Constantinos Vassilakis cvassilakisgrnetgr Greek Research and Technology Network Utrecht Netherlands 56 March 2012 GN3 Green Networking Advances in Environmental Policy and ID: 712192
Download Presentation The PPT/PDF document "Towards energy efficient Internet Servic..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Towards energy efficient Internet Service Providers – ECOnet Perspective
Constantinos Vassilakiscvassilakis@grnet.grGreek Research and Technology Network
Utrecht, Netherlands, 5-6 March 2012
GN3 Green Networking: Advances in Environmental Policy and PracticeSlide2
Outline
The ECONET ProjectEnergy consumption and energy efficiency demandDecomposing the Energy Consumption in the Wired Network
A Taxonomy of Undertaken
ApproachesECONET approach
Potential Impact on the Wired
Network
Utrecht, Netherlands,
5-6 March 2012
GN3 Green Networking: Advances in Environmental Policy and PracticeSlide3
GN3 Green Networking: Advances in Environmental Policy and Practice
Increasing the energy efficiency and the sustainable growth of our world is a global process where Telecommunications technologies (and the ICTs in general) play a key role.But to obtain optimum results the process should involve the “two faces of the same coin”:Green ICT – reducing the carbon footprint of ICTICT for Green – using ICT for reducing third party-wastes.ECONET is dealing with the first aspect Focused on short and medium time exploitation
The ECONET projectSlide4
GN3 Green Networking: Advances in Environmental Policy and Practice
Participant organisation name
Short name
Country
Consorzio Nazionale Interuniversitario per le
Telecomunicazioni
– UdR
at DIST University of Genoa (Coordinator)
CNIT
Italy
Mellanox
Technologies
MLXIsraelAlcatel LucentALUItalyLantiqLQDEGermanyEricsson Telecomunicazioni S.p.A.TEIItalyTelecom ItaliaTELITItalyGreek Research & Technology NetworkGRNETGreeceResearch and Academic Computer NetworkNASKPolandDublin City UniversityDCUIrelandVTT Technical Research CentreVTTFinlandWarsaw University of TechnologyWUTPolandNetVisorNVRHungaryEthernityETYIsraelLightCommLGTItalyInfoComINFOItaly
Manufacturers
Operators
Academic
/research centers
Small/Medium Enterprises (SMEs)
The Consortium
The ECONET projectSlide5
GN3 Green Networking: Advances in Environmental Policy and Practice
Goals: re-thinking and re-designing network equipment towards more energy-sustainable and eco-friendly technologies and perspectives.The overall idea is to introduce novel green network-specific paradigms and concepts enabling the reduction of energy requirements of wired network equipment by 50% in the short/mid-term (and by 80% in the long run) with respect to the business-as-usual scenario.
To this end, the main challenge is to design, develop and test novel technologies, integrated control criteria and mechanisms for network equipment
allowing energy saving by dynamically adapting the device capacities and consumptions to current traffic loads and user requirements.The ECONET projectSlide6
GN3 Green Networking: Advances in Environmental Policy and Practice
Energy consumption and energy efficiency demandThere are two main motivations that drive the quest for “green” ICT: the environmental one, which is related to the reduction of wastes, in order to impact on CO2 emission; the economical one, which stems from the reduction of operating costs (OPEX) of ICT services.
Gartner Group, Inc. (2007)
“The global information and communications technology (ICT) industry accounts for approximately
2%
of global carbon dioxide (CO
2) emissions, a figure equivalent to aviation.”
Note that the ICT sector raises much faster than
aviation
How
much
is
2% of CO2?Slide7
GN3 Green Networking: Advances in Environmental Policy and Practice
Energy consumption and energy efficiency demandThe figures refer to the whole corporate consumption. As such, they account for numerous sources, other than the operational absorption of the networking equipment (e.g., offices’ heating and lights). Notwithstanding, they give an idea of the general trend.Slide8
GN3 Green Networking: Advances in Environmental Policy and Practice
Energy consumption and energy efficiency demandElectrical energy consumption evolution and future trends for TELIT’s fixed network. Source: Telecom Italia
Source: C. Bianco, F
. Cucchietti, G. Griffa, ” Energy consumption trends in the Next GenerationAccess Network - a Telco
perspective, ”
IEEE INTELEC 2007
.Slide9
GN3 Green Networking: Advances in Environmental Policy and Practice
Decomposing the Energy ConsumptionThe Wired Network
Typical access, metro and core device density and energy requirements in today’s typical networks deployed by telcos
, and ensuing overall energy requirements of access and metro/core networks.Source: R. Bolla, R. Bruschi, F. Davoli, F. Cucchietti
, “Energy Efficiency in the Future Internet: A Survey of Existing Approaches and Trends in Energy-Aware Fixed Network Infrastructures,”
IEEE Communications Surveys & Tutorials,
vol. 13, no. 2, pp. 223-244, 2nd Qr. 2011.Slide10
GN3 Green Networking: Advances in Environmental Policy and Practice
Decomposing the Energy Consumption High-end Routers
Estimate of power consumption sources in a generic platform of high-end IP router. Source: R. Tucker, “Will optical replace electronic packet switching?”,
SPIE Newsroom, 2007.Slide11
GN3 Green Networking: Advances in Environmental Policy and Practice
Decomposing the Energy Consumption Is the energy consumption currently load-dependent?
Network engineers only speak about the capacity of a device or of a link interface…
…as a matter
of
fact, device and link are specifically designed to work at the maximum speed…Source: The ECONET Consortium, ”End-user requirements, technology specifications and benchmarking methodologies
,” Deliverable 2.1.
Power consumption in GRNET core routers (
24-hour period)
Daily traffic
profile of core GRNET network router (peering with
GEANT)Slide12
GN3 Green Networking: Advances in Environmental Policy and Practice
Decomposing the Energy Consumption Is the energy consumption currently load-dependent?Power Consumption of Cisco Catalyst 2970
Switch
Source: K. Christensen, P. Reviriego, B.
Nordman
, M. Bennett, M.
Mostowfi, J.A. Maestro, "IEEE 802.3az: the road to energy efficient ethernet," IEEE Communications Magazine, vol.48, no.11, pp.50-56, November 2010.
There is no significant difference in power consumption whether a port is running at 10 Mbps or 100 Mbps.
The
switch power consumption is increased by connecting a new link, even if there is no data being transmitted on this link.
The
difference in power consumption is quite low when a 1
Gbps
link is fully utilized compared to when it is zero utilized.Slide13
GN3 Green Networking: Advances in Environmental Policy and Practice
Decomposing the Energy Consumption Day & Night Traffic ProfilesPercentage
w.r.t. peak level. The profiles exhibit regular, daily cyclical traffic patterns with Internet traffic dropping at night and growing during the day.
Traffic load fluctuation at peering links for about 40 ISPs from USA and Europe
Source
: http://asert.arbornetworks.com/2009/08/what-europeans-do-at-night/Slide14
GN3 Green Networking: Advances in Environmental Policy and Practice
Decomposing the Energy Consumption Energy wastesNetworks and devices are lightly utilized.Often peak loads during rush hours are generally much lower than capacities of links and devices.
It is well known that the «overdimensioning» is the best design strategy for assuring QoS levels…Moreover, traffic loads follow well-known day & night fluctuations.
On the other hand, the energy requirements of network devices remain substantially flat according to their workload.Furthermore, networks are highly overprovisioned /redundant to assure service availability.Slide15
GN3 Green Networking: Advances in Environmental Policy and Practice
A Taxonomy of Undertaken ApproachesSource: R.
Bolla, R. Bruschi, F. Davoli, F. Cucchietti
, “Energy Efficiency in the Future Internet: A Survey of Existing Approaches and Trends in Energy-Aware Fixed Network Infrastructures,” IEEE Communications Surveys & Tutorials, vol. 13, no. 2, pp. 223-244, 2nd Qr. 2011.
The largest part of undertaken approaches regarding engineered improvements is funded on few base concepts, which have been generally inspired by energy-saving mechanisms and power management criteria that are already partially available in computing systems.Slide16
GN3 Green Networking: Advances in Environmental Policy and Practice
Re-engineering approaches aim at: introducing and designing more energy-efficient elements for network device architecturessuitably dimensioning and optimizing the internal organization of devicesreducing their intrinsic complexity levels.
A Taxonomy of
Undertaken ApproachesRe-engineeringSlide17
GN3 Green Networking: Advances in Environmental Policy and Practice
The dynamic adaptation of network/device resources is designed to modulate capacities of packet processing engines and of network interfaces, to meet actual traffic loads and requirements. This can be performed by using two power-aware capabilities, namely, dynamic voltage scaling and idle logic, which both allow the dynamic trade-off between packet service performance and power consumption.
A
Taxonomy of Undertaken ApproachesDynamic AdaptationSlide18
GN3 Green Networking: Advances in Environmental Policy and Practice
Standard operations
Idle logic
Power
scaling
Idle + power scaling
Wakeup
and sleeping
times
Increased
service
times
Wakeup
and sleeping + increased service timesA Taxonomy of Undertaken ApproachesDynamic AdaptationSlide19
First
version: Adaptive Link Rate proposed by Christensen and NordmanFinal Version: based on the “low power idle” concept, proposed by Intel.
Idea: transmit data at the maximum speed, and put the link to sleep when it is idle.
GN3 Green Networking: Advances in Environmental Policy and Practice
19
0
5
10
15
Link speed (Mb/sec
)
Power
(W)10100100010000A Taxonomy of Undertaken ApproachesDynamic Adaptation: Green Ethernet (IEEE 802.3 az)Tw and Ts for 10 Gb/s in IEEE Std 802.3az-2010 are 4.48 μs and 2.88 μs, respectivelyLPI can possibly be asynchronousUtrecht, Netherlands, 5-6 March 2012Slide20
In PC-based devices, the Advanced Configuration and Power Interface (ACPI) provides a standardized interface between the hardware and the software layers
.ACPI introduces two power saving mechanisms, which can be individually employed and tuned for each core:Power States (C-states)C0 is the active power stateC1 through Cn are
processor sleeping or idle states (where the processor consumes less power and dissipates less heat).Performance States (P-states)while in the C
0 state, ACPI allows the performance of the core to be tuned through P-state transitions.P-states allow to modify the operating energy point of a processor/core by altering the working frequency and/or voltage, or throttling the clock.
A
Taxonomy of Undertaken Approaches
Dynamic Adaptation:
SW routers & ACPI
GN3 Green Networking: Advances in Environmental Policy and PracticeSlide21
21
Source: R. Bolla, R. Bruschi, A. Ranieri, “Green Support for PC-based Software Router: Performance Evaluation and Modeling”, Proc. IEEE ICC 2009, Dresden, Germany, June 2009. Best Paper Award.
[MHz]
A
Taxonomy of Undertaken Approaches
Dynamic Adaptation:
SW routers & ACPI
GN3 Green Networking: Advances in Environmental Policy and PracticeSlide22
Sleeping/standby approaches are used to smartly and selectively drive unused network/device portions to low standby modes
, and to wake them up only if necessary. However, since today’s networks and related services and applications are designed to be continuously and always available, standby modes have to be explicitly supported with special techniques able to maintain the “network presence” of sleeping nodes/components.
A
Taxonomy of Undertaken ApproachesSleeping/Standby
GN3 Green Networking: Advances in Environmental Policy and PracticeSlide23
Scenario:
networked hosts (PCs, consumer electronics, etc.);Problem: when an end-host enters
standby mode, it freezes
all network services, and it is not able
to
maintain its network presence;Idea: introduce a Network Connection Proxy (NCP), which
is devoted
to
maintain
the network
presence
of sleeping hosts.Sleeping hostNCPInternetContinuous and full connectivityWakeup/sleepmessagesApplication-specific messagesZzzzz…I want to sleepSource: M. Allman, K. Christensen, B. Nordman, V. Paxson, “Enabling an Energy-Efficient Future Internet Through Selectively Connected End Systems,” Proc. ACM SIGCOMM HotNets, Atlanta, GA, Nov. 2007.A Taxonomy of Undertaken ApproachesSleeping/Standby: Proxying the Network PresenceGN3 Green Networking: Advances in Environmental Policy and PracticeSlide24
Scenario:
Core NetworksIdea: put links, interfaces and part of nodes (e.g., line-cards) to
sleepProblem: Network stability,
convergence times at multiple levels (e.g., MPLS traffic engineering + IP
routing
)
Source: R. Bolla, R. Bruschi, A.
Cianfrani, M. Listanti, “Putting
Backbone
Networks
to Sleep,” IEEE Network Magazine, Special Issue on “Green Networking”, vol. 25, no. 2, pp. 26-31, March/April 2011.A Taxonomy of Undertaken ApproachesSleeping/Standby: Proxying the Network PresenceGN3 Green Networking: Advances in Environmental Policy and PracticeSlide25
Solution: they exploited two features already present in today’s networks and devices:
network resource virtualizationmodular architecture of network nodes. This approach allows to:Put physical resources to sleep (e.g., links, linecards, etc.);Move the logical entities working on physical elements going to sleep, to other physical elements on the device.If suitable L2 protocols are used, the complexity of standby management can be hidden from the IP layer, and totally managed inside traffic engineering procedures.
A
Taxonomy of Undertaken ApproachesSleeping/Standby:
Proxying
the Network Presence
GN3 Green Networking: Advances in Environmental Policy and PracticeSlide26
Standby states have usually much lower energy
requirements than active states.Network-wide control strategies (i.e., routing and traffic engineering) give the possibility of moving traffic load among network nodes.When a network is under-utilized, we can move network load on few “active” nodes, and put all the other ones in standby.Different network nodes can have heterogeneous energy capabilities and profiles.Recent studies, obtained with real data from Telcos (topologies
and traffic volumes) suggested that network-wide control strategies could cut the overall energy consumption
by more than 23%.
Standby
state
Performance scaling
Power Consumption
Energy-aware state
GN3 Green Networking: Advances in Environmental Policy and Practice
A
Taxonomy of Undertaken
Approaches
Green network-wide control: Traffic engineering & routingSlide27
Only local control policies
Local + network-wide control policies
Once network devices will include energy management primitives, further energy reduction will be possible by moving traffic flows among the network nodes, in order to minimize the energy consumption of the entire infrastructure.
GN3 Green Networking: Advances in Environmental Policy and Practice
A
Taxonomy of Undertaken
Approaches
Green
network-wide
control: Traffic
engineering &
routingSlide28
GN3 Green Networking: Advances in Environmental Policy and Practice
The ECONET approachSlide29
GN3 Green Networking: Advances in Environmental Policy and Practice
The ECONET approachSlide30
GN3 Green Networking: Advances in Environmental Policy and Practice
The ECONET approachSlide31
GN3 Green Networking: Advances in Environmental Policy and Practice
Green Abstraction
Layer
The ECONET approachSlide32
GN3 Green Networking: Advances in Environmental Policy and Practice
The ECONET approachECONET Test
Bench@ TELIT Test PlantSlide33
GN3 Green Networking: Advances in Environmental Policy and Practice
Potential Impact on the Wired NetworkThe previously mentioned green
technologies allow designing
new-generation network devices characterized by “energy
profiles
”Reference: R. Bolla, R. Bruschi, A. Carrega, F. Davoli, D. Suino, C. Vassilakis, A. Zafeiropoulos, “Cutting the Energy Bills of Internet Service Providers and Telecoms through Power Management: an Impact
Analysis”, Elsevier Computer Networks, Special Issue on “Green Communication
Networks”, to appear Slide34
GN3 Green Networking: Advances in Environmental Policy and Practice
Potential Impact on the Wired NetworkTELIT reference scenarioNetwork load statistics and topology data
2015-2020 network forecast: device density and energy requirements
(example based on Italian network)
customers per DSLAM
640
average usage of a network access
30%
average traffic when
a
user is
connected
10%redundancy degree for metro/transport devices13%redundancy degree for core devices100%redundancy degree of metro/transport device links100%redundancy degree of core device links50%average traffic load in metro networks40%average traffic load in core networks40%standby efficiency85% performance scaling efficiency50% network-wide control efficiency20% air cooling/power supply efficiency15%Home/AccessMetro/Transport/CoretargetSource: forecast based on: carrier grade topologies; traffic analysis and indicators (ETSI TR 102530, ODYSSEE) and projected traffic load.power consumption (Wh)number of devicesoverall consumption (GWh/year)Home1017,500,0001,533Access1,28027,344307Metro/transport6,0001,75092
Core
10,000
17515
Sources
: 1) BroadBand Code of Conduct V.3 (EC-JRC) and “inertial” technology improvements to 2015-2020 (home and access cons.)
2) Telecom Italia measurements and evaluations (power consumption of metro/core network and number of devices)
Data
Plane
Control
Plane
Cooling
/
Power
Supply
Home
79%
3%
18%
Access
84%
3%
13%
Metro/
transport
73%
13%
14%
Core
54%
11%
35%
Device internal sources of energy consumption
Sources:
Information from vendors.
Sources:
BroadBand Code of Conduct V.3 (EC-JRC) and technology improvements to 2015-2020.Slide35
GN3 Green Networking: Advances in Environmental Policy and Practice
Potential Impact on the Wired
Network
TELIT network topology and traffic profilesSlide36
GN3 Green Networking: Advances in Environmental Policy and Practice
Yearly
Energy consumption estimation for TELIT
Potential Impact on the Wired NetworkIs There Room for Energy Saving Optimization
?
Room
for Energy Saving Optimization
Home/access
Metro/Transport
CoreSlide37
GN3 Green Networking: Advances in Environmental Policy and Practice
Potential Impact on the Wired
NetworkEnergy consumption model outline
Source
: R. Bolla, R. Bruschi, A. Carrega, F. Davoli, D. Suino, C. Vassilakis, A. Zafeiropoulos,
“Cutting
the Energy Bills of Internet Service Providers and Telecoms through Power Management: an Impact
Analysis”, Elsevier Computer Networks, Special Issue on “Green
Communication
Networks”,
to appear Slide38
GN3 Green Networking: Advances in Environmental Policy and Practice
DPS primitives only
Standby primitives only
DPS & Standby primitives
Potential Impact on the Wired
Network
Estimated energy saving for the TELIT network
We
suppose standby capabilities to be applied only where “alternative paths” are
present.Slide39
GN3 Green Networking: Advances in Environmental Policy and Practice
Potential Impact on the Wired
Network
The GRNET network case
Yearly
Energy consumption estimation for
GRNET
DPS & Standby primitives
GRNET network does not have Access/Home partsSlide40
Utrecht, Netherlands, 5-6 March 2012
GN3 Green Networking: Advances in Environmental Policy and PracticeThank you for your attention!Questions?cvassilakis@grnet.gr http://econet-project.eu
http://green.grnet.grSlide41
Utrecht, Netherlands, 5-6 March 2012
GN3 Green Networking: Advances in Environmental Policy and PracticeBackup slidesSlide42
GN3 Green Networking: Advances in Environmental Policy and Practice
Decomposing the Energy Consumption Access Technologies
Power consumption of DSL, HFC, PON, FTTN,
PtP,
WiMAX
,
and UMTS as a function of access rate with an oversubscription rate of 20. The technology used is fixed at 2010 vintage for all access rates.Source: Baliga, J.; Ayre
, R.; Hinton, K.; Tucker, R.S.; , "Energy consumption in wired and wireless access networks," IEEE Communications
Magazine
,
vol
. 49
, no. 6, pp. 70-77, June 2011.Slide43
GN3 Green Networking: Advances in Environmental Policy and Practice
Adoption of pure optical switching architectures:They can potentially provide terabits of bandwidth at much lower power dissipation than current network devices. But their widespread adoption is still hindered by technological challenges: problems mainly regard the limited number of ports and the feasibility of suitable
buffering schemes.Decreasing feature sizes in semiconductor technology have
contributed to performance gains: allowing higher clock frequencies designing improvements such as increased parallelism. the same technology trends have also allowed for a decrease in voltage that has reduced the power per byte transmitted by half every two years, as suggested by
Dennard’s
scaling
law.
A
Taxonomy of Undertaken
Approaches
Re-engineeringSlide44
Source: R.
Bolla, R. Bruschi, A. Carrega, F. Davoli, “Green Network Technologies and the Art of Trading-off,” Proc. IEEE INFOCOM 2011 Workshop on Green Communications and Networking, Shanghai, China, April 2001, pp. 301-306.
t
Φ(t)
Φ
a
(
P
y
)
Φ
idle
(
C
x
)
Φ
t
(
C
x
)
τ
on
τ
off
τ
conf
T
R
T
I
T
B
A
Taxonomy of Undertaken Approaches
Dynamic Adaptation:
Understanding the Power-Performance
Tradeoff
Modeling and control
R
ecently a
simple
model has been proposed by
Bolla
et al, which is based
on classical
queueing
theory
and
allows representing the trade-off between energy and network performance in the presence of both AR and LPI capabilities.
The model is aimed at describing the behaviour of packet processing engines.
It is based on a
M
x
/D/1/SET
queueing
system.
GN3 Green Networking: Advances in Environmental Policy and PracticeSlide45
A
Taxonomy of Undertaken Approaches
Dynamic Adaptation:
Understanding the Power-Performance TradeoffModeling and control
GN3 Green Networking: Advances in Environmental Policy and PracticeSlide46
GN3 Green Networking: Advances in Environmental Policy and Practice
A Taxonomy of Undertaken ApproachesRe-engineering: Optical Backbone Networks
The creation of optical paths (via DWDM) within optical backbone networks has been utilized for the dynamic establishment of high capacity circuits with reduced energy demands
Slide47
GN3 Green Networking: Advances in Environmental Policy and Practice
Standardization effortsThe European Union already published a number of Codes of Conduct covering different categories of equipment, including broadband equipment, data centres, power supplies, UPS. The Code of Conduct on Energy Consumption of Broadband Equipment has been defined by the EU, which sets targets in reducing energy consumption in the access network IEEE has also ratified the Energy Efficient Ethernet (EEE) standard in October 2010, also known as IEEE 802.3az,
which is a set of enhancements to the twisted-pair and backplane Ethernet networking standards that will allow for more than 50% less power consumption during periods of low data activity, while retaining full compatibility with existing equipment.
ENERGY STAR is a joint program of the U.S. Environmental Protection Agency and the U.S. Department of Energy that has defined the ENERGY STAR Product Specifications. IETF has recently established the Energy Management (EMAN) Working Group. Different interesting issues are under consideration by the Environmental Engineering Technical Body in ETSI
The Home Gateway Initiative (HGI) launched an internal task force called ”Energy Saving”
with
the objective of setting up requirements and specifications for energy efficiency in the home gatewaysITU-T Study Group 15 (Optical transport networks and access network infrastructures)ITU-T created in September 2008 a new Focus Group, namely, FG ICT & Climate ChangeSlide48
ECONET approach
GN3 Green Networking: Advances in Environmental Policy and Practice