful and practical for industry decision-

makers?

What is the appropriate balance between secondary and primary data?
What level of transparency should be required?

Case Study material would be welcome for inclusion in the guidance

SLIDE 35

w w w .gh ghgpr gprot

t ocol
l.or
rg

35 35

Key guidance for Transportation Substitution chapter

SLIDE 36

w w w .gh ghgpr gprot

t ocol
l.or
rg

36 36

Lead chapter addressing “enabling effect” of I CT (“The 98% ”)
Compares business as usual (BAU) and I CT solution
All emissions considered

– Direct GHG emissions

Life-cycle emissions of the ICT solution

– Enabling effects (avoided emissions)

Immediate enabling effect
Longer-term enabling effect

– Rebound effects

Immediate rebound effect
Longer-term rebound effect

Transport Substitution Chapter

SLIDE 37

w w w .gh ghgpr gprot

t ocol
l.or
rg

37 37

transpor- tation 34% buildings residential 21% buildings commercial 19% industrial 26% passenger transport 26% freight 11% buildings household 22% buildings services 14% manufac- turing 26%

Source: U.S. Energy Information Agency (EIA) Emissions of Greenhouse Gases Report Table 7 (U.S., 2009)

Global GHG Emissions – Energy

Source: International Energy Agency (IEA) Energy Use in the New Millennium Figure 2.3 and p. 24 description (IEA14, 2004)

EIA (U.S.) IEA14

Transport Substitution (“Total Available Carbon Market”)

SLIDE 38

w w w .gh ghgpr gprot

t ocol
l.or
rg

38 38

Chapter scope: ICT use by people to avoid travel

(1) Avoid business travel

Local Attendees Travelling Attendees Site A Site B Site A Site B

kWh kWh

Local Attendees Local Attendees

kWh

ICT Solution: Remote Collaboration Meeting BAU: Business Travel Meeting

HVAC HVAC

Telecom Network Services (Chapter X)

Traversal Server Call Control Mgmt. Server Recording/ Streaming Multipoint Server

WAN/Cloud

audio video desktop Traversal Server Call Control Mgmt. Server Recording/ Streaming Multipoint Server Traversal Server Call Control Mgmt. Server Recording/ Streaming Multipoint Server Traversal Server Call Control Mgmt.Server Recording/ Streaming Multipoint Server

Transport Substitution Chapter

SLIDE 39

w w w .gh ghgpr gprot

t ocol
l.or
rg

39 39

Sum m ar ary of

f em ission
ns cat

at egor

ry fac

act or

rs fo

for avoiding busi siness ss t ravel

GH GHG G em issio ission cat eg egory Avoid id busin siness ss t ravel Descript pt ion Avoid id busin siness ss t ravel I nc nclud ude / Exclud ude

Direct

ICT solution end points
enterprise ICT solution infrastructure
enterprise network infrastructure (shared )
associated HVAC allocation
service provider power consumption with all life cycle and non-

use-mode burdens

include
include
include
include
include

I mmediate enabling

reduced employee transport

a. air b. boat c. rail d. road (private vehicle, bus)

include.
travel to/from airport (or rail station), ground transportation,

hotel

Exclude as conservative

Longer-term enabling

reduced operating emissions at airport facilities
reduced future infrastructure (airplanes and airports)
reduced personal air travel due to heightened environmental

awareness

reduced travel from loss of frequent flier miles
exclude as conservative
exclude as conservative
exclude as conservative. Also listed as

possible rebound effect.

may offset longer-term rebound effect

(increased personal air travel)

I mmediate rebound

increased ICT solution use for business reasons other than

avoiding travel

use of alternative means of business travel (rail, boat, road)
include
include

Longer-term rebound

increased employee personal travel, especially by air
exclude.

SLIDE 40

w w w .gh ghgpr gprot

t ocol
l.or
rg

40 40

Chapter scope: ICT use by people to avoid travel (1) Avoid business travel

(2) Avoid employee commuting (telework)

Commuting Employees

kWh

ICT Solution: Teleworking BAU: Employee Commuting

HVAC HVAC

Company Campus

intra- campus travel

HVAC HVAC HVAC

kWh kWh Telecom Network Services (Chapter X)

Company Campus

employee home remote telework location employee home kWh kWh

Voice | Video |Apps | Data | TelePresence

Cloud

Cellular/ Wi-Fi secure mobility for phone/tablet secure mobility

n laptop

WiFi extensibility to remote laptop basic virtual office with desktop HD video full virtual office with telepresence

Transport Substitution Chapter

SLIDE 41

w w w .gh ghgpr gprot

t ocol
l.or
rg

41 41

Sum m ar ary of

f em issi

ssions s cat at egor

ry fac

act or

rs fo

for avoiding em ployee co com m ut ing ( t elew ork)

GH GHG G em issio ission cat eg egory Avoid em em ployee ee com m ut ing Descript pt ion Avoid em em ployee ee com m ut ing Disp isposit sit io ion

Direct

ICT solution remote telework end points, wireless, routing, access
Enterprise ICT solution infrastructure
Enterprise network infrastructure (shared)
Associated HVAC allocation
Service provider power consumption with all life cycle and non-

use-mode burdens

Include (partial).
Include
Exclude. Internal network traffic is the same for

BAU and ICT solution use case

Include
Include

I mmediate enabling

Reduced employee commuting
Boat/ferry; rail; road
Reduced HVAC load at company campus
Include. Most commuting is by personally owned

vehicle (POV)

Include (paired with inclusion of home HVAC

rebound effect)

Longer-term enabling

Reduced future, company real estate portfolio (company buildings;

redeploy existing, company, real-estate portfolio to third parties

Reduced operating emissions at company office facilities (building

electricity)

Reduced public transit service due to reduced passenger load
Improved passenger transport and freight logistics from reduced

road congestion

Reduced future infrastructure (roads and public transit))
Exclude as conservative
Include if paired with reductions in real estate

portfolio (building size, building count)

Exclude. Public transit is less sensitive to small

changes in demand.

Exclude as conservative.
Exclude as conservative

I mmediate rebound

Increased home energy use for HVAC
Increased use of POV during business day
Include
Exclude for basic analysis.

Longer-term rebound

Increased GHG emissions from urban sprawl (since proximity to
ffice less important)
Increased GHG emissions from construction and operation of

larger homes (for home-office space)

Exclude. Potentially offset by “live local” option

now available.

Exclude. Monitoring and further study

suggested.

SLIDE 42

w w w .gh ghgpr gprot

t ocol
l.or
rg

42 42

General Question & Answer

SLIDE 43

w w w .gh ghgpr gprot

t ocol
l.or
rg

43 43

Technical Support Chapters

SLIDE 44

w w w .gh ghgpr gprot

t ocol
l.or
rg

44 44

1. Hardware 2. Software (Energy Used by) 3. Data Centers (Stand alone)

Technical Support Chapters

SLIDE 45

w w w .gh ghgpr gprot

t ocol
l.or
rg

45 45

Provides guidance for ICT practitioners in calculating

GHG emissions associated with ICT Hardware

Follows the GHG Protocol Product Standard and forms an

essential building block to the ICT Sector Guidance chapters

Offers methodology approach and options to assess GHG

emissions over the life cycle of an ICT hardware asset

Provides a means of understanding emissions sources and

prioritizing them for reduction

Aligns with ITU-T and ETSI GHG measurement and reporting

standards and with iNEMI and MIT’s PAIA ICT Benchmarking Partnership developments

Methodologies divided into life cycle assessment and

estimation approaches

Guide follows GHG Protocol Standard:

– Scope, boundary setting, functional unit definition, etc. – Calculating LCA inventory results - methodology, approach

hierarchy, references, examples

– Appendix - analysis of a case study (wireless router)

Life Cycle Stages for an ICT Hardware Asset

I CT Hardware Guide - Obj ec

ect ives es

SLIDE 46

w w w .gh ghgpr gprot

t ocol
l.or
rg

46 46

Product Components

Flat rolled metals (Fe,

Al, Cu)

Other metals (Zn, Sn, Ag, Au,

Rare Earth Metals, etc.)

Silicon wafer (processed)
Plastics (~10 types)
Proprietary goods
…

Transport ICT Component Manufacturing ICT Subassembly Manufacturing ICT Product Assembly & Test Transport ICT Product Packing/Storage Transport Transport Transport ICT Product Installation ICT Product Retailing Transport ICT Product Use ICT Product Dismantling ICT Component Shredding Landfill ICT Product Servicing Transport Transport Transport Transport Transport To Front-end Processes Incineration / Recycling ICT Product De-installation Waste

Material Acquisition & Preprocessing Production Distribution & Retail Use End-of-Life

Material Acquisition & Preprocessing of Product Components

Transport

Transport Transport

I CT Hardware Guide - Process

ess Map ap

SLIDE 47

w w w .gh ghgpr gprot

t ocol
l.or
rg

47 47

Primary Data Guidance
Site-specific process data - includes direct emission measurements and/or process activity data
Site-average data – representative averages from entities operating equivalent processes
Secondary Data Guidance
Generic process data – obtained from sources other than direct measurement of original sources

(e.g. regional statistical info or averages from more generalized processes  good for confidential ICT processes)

Process data from literature studies / expert estimates
Environmentally extended input-output data

If applicable data is not available, the following data sources can be used to bridge data gaps:

Extrapolated data
Proxy data

Data quality should match the purpose of specific ICT hardware GHG emissions assessment. Use the most representative, reliable and highest quality data available relative to the analysis being performed.

I CT Hardware Guide - Dat

at a a Qualit y

SLIDE 48

w w w .gh ghgpr gprot

t ocol
l.or
rg

48 48

Assessing GHG Emissions of I CT Hardware

Approach Key Capabilities Key Drawbacks

LCA using primary data sources (PDS LCA)

In-depth assessment of full life cycle impact
Highest degree of accuracy and least amount
f uncertainty
Preferred for new product / technology

analysis; comparative evaluation; detailed understanding of one or more contributions from a specific LC stage process

Large amount of info, time, and expertise

needed

Requires primary data – which may not be

available for new technologies, materials, and energy efficiency features

LCA using primary & secondary data sources (PSDS LCA)

Same as above, but with lesser accuracy and

higher uncertainty

Large amount of info, time, and expertise

needed

Requires primary & secondary data – which

may not be available for new technologies, materials and energy efficiency features

I CT Hardware Guide -GHG Em issi

ssions s Assessm ssessm en ent - LCA A Appr pproaches

SLIDE 49

w w w .gh ghgpr gprot

t ocol
l.or
rg

49 49

I CT Hardware Guide - Dat

at a Q a Qual alit y

Assessing GHG emissions of I CT Hardware

Approach Key Capabilities Key Drawbacks

LCA estimation using component characterization

Uses commonality amongst ICT

components (e.g. materials, processes, mfg locations)

When characterization is based on

primary data, approach may offer sufficient data quality

Uniqueness of product may go beyond validity of

estimation algorithms

Practitioner needs to be aware of applicability
Can capture specificity of component characteristics

but not of processes used throughout life cycle

Trend shifts may not be visible or identified

LCA estimation using hardware parameterization

Uses modularity and commonality

amongst ICT equipment

When parameterization is based on

primary data, approach may offer sufficient data quality

Similar to above

LCA estimation using life cycle stage ratio profiles

Uses commonality amongst ICT

equipment and associated LCA stage ratio profiles

Higher level of uncertainty – use only for rough

estimates / screening evaluations

User must be aware of typical parameters for which

ratio data was developed

Other DBs similar to above

Estimation using environmentally extended input-

utput (EEI O)
Uses input-output data (e.g. financial

data) from targeted industry sectors

Provides a high level estimation

based on key parameters (e.g. energy or material flow)

EEIO tables limited to certain regions and industry

sectors

EEIO tables may not be up to date with ICT’s

newest technologies / materials

SLIDE 50

w w w .gh ghgpr gprot

t ocol
l.or
rg

50 50

Software

Aim: Allocate specific software tasks or services a portion of the total power

consumption of a device.

The steps involved to measure the power consumption of devices under

specific software loads are described.

Chapter describes the scientific methods to undertake benchmarking to

allocate device power consumption

The chapter provides basic to advanced methods to use which correlates with

result uncertainty levels.

Aimed at a higher technical ability level (both computer and device)
Output is not device independent.

SLIDE 51

w w w .gh ghgpr gprot

t ocol
l.or
rg

51 51

Software Chapter Overview

1. Measurement Methods

–

Ranges from secondary data to device component measurement

2. OS Measurement

–

This chapter separates OS from Application power consumption. This section describes how to create the OS baseline from which to measure application measurement.

3. Application Measurement

–

Multiple methods to allocate device power to either one or a set of defined applications.

–

Includes a method to apportion calculated application power to transactions within the application.

4. Virtual Machine Measurement

–

Focused upon Sever devices – but can be applied to any.

–

Basic methods to apportion device power to defined virtual machines.

SLIDE 52

w w w .gh ghgpr gprot

t ocol
l.or
rg

52 52

Software Challenges

Complexity level – too high?
Lack of case studies

SLIDE 53

w w w .gh ghgpr gprot

t ocol
l.or
rg

53 53

Definition of a Data Center
Boundary definitions
Life Cycle Stages

– Production: construction of the data center

Deployment of IT equipment and site preparation
Use stage emissions

– Energy used directly by the ICT equipment. – Energy “overhead” for environmental control and other site-wide systems.

End of Life (demolition) stage

Data Centers

SLIDE 54

w w w .gh ghgpr gprot

t ocol
l.or
rg

54 54

Appendices - Additional Slides for TNS Chapter

SLIDE 55

w w w .gh ghgpr gprot

t ocol
l.or
rg

55 55

Telecommunications Network Services Guide

Cust st om er dom ain in use st age det aile iled calc lcula lat io ion

Example: a small chassis router has typical active power consumption of 150 W at a utilization rate of 24 hours per day and shared services as shown below. What is the apportioned use stage GHG emissions for Service “C” per annum? Step 1: calculate the router’s use-phase GHG emissions per annum: From measured power consumption data (e.g. OEM performance tests): Euse = 150 W × 8760 hrs/yr × 1 kwh/1000 W-hrs × 0.6 kg CO2e/kwh* Euse = 788 kg CO2e per annum

*GHG conversion factor for actual region of use

Step 2: for apportionment, apply proportion of energy consumption to assessed service (e.g. Service “C” at 20 Mbps: Euse-C = 788 kg CO2e per annum × 5% apportionment = 39.4 kg CO2e per annum

Service A: 400 Mbps Service B: 150 Mbps Service C: 20 Mbps Small Chassis Router (Power: 150 W) Service A: 105 W (70%) Service B: 38 W (25%) Service C: 7 W (5%)

SLIDE 56

w w w .gh ghgpr gprot

t ocol
l.or
rg

56 56

Telecommunications Network Services Guide

Cust st om er dom ain in em bo bodi died d st a t age est i t im at i t ion by by LCA rat i t io

Example: small chassis router and Service “C” (from previous slide). What is the apportioned embodied GHG emissions for Service “C” per annum? Step 1: calculate the router’s use-phase GHG emissions  from the previous slide: Euse = 788 kg CO2e per annum Step 2: estimate the router’s embodied-phase GHG emissions using historical LCA data showing the LCA ratio for use / embodied emissions for different equipment types (see table): Eemb : Euse  15% : 85% = 788 kg CO2e × (15% / 85%) = 139 kg CO2e per annum For Service “C”, the apportioned embodied stage GHG emissions is: Eemb-C = 139 kg CO2e × 5% = 7 kg CO2e per annum

Equipment Category Use / Embodied Phase LCA Ratio (Cu and Ce) Wireless Access Point 80% / 20%

Router, small chassis 85% / 15%

Telepresence system 90% / 10%

SLIDE 57

w w w .gh ghgpr gprot

t ocol
l.or
rg

57 57

Point-to-Point Tree / Star Mesh

Network Types

Telecom Network Services Guide

Service Plat form Use St age: Modeling a approach

Top-down
Bottom-up
Refined Bottom-up:
Coarse-grained
Fine-grained

Model Types

SLIDE 58

w w w .gh ghgpr gprot

t ocol
l.or
rg

58 58

Increasing Precision / Data Needs

Telecom Network Services Guide - Modeling Need

eeds s

Parameters Top- Down Bottom- Up Bottom-up

Coarse se

Bottom-up

Fi Fine

Overall network power

√ √ √ √

Total mean network traffic

√ √ √ √

Mean network traffic (“nth” service)

√ √ √ √

Ingoing traffic (“nth” service)

√

Outgoing traffic (“nth” service)

√

Number of devices

√ √

Subset of number of hop counts

√

Weighting for traffic thru “h“ hops

√

Weighting for “nth” service thru “h” hops

√

Network architecture

√

Class of equipment /equipment category

√ √

Total number of equipment categories

√ √

Mean traffic for “nth” service thru each equipment category

√ √

Total mean traffic for each equipment category

√ √

Mean power/footprint for “kth” equipment category

√ √

Power/footprint for “ith” device

√

Total mean traffic for “ith” service thru “ith” device

√

Total mean traffic for “ith” device

√

Network / Service based Hops based Equipment based Device based