idnits 2.17.00 (12 Aug 2021)
/tmp/idnits2040/draft-zhang-panet-problem-statement-03.txt:
Checking boilerplate required by RFC 5378 and the IETF Trust (see
https://trustee.ietf.org/license-info):
----------------------------------------------------------------------------
No issues found here.
Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt:
----------------------------------------------------------------------------
No issues found here.
Checking nits according to https://www.ietf.org/id-info/checklist :
----------------------------------------------------------------------------
** The document seems to lack an IANA Considerations section. (See Section
2.2 of https://www.ietf.org/id-info/checklist for how to handle the case
when there are no actions for IANA.)
Miscellaneous warnings:
----------------------------------------------------------------------------
== The copyright year in the IETF Trust and authors Copyright Line does not
match the current year
-- The document date (October 15, 2013) is 3133 days in the past. Is this
intentional?
Checking references for intended status: Informational
----------------------------------------------------------------------------
== Unused Reference: 'EMAN-WG' is defined on line 295, but no explicit
reference was found in the text
== Unused Reference: 'TM' is defined on line 326, but no explicit reference
was found in the text
Summary: 1 error (**), 0 flaws (~~), 3 warnings (==), 1 comment (--).
Run idnits with the --verbose option for more detailed information about
the items above.
--------------------------------------------------------------------------------
2 Internet Engineering Task Force B. Zhang
3 Internet-Draft J. Shi
4 Intended status: Informational Univ. of Arizona
5 Expires: April 18, 2014 J. Dong
6 M. Zhang
7 Huawei
8 M. Boucadair
9 France Telecom
10 October 15, 2013
12 Power-Aware Networks (PANET): Problem Statement
13 draft-zhang-panet-problem-statement-03
15 Abstract
17 Energy consumption of network infrastructures is growing fast due to
18 exponential growth of data traffic and the deployment of increasingly
19 powerful equipment. There are emerging needs for power-aware routing
20 and traffic engineering, which adapt routing paths to traffic load in
21 order to reduce energy consumption network-wide. This document
22 outlines the design space and problem areas for potential IETF work.
24 Status of this Memo
26 This Internet-Draft is submitted in full conformance with the
27 provisions of BCP 78 and BCP 79.
29 Internet-Drafts are working documents of the Internet Engineering
30 Task Force (IETF). Note that other groups may also distribute
31 working documents as Internet-Drafts. The list of current Internet-
32 Drafts is at http://datatracker.ietf.org/drafts/current/.
34 Internet-Drafts are draft documents valid for a maximum of six months
35 and may be updated, replaced, or obsoleted by other documents at any
36 time. It is inappropriate to use Internet-Drafts as reference
37 material or to cite them other than as "work in progress."
39 This Internet-Draft will expire on August 29, 2013.
41 Copyright Notice
43 Copyright (c) 2013 IETF Trust and the persons identified as the
44 document authors. All rights reserved.
46 This document is subject to BCP 78 and the IETF Trust's Legal
47 Provisions Relating to IETF Documents
48 (http://trustee.ietf.org/license-info) in effect on the date of
49 publication of this document. Please review these documents
50 carefully, as they describe your rights and restrictions with respect
51 to this document. Code Components extracted from this document must
52 include Simplified BSD License text as described in Section 4.e of
53 the Trust Legal Provisions and are provided without warranty as
54 described in the Simplified BSD License.
56 Table of Contents
58 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3
59 2. Motivation and Problem Scope . . . . . . . . . . . . . . . . . 3
60 3. Potential Solution Approaches . . . . . . . . . . . . . . . . 4
61 4. Problem Areas for IETF . . . . . . . . . . . . . . . . . . . . 6
62 5. Security Considerations . . . . . . . . . . . . . . . . . . . 7
63 6. Informative References . . . . . . . . . . . . . . . . . . . . 7
64 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 9
66 1. Introduction
68 Driven by exponential growth of Internet traffic, networks worldwide
69 are expanding their infrastructures at a fast pace by deploying more
70 high-capacity, power-hungry routers, which also leads to increasing
71 energy consumption. For example, in the US, the energy bill for
72 powering the wired network reaches up to 2.4 billion dollars per year
73 [Doverspike10]. Telecom Italia, the largest ISP in Italy, is now the
74 second largest consumer of electricity after the National Railway
75 system [Pileri07]. As one of the biggest energy consumers in the
76 United Kingdom, British Telecom consumed about 0.7% of the entire
77 nation's electricity in 2007 [Bolla11]. In Japan, predictions say
78 that routers will consume 9% of the total electricity by 2015
79 [Nakamura07]. Besides operational costs and environmental impacts,
80 the ever-increasing energy consumption has become a limiting factor
81 to long-term growth of network infrastructure due to challenges in
82 power delivery and heat removal for both router components and
83 hosting facilities [Gupta03] [Epps06].
85 Traditionally energy efficiency is improved at the device level or
86 the link level. For example, energy management techniques can be
87 applied to adjust router CPU's power status or CPU frequency in
88 response to different CPU workload; Links can be put to sleep mode
89 when it has been idle for a while. More recently, there have been a
90 number of research work that look beyond a single router or linecard
91 for network-wide solutions towards energy proportionality.
93 The purpose of this document is to discuss the problem scope, outline
94 potential approaches, and problem areas for IETF work on power-aware
95 networks.
97 2. Motivation and Problem Scope
99 Today's ISP networks have redundant routers and links, over-
100 provisioned link capacity, and load-balancing traffic engineering. As
101 a result, routers and links operate at full capacity all the time
102 with low average usage, typically less than 40% of link utilization.
103 This practice makes networks resilient to traffic spikes and
104 component failures, but also makes networks far from energy-
105 efficient.
107 Power-aware routing and traffic engineering have been proposed to
108 improve network's energy efficiency, for example, by aggregating
109 traffic onto a subset of links and putting other links with no
110 traffic into sleep. Data from various sources (e.g., [Heddeghem12]
111 [Chabarek08]) have shown that line cards are a significant source of
112 router's power consumption, accounting for 40% - 70% of total power
113 consumption. Most of the energy is consumed even in standby state,
114 and forwarding packets at full speed only increases the energy
115 consumption by a small percentage. This implies that being able to
116 put links into sleep mode can potentially save a lot of energy. In
117 face, this has been demonstrated in several research works such as
118 [GreenTE] [Nedevschi08] [Chabarek08].
120 Designing practical protocols, however, has been challenging, because
121 making routing protocols power-aware brings significant changes to
122 the routing system and the entire network, thus it involves hardware
123 support, protocol design, network monitoring, and operational
124 practices. These issues often depend on the specific network
125 environments under discussion. In order to focus on protocol-related
126 issues, we suggest that as the first step we limit the scope of the
127 discussion to intra-domain routing within one administrative domain,
128 to avoid inter-domain policy issues. This includes transit networks
129 as well as edge networks. We leave data center networks out of this
130 draft since that usually requires concerted efforts beyond network
131 protocols.
133 3. Potential Solution Approaches
135 The high-level idea of power-aware networks is to adjust routing
136 paths based on traffic level. When traffic level is high, use more
137 links to carry the traffic; when traffic level is low, merge traffic
138 onto a subset of all links so that other links can be put to sleep or
139 reduce rate in order to save power. This needs to be done without
140 significantly impacting network QoS, network resiliency, and
141 interoperation with other protocols.
143 In the last few years a number of power-aware network designs have
144 emerged. Instead of listing them individually, here we categorize
145 the solutions along three different dimensions.
147 Link Sleep vs. Rate Adaptation
149 Sleeping and rate adaptation are two major ways to save energy in
150 computer systems. Many hardware, including line cards and chassises,
151 consumes a significant amount of power when they stand by without
152 doing any actual work. When put into sleep mode, they will consume
153 only a little power. Thus putting an idle component to sleep is a
154 common way to save energy. If there is a need to use this component,
155 it can be waken up and become usable after a transition time. The
156 longer a component is in sleep mode, the more power saved. A power-
157 aware protocol adjusts routing paths to increase the sleep time for
158 certain links in the network.
160 A network interface often supports multiple data rates. Operating at
161 a lower data rate usually consumes less energy, though the actual
162 rate-power curve varies from device to device. Rate-adaptation-based
163 approaches operate interfaces at lower data rates when the traffic
164 demand is low and increase the data rate when traffic demand is high.
165 Thus the routers can save power during low utilization period.
167 These two approaches are also related in the case of "bundled links"
168 [Fisher10]. A bundled link is a virtual link comprised of multiple
169 physical links. A sleep-based approach can put some physical links
170 into sleep to save power, which is same as conducting rate adaptation
171 on the virtual link with adjustment unit of a physical link.
173 Configured vs. Adaptive
175 The key in power-aware routing and traffic engineering is to adjust
176 routing paths in response to traffic changes, so that the power state
177 of routers (or router components) will also change accordingly to
178 achieve energy saving. Different approaches differ at the
179 granularity of the adjustment.
181 Some approaches take the long-term traffic average as input, and
182 output a routing configuration that is applied to the network
183 regardless of short-term traffic variation. This is mostly useful
184 when network traffic exhibits a stable, clear pattern, e.g., diurnal
185 pattern where traffic is high during work hours and low during off
186 hours. It can only exploit the target traffic pattern; it cannot
187 react dynamically to short-term traffic changes to either save energy
188 (by putting links to sleep) or avoid congestion (by waking links up),
189 but the design and implementation should be simple.
191 Another type of approach is to adapt to traffic changes dynamically
192 on much smaller time granularity. This approach may be able to save
193 more energy and have better performance because it is more
194 responsive, but the design and implementation usually are more
195 complicated. This approach needs to continuously collect traffic
196 data in order to adjust routing dynamically. The adjustment may be
197 done periodically or whenever significant traffic changes are
198 observed.
200 Distributed vs. Centralized
202 In distributed solutions, routers make power-aware adjustment
203 decisions, such as link sleep/wake-up and rate increase/decrease,
204 locally without a central controller. These routers need to exchange
205 information in order to achieve consistent network states.
206 Distributed approach fits the Internet operation model well but its
207 design is the most challenging. Traditional routing does not respond
208 to traffic variation while power-aware routing does, and it needs to
209 do so without causing loops or congestions.
211 In centralized solutions, a controller computes the routing paths
212 considering the network topology and traffic demand, and informs
213 routers how to adjust their routing paths. A centralized server
214 usually has more complete information, more computation power, and
215 more memory and storage than routers, thus it may make better
216 decisions than distributed approach. The server locates in the
217 network NOC and can be backed up by server replicas. Nevertheless,
218 this approach requires high reliability of the server.
220 Both distributed and centralized solutions may find their places in
221 ISP networks. For example, centralized solution can be integrated
222 into the Path Computation Element (PCE) framework [PCE-WG]. There
223 can also be hybrid designs, e.g., using a centralized solution based
224 on long-term traffic pattern, and distributed mechanisms to handle
225 short-term traffic variations.
227 4. Problem Areas for IETF
229 Power-aware networks have great potentials to improve network energy
230 efficiency while maintaining network services at desired levels. Its
231 effectiveness, however, depends on various supports from hardware and
232 software, especially protocol designs that address operational
233 issues. In this section we list a few problem areas that will
234 benefit from additional input from the IETF community, or have the
235 potential to become work items in related IETF working groups.
237 Motivation and Problem Scope
239 o What are the motivations for Power-Aware Networking (PANET)?
241 o To what extent power consumption is a key factor for Internet
242 scaling?
244 o To what extent power-aware system at router level and link level
245 are not sufficient to reduce the overall energy consumption of
246 networks?
248 Technical Development
250 o What are the technical requirements for an efficient PANET
251 solution?
253 o What are the technical tracks to reduce the overall power
254 consumption at the level of an IP network?
256 o How protocols can be designed to be power-aware and still maintain
257 enough network resiliency?
259 o What are the technical challenges for deploying efficient PANET
260 solutions?
262 o How routing protocols (e.g., OSPF) can be extended to disseminate
263 power-related information?
265 o How PCE architecture can be used to compute power-aware paths?
267 o How PANET can be deployed in centralized or in distributed model?
269 Operation Practice
271 o What will be the impacts of PANET to network operations?
273 o What will be the guidelines for deploying PANET systems?
275 5. Security Considerations
277 This draft is a discussion on the Internet's necessity to follow an
278 evolutionary path towards the future. There is no direct impact on
279 the Internet security.
281 6. Informative References
283 [Bolla11] Bolla, R. and et al. , "Energy Efficiency in the Future
284 Internet: A Survey of Existing Approaches and Trends in Energy-
285 Aware Fixed Network Infrastructures", IEEE Communications Surveys
286 and Tutorials, 2011.
288 [Chabarek08] Chabarek, J. and et al. , "Power Awareness in Network
289 Design and Routing", IEEE INFOCOM 2008.
291 [Doverspike10] Doverspike, R., Ramakrishnan, K., and C. Chas,
292 "Structural overview of ISP networks", Guide to Reliable Internet
293 Services and Applications, Springer, 2010.
295 [EMAN-WG] "IETF Energy Management Working Group", 2012,
296 .
298 [Epps06] Epps, G. and et al. , "System Power Challenges", 2006,
299 .
302 [Fisher10] Fisher, W. and et al. , "Greening Backbone Networks:
303 Reducing Energy Consumption by Shutting Off Cables in Bundled
304 Links", Green Networking 2010.
306 [GreenTE] Zhang, M. and et al. , "GreenTE: Power-Aware Traffic
307 Engineering", ICNP 2010.
309 [Gupta03] Gupta, M. and S. Singh, "Greening the Internet", ACM
310 SIGCOMM 2003.
312 [Heddeghem12] Van Heddeghem, W. and F. Idzikowski, "Equipment power
313 consumption in optical multilayer networks - source data", IBCN
314 Technical Report 2012.
315 [Nakamura07] Nakamura, M., "Advanced photonic technologies for the
316 information era", Nature Photonics Technology conference, 2007.
318 [Nedevschi08] Nedevschi, S. and et al. , "Reducing Network Energy
319 Consumption via Sleeping and Rate- Adaptation", USENIX NSDI 2008.
320 [PCE-WG] "IETF Path Computation Element Working Group", 2012,
321 .
323 [Pileri07] Pileri, S., "Energy and communication: engine of the human
324 progress", 2007.
326 [TM] Roughan, M., Thorup, M., and Y. Zhang, "Traffic Engineering with
327 Estimated Traffic Matrices", IMC 2003.
329 Authors' Addresses
331 Beichuan Zhang
332 Univ. of Arizona
334 Email: bzhang@cs.arizona.edu
336 Junxiao Shi
337 Univ. of Arizona
339 Email: shijunxiao@cs.arizona.edu
341 Jie Dong
342 Huawei
344 Email: jie.dong@huawei.com
346 Mingui Zhang
347 Huawei
349 Email: zhangmingui@huawei.com
351 Mohamed Boucadair
352 France Telecom
354 Email: mohamed.boucadair@orange.com