Interim Report

COMP 10024 Computer Networking 4 Project

“Power Consumption of Network Devices”
Interim Report

Author: Andrew J. Jess (B00113374) th Due Date: 16 December 2009 Word Count: 8,841

Table of Contents
1 Introduction
1.1 1.2 1.3 Background Justification Objectives

Page

4 5 6

2

Literature Review
2.1 2.2. Power Management of Devices
2.1.1 2.1.2 2.2.1 2.2.2 2.2.3 Advanced Configuration and Power Interface (ACPI) CISCO EnergyWise EnergyStar Wake-on-LAN Power-over-Ethernet 802.3az Adaptive Link Rate Pause Power Cycle Proxying

6 8 11 12 13 14 15 17 18 20

Present Initiatives

2.3

Future Initiatives
2.3.1 2.3.2 2.3.3 2.3.4

2.4

Conclusions

3

Progress
3.1 Basic Objective 1
3.1.1 3.1.2 3.1.3 3.1.4 Overview Energy Consumption by Office and Telecommunications Equipment in Commercial Buildings Case Study of Data Centres’ Energy Performance Conclusions & Chosen Methodology

21 22 25 28 28

3.2 3.3

Basic Objective 2 Basic Objective 3
Experiment 1: Measuring Power Usage of Various S-states Experiment 2: Data Load’s Impact on Power Consumption of a Network Switch

29 33 35 36 37 39

3.4

Advanced Objectives

4 5 6

Summary & Completion Plan References Appendices

Page 2

Table of Figures
Tables
Table 2.1: Table 2.2: Table 3.1: Table 3.2: Table 3.3: S-States of the ACPI Standard CISCO EnergyWise Category & Power Level Table Sun & Lee’s Device Criteria Results of Experiment 1 Results of Experiment 2

Page

6-7 9 26 31 34

Figures
Figure 2.1: Figure 2.2: Figure 2.3: Figure 3.1: Figure 3.2: Figure 3.3: Figure 3.4: Figure 3.5: Figure 3.6: The form of a typical magic packet PPC in operation The operation of a proxy AEC of Office Equipment AEC of Network Devices Breakdown of energy use of a data centre Setup of Experiment 1 Results of Experiment 1 Setup of Experiment 2
14 17 18 22 24 27 29 31 33

Appendices
Appendix 1: Appendix 2: Dell Optiplex 755 Datasheet Netgear WGR614v9 Datasheet
39 40

Page 3

1

Introduction

1.1 Background
As prices for energy continually rise in today’s world, companies and manufacturers alike are under enormous pressure to make their operations more environmentally friendly from international energy efficiency organisations. “Greening” operations offers numerous advantages to companies resulting not only in savings on energy bills, but also in terms of reducing the company’s emissions “footprint” and increasing the company’s environmental reputation (which is of inestimable value). However, with Information Technology (IT) becoming ever more abundant within enterprises, and with a mounting need for a strong network backbone to serve and process these installations’ data, more and more electricity is required to power them. One of the more notable studies on the power consumption of office and telecommunications equipment estimated the United States’ annual consumption at 97TW-h in 2002 [1], an annual cost of $7.65 million† (£4.62 million). A projection of energy prices published in 2005 projected electricity price increases of 10% between 2005 and 2010 [3]. Coupled with a staggering adoption rate in the IT sector (an investment proportion of 40% in 1998, and rising [4]), it can be assured that energy bills will follow a similar upwards pattern. This report seeks to investigate the electricity requirements of network devices, of data communication in general, and consider whether the workload placed upon networks impacts the amount of energy consumed.

†

2002 price of 7.89 cents per kW-h

[2]

Page 4

1.2 Justification
Although several studies with enormous scope have been performed considering the “energy footprint” of office equipment and data centre operation, there has been little to no work on the subject of network infrastructure devices specifically. As a precursor to investigating typical electricity costs in an organisation, examinations of what wider studies could be located will be made, with a view to adopt and adapt an author’s methodology to be applied in the context of this project. Also of interest are the numerous Ethernet technologies being developed, which will result in more energy efficient data links being introduced into networks. A detailed exploration of several of these breakthroughs will be contained within this report’s literature review.

1.3 Objectives
The objectives of this project are as follows: Basic Investigate the costs involved in maintaining the operation of a typical organisation’s IT infrastructure Investigate and calculate the theoretical power requirements of data transmission. Observe and measure the power consumption of devices in a typical network, both under load and whilst idle. Advanced Compare and analyse observed power usage data against theoretical projections Compare the power usage of idle devices with those under load.

Page 5

2

Literature Review

2.1 Power Management of Devices
2.1.1 Advanced Configuration and Power Interface (ACPI)
Network devices, regardless of their type can all be considered to have certain modes of operation. Although certain types of IT equipment can only be considered “on” or “off”, more advanced devices such as workstations and network infrastructure devices can be powered down to intermediate steps where less power is consumed. Any one type of device can have different rates of energy consumption depending on the mode it is being run in. The Advanced Configuration and Power Interface (ACPI) standard defines a set of power states for systems and devices and was developed in conjunction with major software vendors including Hewlett-Packard, Intel and Microsoft. Support also exists for migrating Linux machines to this standard [5]. Table 2.1 describes a list of system “S-States” which define the power status of an ACPIcompliant workstation.

ACPI Level

Mode Name

Function The system is operating normally, all components are receiving power. Although not mentioned in the ACPI standard, Roth et al. note the distinction between active-idle (where the system is not actively processing) and active-processing (where the system is performing computations [6]) and as such there can be a wide difference in power consumption from devices in this mode. Levels S1 to S4 all define sleep levels of variable depth. S1 preserves most operation and conserves the least amount of power, whereas S4 provides the largest savings and powers down every possible component of the system.

S0

Working

S1 to S4

Sleep

Microsoft Windows computers commonly use Standby as their primary power saving mode. This mode operates at level S3. Here, user data is stored in RAM and non-essential components of the system are shut down. The CPU is provided no power, hard disks are switched off but RAM is in a constant “refresh mode” to keep the user’s data intact.

Page 6

S4 mode is supported by more recent operating systems and is known as “hibernate”. This mode saves more power than its standby equivalent by saving an image of the system’s memory to hard disk before powering down, eliminating the need for the system’s RAM to be refreshed. System is completely powered down and requires a full reboot to return to S0 state. The system is still connected to the mains supply and draws a nominal amount of power (see Phantom Load) Not a part of the official ACPI standard, state S6 is sometimes used to refer to the global “mechanical off” state G3 [7] and implies that the electricity supply is physically removed from the system.
Table 2.1: S-States of the ACPI Standard
[8]

S5

Soft Off

“S6” (G3)

Mechanical Off

It should be observed that even if a computer is considered to be in S5 mode, it will still draw a nominal amount of power. Roth’s measurements show that even when powered off (but still plugged in) personal computers and notebooks still draw 2W [6]. This phenomenon is known as “phantom load” and is common to all electronic devices. Also known as “standby power” or “vampire power”, it has been identified as a major source of energy wastage and has been a focus of many governments’ energy efficiency undertakings [9]. As such, discussions considering the benefits and drawbacks of standby modes are relatively frequent. With the difference of power consumption between S3 and S5 modes being so small, leaving a system in standby mode overnight may be almost as energy efficient as shutting it down. Harris & Cahill even suggest that power mode transitions from deeper ACPI modes typically consume extra energy (due to device start-ups) and can even reduce a system’s mechanical lifetime (due to mechanical wear) [10]. A moot point to these arguments is that putting a device into “S6 mode” always garners the most savings. Removing a device from the electricity supply always reduces its power consumption to 0W. The ACPI model pertains mainly to workstation systems developed by the contributing vendors. Of interest to other network infrastructure devices is the CISCO EnergyWise initiative which uses a scale similar to (but greater in scope than) the ACPI model which all network devices would comply to.

Page 7

2.1.2 CISCO EnergyWise
IT devices and its infrastructure are not the only consumers of electricity in an organisation. As lighting and heating alone account for 66% of an organisation’s electrical energy consumption (compared with IT equipment’s 25-30%) [11], the prospect of managing complete organisational power consumption underneath one central system is an appealing idea. The EnergyWise initiative (developed by CISCO Systems) is a proposed energy management architecture which seeks to measure and collect power information from all its connected devices, with an aim to allow administrators to better optimize the power consumption of an organisation. It looks beyond simply conserving power of network IT equipment and instead aspires to control all aspects of an organisation’s power usage. In order to do this, EnergyWise defines several attributes that are used to model the organisation’s systems.

1

Categories & Power Levels

Similarly to the ACPI protocol discussed earlier in this review, a common language used to define power states between devices is required to standardise their management. ACPI, having applications only for PC workstations and compliant mobile devices would be an inappropriate choice for EnergyWise. Instead, CISCO developed a new set of power levels for their management system to utilise, creating a “common lexicon” [11] so that power levels can be understood between devices. In particular, this meant that existing power management standards (such as ACPI) could be mapped directly onto the EnergyWise system.

Page 8

Table 2.2: CISCO EnergyWise Category & Power Level Table

[11]

It should be noted that the Table 2.2 has varying levels of complexity depending on the device it is referring to. For example simple devices such as grids of lighting may only use two modes, Operational and Non-Operational. More complex devices such as PCs will have their ACPI modes married up with a “level” in the table above.

2

Entities

Entities represent power consuming devices connected to the EnergyWise network and can consist of several different types. Entities may be IP-based (even differentiating between Power-over-IP enabled and standard IP) or not. A category exists as well for controllers which operate systems unrelated to the IT infrastructure of the network, such as heating or lighting. Devices, no matter what type, are considered children of the EnergyWise enabled network switch that they are connected to. Network witches are typically the entities that management systems will interface with in order to control the EnergyWise system.

Page 9

3

Domains

Each entity as described above must be a member of a domain. This allows devices to be logically arranged into groups to allow more effective management of the network (in turn better facilitating its expansion). For example, under the domain system, sets of switches (and their children) could be grouped together based on the building floor they reside on.

4

Management Communications

EnergyWise also defines its own communication methods in order to send commands from a central management location to its devices. CISCO has suggested two methods to implement this. The first is to send messages using the Simple Network Management Protocol (SNMP) which provides a framework for network administration tasks. EnergyWise provides its own Management Information Bases (MIBs) defining how to handle data produced by the system. This allows for simple management of one switch; however Lippis notes that the limitations of SNMP make it unsuitable for managing domains containing more than one switch [11]. On a central switch, a single “Management Port” can be defined that will allow administrators to gather domain-wide information by issuing commands to it. Support for requesting and changing the power levels for tens of thousands of entities is purportedly possible [11].

5

Management Applications & API

In order to control the EnergyWise network, CISCO have provided a common API in order to allow third-party vendors to develop network management applications utilising EnergyWise information. The API allows power consumption and device efficiency data to be pulled simply from the network and be translated into meaningful colour-coded topologies. This would allow companies which have already published software controlling various aspects of a network to easily allow power-state management to the set of features offered.

Page 10

The advent of EnergyWise promises to expand the role of switches within a network. Instead of switching only traditional IP traffic, switches will soon become responsible for delivering management instructions to devices or gathering reports of power consumption over a period of time. As non-IT devices are incorporated into the EnergyWise topology, switches may soon be able to perform such complex functions as alter the temperature of a building dependant on the time of day. The scope for savings that can be gathered by a system such as this is wide with switches being able to orchestrate the power states of devices automatically on a regular basis.

2.2 Present Initiatives
A number of initiatives have been undertaken in order to kerb the amount of energy used by IT infrastructure, network devices and more widely, electrical devices as a whole.

2.2.1 EnergyStar
EnergyStar is a standard specifying power consumption requirements for a range of electronic devices. Originally created in 1992 as an American government-funded program to encourage computer manufacturers to include power management options in their products [12], it has since expanded to consider consumer and commercial products, as well as devices such as lighting and air conditioners [13]. As a voluntary accreditation, it is not required for manufacturers to subscribe to, but its high reputation amongst consumer groups provides an incentive for compliance. EnergyStar’s current fifth specification revision maintains directives on a number of different computer systems. Desktop computers, notebooks, games consoles and workstations amongst others are all included. However, server computers and more recent mobile devices (PDAs and smart phones) are not included in the specification [14]. Also of note is their specification for notebook computers which requires a low-power mode consuming no more than 15W, which McWhinney notes a large percentage of notebooks comply with [15]. EnergyStar has proven to be a very popular scheme, as demonstrated by the range of devices it now covers, and its expansion internationally. In their 2006 annual report, EnergyStar reported that compliant desktop computers are shown to save between 5% and 55% more than their nonaccredited counterparts. The program in its entirety also published annual savings of $13.7 million in the year of publication, along with considerable emission reductions from the year of 2000

Page 11

onwards [13]. As such, companies with large IT load outs can be assured that purchasing products accredited by EnergyStar, conserves more energy and creates less carbon emissions.

2.2.2 Power-over-Ethernet
Power-over-Ethernet (PoE) describes a set of standards which define a method of transmitting power over an Ethernet link whilst not disturbing the data contained on it. First published as 802.3af in 2003, a new version of the standard known as 802.3at was recently approved in September 2009
[16]

featuring marked improvements to the amount of power supported devices could provide. The

publication of 802.3af/at also serves to encourage standardisation of all previous work performed in the same area, such as CISCO Systems’ “inline power” technology [17]. Originally developed to provide both power and network connectivity to locations where power cabling was impractical or impossible to provide, the main advantage of PoE lies in the ability to discard the traditional AC transformer based method of supplying power to devices. PoE is of particular application to devices such as CCTV cameras and wireless repeaters (which are often positioned in out of reach locations) as well as making Voice-over-IP (VOIP) phones resemble their “plain old telephone system” counterparts more (which similarly drew power from their copper transmission line). Two types of devices exist in the operation of PoE: Power Sourcing Equipment (PSE) – PSE equipment is typically a PoE enabled network switch which supplies electricity to connected devices. Devices known as Midspan Power Sources (MPS) are also used along with traditional Ethernet switches to “inject” power into existing Ethernet networks in the absence of a PSE switch. Powered Device (PD) – These connected devices are known as PDs, and are supplied power from the PSE via twisted pair cable. The 802.3af specification provides only around 13W of power to be supplied [18]. Whilst certainly not enough to power larger devices such as PCs and large printers, PoE has found its niche powering smaller pieces of equipment that only require nominal amounts of energy.

Page 12

Despite the clear advantages that a PoE infrastructure would bring to an organisation, attention must be paid to the backwards compatibility of the platform. As thousands [18] of varieties of standard Ethernet equipment have been deployed across the world, it would be foolish to arbitrarily inject power into them (risking damage and device failures). To prevent this, a discovery process is embedded into DSE devices which maintain their ports in a low-power state until devices are determined to be PoE compliant. A tentative low voltage (in the range of 2.7V to 10.1V) is then applied to a PD upon connection and the PSE checks for a built in a “signature resistance” of 25kΩ before supplying larger amounts of power [19]. Additional concerns have been raised [18] over the safety of using existing RJ-45 connectors to supply power with, particularly as the female socket is large enough for a small finger to be inserted into. However, as the 802.3af standard only provides a small DC voltage (48V) and an extremely low current (up to 300-375mA maximum) [18] [19] through the twisted pair wire, no harm can be caused.

2.2.3 Wake-on-LAN
Wake-on-LAN (WoL) is a technology designed to be used on Ethernet-compliant devices and allows devices to be turned on via network communication from another device. WoL has been available for over a decade with various implementations supported by different hardware vendors [20] [21]. Prior to the introduction of WoL, computers could only be communicated with if they were in an ACPI S0 state. When technicians realised that they required a method to communicate with computers kept in other states, WoL was developed in order to “pull” a device out of its low power state and back into S0 mode. WoL functions by requiring a device’s network adapter to remain operational whilst the rest of the device is powered down. This results in a nominal amount of “standby-power” being drawn by the device to keep it operational. The network adapter of the device would also contain software that continually listened for “magic packets”. Upon the reception of a magic packet, the network adapter would send a signal to its host, prompting it to “wake up” into S0 mode. A “magic packet” requires a certain sequence to be contained within it in order to awaken a system. It can appear anywhere in the packet’s payload, but the sequence must take the form of six “one” bytes (represented by FF) followed by sixteen full iterations of the device’s six byte MAC address (represented in figure 2.1 as 11 22 33 44 55 66).

Page 13

Figure 2.1: The form of a typical magic packet

[20]

WoL could find a valuable place as part of an organisation’s power management plan. It would be possible for administrators to remotely power on machines on an as-needed basis rather than remain powered on indefinitely. However, one of WoL’s limitations is its unidirectional nature, only being able to wake systems. A worthwhile expansion of the technology would allow magic packets to shut down systems remotely. “Proxying” (discussed in Section 2.3.4) can be considered in some regards as a more sophisticated implementation of WoL.

2.3 Future Initiatives
2.3.1 802.3az
In October 2007, the Institute of Electrical and Electronics Engineers (IEEE) approved the 802.3az project to investigate and improve the energy efficiency of the 802.3 Ethernet standards. Its main objectives involve developing techniques for lowering the power use of Ethernet whilst retaining compatibility with the current physical media that use it. As Ethernet is a family of technologies operating at OSI Layer 1 and 2 and is used in the majority of Local Area Networks today, incorporating energy saving techniques into the technologies themselves will yield savings from every network that utilises them.

Page 14

Currently the 802.3az project has published proposals highlighting three techniques which could increase the energy efficiency of Ethernet devices. Although distinct in their application, each proposal brings attention to the fact that most networks are kept on 24 hours a day, even when they aren’t required by users.

2.3.2 Adaptive Link Rate
The first proposal published by the 802.3az project was a paper on using Adaptive Link Rate (ALR) mechanisms as a method of controlling the power usage of Ethernet links [22]. ALR was developed and refined out of the realisation that Ethernet links remain idle or in low use for a very large proportion of the time (with studies showing average Ethernet link utilisation of only 1% [23]). Bennett proposes in the ALR proposal that as the capacity of network media and transmitters increase so will the energy required to power and maintain the links. They observe that Ethernet links operating at 1Gbps require 2W more power at each transmitter than equivalent links operating at 100Mbps. As such they propose that during periods of low network usage, ALR would allow Ethernet links to “step down” transmission speeds in order to save power. Similarly, links would “step up” to higher rates as their services were demanded. ALR operates from both ends of the transmission link. Both transmitter and recipient interfaces would use in-built “policies” to automatically negotiate whether data rates should be stepped up or down. Working as a handshake mechanism, a change would be made only if both parties agreed. Factors such as buffer queue thresholds and actual rate utilisation would considered in this decision. Two scenarios would be possible: Increase from low data rate to high: Bennett proposes that the size of the transmitter’s buffer queue be used in determining the need for a higher data rate. When over a certain amount, the burdened interface would send a frame to the recipient requesting a transition. If a higher rate is available, the request to “step-up” must never be denied by the recipient in order to guarantee maximum throughput. Decrease from high data rate to low: The link utilisation of the interface would be monitored. If below a certain threshold, the interface would send a frame requesting a rate “step-down”. However, if the other interface’s link utilisation did not also fall beneath the threshold, the request must be denied.

Page 15

Using conditions such as the above would guarantee that higher data rates would always take precedence over energy conservation. Also, as “step-up” and “step-down” requests would be implemented using a fast signalling method at the MAC level, transitions could take place promptly. This would make the amount of perceived delay negligible to the user. As the policies for stepping up and stepping down data rates must be contained in both transmitter and receiver Ethernet controllers, both devices would have to be compliant with the ALR protocol. Unfortunately, devices in use today are not. No standard currently exists for ALR and a considerable amount of work on the “open challenges” present in the technology must be performed before one will be developed. Additionally, once a standard has been published, will existing Ethernet devices be able to comply with it? Or must they be upgraded to more recent devices? If the latter scenario is true, it will almost certainly cost a considerable amount of money for most businesses to replace every Ethernet controller present in their network. And even if it becomes possible to upgrade existing controllers, it will take a considerable amount of time until the technology is widespread enough to be enforced vigorously enough to yield the massive monetary savings heralded by its authors. A paper by Nedevschi et al. notes that EnergyStar standard proposals for 2009 discuss requirements for Ethernet links to use slower data rates in order to conserve energy when idle [24]. As such, ALR or a technique similar to it may see inclusion within the EnergyStar specification in the near future.

Page 16

2.3.3 Pause Power Cycle
Pause Power Cycle (PPC) is a method used by LAN switches that involves adapting the power states of its own components in accordance with the states of the active links that are connected to them. The author of the technique, Blanquicet, suggests that rather than remaining powered on 24 hours a day, the main goal of switches should be to transmit data as fast as possible and then return to a “low power idle-mode” [25]. The PPC method is an implementation of this ideology.

Figure 2.2: PPC in operation

[25]

Figure 2.2 shows how PPC might be used in a typical network. The switch periodically sends PAUSE frames to network devices and temporarily powers off the link, conserving energy. After a timer elapses, the link is then powered back on and resumes transmission of data. Blanquicet’s initial calculations on power saving show that the energy conserved by PPC is directly related to the proportion of time it is powered down. He refers to the ratio of uptime to downtime as the switch’s “duty cycle” and cites that if it were set at a value of 50% (essentially halving its uptime), the amount of energy required by the device would be halved. The amount of energy saved through PPC seems to depend on sacrificing network throughput quality. By lowering the amount of time a link is powered on, the effective throughput of the medium is reduced. Banquicet asserts that his technique may result in occasional buffer overflows in clients (resulting in packet loss) and his experiments with PPC’s duty cycle set to 50% show the introduction of erroneous artefacts to streamed video [25]. In high speed environments, such as Local

Page 17

Area Networks, this may not be such an issue, as data can be retransmitted quickly over media with large capacities. However, in Wide Area Network environments where the available bandwidth is considerably lower, these errors suggest that PPC may need its duty cycle set to a less aggressive setting (or be disabled altogether) to provide acceptable throughput. In conclusion, this technique is a direct trade-off between link quality and device power consumption.

2.3.4 Proxying
September 2007 saw a proposal exploring a process known as “proxying”. The concept of “proxying” provides a method for network terminals to be able to retain their network connectivity regardless of their power mode. An additional device (known as the “proxy”) would act as an intermediary to the terminal and preserve the network state of its parent device. Nordman argues that many messages destined for a workstation don’t require the use of many of its many “power hungry” components (such as CPU, hard drive and memory) and can be handled by the network interface card (NIC) itself [26]. The proxy’s main task would be to identify these messages, generate routine replies for them and determine whether the device requires to waking up. This would allow a workstation to remain in a standby power mode while the proxy dealt with maintaining its network presence.

Figure 2.3: The operation of a proxy

[26]

Page 18

Figure 2.3 demonstrates the proxying process as currently proposed. Three distinct entities are present: the proxy, the sleeping device and the external network. Five steps are defined in the process of proxying: 1 A scenario arises where the device is in the process of going into sleep mode (e.g. under user direction, after a period of inactivity) 2 Before completely powering down, the device passes network state and notice of sleep to the proxy device. 3 4 On behalf of the device, the proxy maintains full network presence. If the proxy receives a packet that requires device wakeup, it signals to the device to awaken. 5 Once the device has woken up, the proxy passes the network state back to it and normal network operation is resumed.

Several different types of proxying are suggested in Nordman’s document: Self-Proxying: Where the proxy exists as part of one of the device’s components (typically its NIC) and is controlled under the same operating system. Power would remain supplied to the proxy component whilst all of the workstation’s other components would remain off. Switch-Proxying: Where the proxy exists as part of a network switch’s port that the device is connected to. Nordman suggests that the mobility of connected devices may pose an issue to how proxying would be implemented [26]. Existing “Wake-On-LAN” requests may be utilised in its operation. Third Party Proxying: Where the proxy exists in a third party device such as another workstation on the network or even a dedicated device.

Proxying still appears to be in its conceptual form as an addition to Ethernet. Although planned to be become a requirement of EnergyStar compliant devices [27], many of the proxying processes’ procedures have yet to be defined. In particular, its authors acknowledge that problems may arise in the implementation of switch proxying and consider third party proxying as outside the scope of their paper due to its complexity. They also concede that the fifth stage of the process still lacks the definitions for the proxy’s role after the host device has woken up.

Page 19

Proxying as a concept is certainly a fascinating idea, as the network infrastructure of the Internet currently has no conception of the power states of devices connected to it. The ability to place entire racks of servers into sleep mode until required would considerably reduce the amount of electricity consumed in data centres, for example. However, despite its potential to be furnished in future Ethernet hosts, its lack of maturity (and lack of a published standard) make it an unrealisable method to save power in computer networks in its present form.

2.4 Conclusions
Along with examining how typical host workstations represent their power states, an exploration of CISCO’s EnergyWise shows exactly how the future of network power management may look. Several initiatives and technologies aiming to make IT equipment more efficient have been present for many years. EnergyStar has been a very successful initiative encouraging manufacturers to develop more energy efficient equipment. Wake-on-LAN technologies have also been used by network professionals for years to reduce the constant power draw of infrequently used PCs. However, much of the future progress towards making network links themselves more efficient is being fore-fronted by the IEEE 802.3az group. Despite a selection of papers exploring a range of interesting technologies, its very recent inception means that although standards yet exist for them. As such, it will likely be several years before their draft proposals are accepted by the IEEE for introduction in Ethernet devices.

Page 20

3

Progress

3.1 Basic Objective 1:
“Investigate the costs involved in maintaining the operation of a typical organisation's IT infrastructure”

3.1.1 Overview
It was decided early in the project to model the University’s IT infrastructure using an existing methodology on power consumption to model typical organisational costs. Meetings with representatives from the University have yielded useful data including a spreadsheet containing the number of client machines and printers distributed across various labs. A large, abstracted inter-campus diagram was provided along with model numbers of a selection of more “invisible” devices, however even IT Services lack a reliable inventory of switches or routers in use over the network. Estimates for number of switches were obtained for certain portions of the University, so perhaps narrowing the scope of the “organisation” to these portions would be a worthwhile pursuit. However, much of the work performed to date involved analysing existing studies regarding power consumption and seeking to adopt an author’s methodology for use in this project. Although performed as part of the literature review, this analysis shows the progress made towards this objective so far (and hence has been included below). A note on existing literature In terms of analysing how much power network devices use, there is little dedicated work. Rather, studies are carried out with much wider scope, generally focusing on the power consumption of an entire country, or in smaller cases, single data centre installations. A selection of relevant literature is evaluated below. However, although these studies suggested that although network infrastructure-specific devices and workstations did utilise a portion of a site’s power, they were by no means the main consumers of energy.

Page 21

3.1.2 Energy Consumption by Office and Telecommunications Equipment in Commercial Buildings
The 2002 publication on the energy consumption of office and telecommunications equipment by Roth et al. consists of a 211 page “bottom up” study which examines the Annual Electricity Consumption (AEC) of a range of equipment categories. Its broad scope encompasses many items of interest to this project including workstations, server computers and computer network equipment but also discusses the impact of workstation displays, printers, copiers and much more. Its AEC analysis found that computer network equipment used a comparatively small amount of electricity compared to other office equipment. Figure 3.1 shows that computer networks and their associated devices only use 6.4TW-h (terawatt hours) of the total 97TW-h consumed by office equipment in the year 2000 (about 6.5%).

Figure 3.1: AEC of Office Equipment

[1]

Figure 3.1 also shows the impact of workstations and servers have on energy consumption in an office environment (over 30% combined). As these devices are can be considered as endpoints on a network, their power consumption is also relevant to this project. Nevertheless, Roth et al. did identify network devices as an area worthy of further investigation, and devoted a section of their study to measuring their impact. Notably, Roth subdivides the area of computer network into distinct device types, taking hubs, switches (both LAN and WAN) and routers into consideration. Further to this, Cable Modem Termination Systems and Remote Access Servers were also shown to have an extremely small (1.25%) contribution to power consumption.

Page 22

AEC of Network Hubs

AEC = N× Pport × tOH
Roth’s methodology for measuring the AEC of hubs consisted of calculating a watt per port value for each device (Pport). An average power draw value for the entire device would be taken and then be divided by the number of ports present on the device. The resulting value would allow larger capacity devices which used more power to be compared fairly to smaller, less energy consuming ones. Despite this attempt to levelling the playing field, Roth’s findings showed that larger capacity devices used a lower amount of power per port†. In order to calculate the AEC of the hub, a generous value for Pport was used (1.25W) to account for the variety of hub models deployed across the country. This value was then multiplied by industry estimates for the number of ports installed in all commercial buildings in the United States (N) and multiplied by the number of hours in operation per year (tOH). In his calculations, Roth realised the necessity of computer networks being available at all times. As such, his operational hours are always taken to be the “always on” value of 8,760 hours per year. The resulting value of these calculations could be considered the AEC value of the all hubs in the country.

AEC of LAN Switches This methodology was used in a similar context for LAN switches also, and Roth’s findings showed that switches tended to use more power per port than hubs, with an average Power/Port value being taken as 4W. A similar outlook applied to Ethernet switches raises an interesting question. Spanning tree protocol allows for individual ports on a switch to be disabled or set to various “standby” modes. By only enabling the number of ports required by a network, it may be possible to reduce the power per port value of a device (and increase its overall efficiency). Although not covered by this project, controlling the power consumption of a switch through its protocol configuration could be a worthy investigation.

†

84-port hub used 1.23W/port, whilst a 96-port hub used only 1.13W/port. Both hubs were from the same manufacturer.

Page 23

AEC of Routers Roth used a different methodology for calculating the AEC of routers. As these devices do not generally have as many ports as switches and routers, a power per port value would be misrepresentative. Instead, he simply considered an average power draw for a typical router (taken as 40W) and multiplied that by an estimated number of routers in operation and the same constant tOH value of 8,760.

Findings

Figure 3.2: AEC of Network Devices

[1]

Roth’s analysis of network devices identifies LAN switches as the largest consumers of electricity. Since the number of hub ports and switch ports is very similar (93.5 million hub ports [28] to a mean of 92,500,000 switch ports†) , the main reason for this is the fact that Roth’s investigations showed switch ports to use more than three times as much power as hub ports. One disadvantage of Roth’s report was the amount of estimation required in gathering an inventory of each type of device. Because the scope of the study was so huge (calculating AEC values for devices deployed across all of the United States), the margin of error in estimating amounts of devices would no doubt be considerable.

†

Studies showed a range of 90,000,000 in 1999/2000

[29]

to 95,000,000 (ADL Estimate based on

[29]

) switch ports in operation

Page 24

The sources Roth cited in his estimations also tended to be published comparatively far apart. His estimations for hub ports were based on a report carried out by Silva in 1998 whilst his switch port estimates were gathered over 1999 and 2000. It would be expected that a lot more hub ports would be installed over 1999 and 2000, something his AEC calculations should reflect. As a result, Figure 3.2 should show an increased proportion of power being consumed by hubs. Of additional concern, the power per port values calculated for all of these devices would be gathered from only one or two different models of device. This abstraction fails to represent the diversity of devices deployed across the country and as such his power per port values could be misrepresentative of the country’s actual average. However, Roth’s methodology would be extremely accurate if used in a context where exact inventory, power draw and model types of devices were known. This study also contains similarly useful measurements for considering the contribution of workstation PCs and servers and would allow a diverse analysis of the campus’s power usage to be made.

3.1.3 Case Study of Data Centres’ Energy Performance
In 2005, Sun & Lee examined in detail the power consumption of two data centres and found them to be high energy consuming facilities. Interestingly, they noted that the energy requirements of data centre floor space (per m2) could exceed that of traditional commercial office space by eighteen times†. Sun & Lee’s study differed from Roth’s considerably, most notably the devices examined were abstracted considerably more. Also present was a more detailed examination of lighting circuits and heating, ventilation and air conditioning (HVAC) systems which were only touched on in Roth’s study. They considered the data centre to have four broad areas where power was consumed:

†

Commercial office space typically measured 50W/m to 110/m while data centre power demand had a much 2 2 [30] wider range of 120W/m to 940W/m .

2

2

Page 25

Area IT Equipment

UPS Loss

HVAC

Lighting

Description Defined by Sun & Lee as “servers, data storage, network devices, monitors, etc” [30], the devices which provide the actual services to users. The wastage of power directed into the Uninterruptable Power Supply. Whilst giving power redundancy to all devices they are connected to, UPS efficiency varies greatly. As IT Equipment generates a lot of heat in its operation, facilities are required to regulate the temperature of a room and ensure its proper ventilation. The energy impact of overhead lighting used by staff in the data centres.
Table 3.1 Sun & Lee’s Device Criteria

This study also asserts the conception of “always on” computing, with both data centres showing that their IT equipment (along with supporting HVAC and UPS devices) were kept powered on 24 hours a day, seven days a week [30]. Only lighting equipment was powered on or off on a scheduled basis, dictated by its occupancy. The major findings uncovered from Sun & Lee’s study was that supporting the operation of IT Equipment often consumed more energy than the IT Equipment itself. Figure 3.3 shows a breakdown of the entire energy consumption of a data centre that they examined. From this we can see that just over a quarter of data centre power usage supplies the devices themselves, and that the remainder of energy is used in providing stable operating conditions for devices (HVAC),visibility for users (Lighting) and redundancy of the power supply in case of failure (UPS). Sun & Lee made several recommendations on how to reduce energy expenditure in data centres. Most of these suggested the reconfiguration of the support services to be more efficient. They reinforced the necessity of keeping the ratio of support services cost to IT Equipment cost as low as possible by saying “Generally, a larger contribution from the IT equipment to the total energy use indicates a better overall energy performance” [30]. No recommendations were made on limiting the operation of IT equipment (such as turning off or suspending workstations when not in use).

Page 26

Findings

Figure 3.3: Breakdown of energy use of a data centre

Several criticisms can be made of Sun & Lee’s study which could question the reliability of their conclusions. The first item of note is that both data centres were located in Singapore. Due to the tropical nature of the temperatures in this country, coupled with the fact that their study was carried out in the middle of summer, it could be suggested that HVAC systems would be much more abundant in this country (and more under load at this time) to keep the temperature of the data centres at an operable level. For example, a similar data centre located in a more temperate region could see the HVAC contribution decreased. Sun & Lee’s methodology in measuring power consumption between the two data centres could also be seen as slightly misleading. The metrics for graphing both centres’ energy use failed to take the floor space of the facility into consideration. Considering that data centre 1’s total floor space was 97m2 and data centre 2’s floor plan was more than ten times that at 1048m2, different HVAC and UPS requirements for larger premises might explain the seeming “inefficiency” of data centre 2’s graph. Sun & Lee also make no differentiation between network infrastructure devices and workstation computers. Indeed judging by their definition, even printers, monitors, projectors and scanners could be included in the final energy measurements. This meant that no real conclusions regarding the power consumption of network devices could be made.

Page 27

It would be interesting to see Sun & Lee’s methodology put to use in a larger range of data centres. In this report, only two facilities were investigated. However if the trend of IT Equipment’s energy consumption being over shadowed by support services was as consistently high as shown in Figure 3.3, then an investigation into increasing HVAC and UPS efficiency would be a worthy task.

3.1.4 Conclusions & Chosen Methodology
Roth’s methodology for AEC calculation of Office and Telecommunications Equipment has been identified as the best choice here. It offers the flexibility to analyse a range of devices present on a network and includes a detailed examination of network infrastructure devices. Roth’s methodology, although broad in scope, offers the potential to be scaled down for the purposes of this project. With its heavy bent towards estimation, it would also be forgiving if complete data for the university could not be obtained.

3.2 Basic Objective 2:
“Investigate and calculate the theoretical power requirements of data transmission.”

Not as much progress has been completed as desired. Most of the initial reading performed was aimed towards completing the literature review study, and most of these papers read did not go into enough detail to allow for theoretical calculations of data transmission. A high priority for this objective is to find a suitable textbook which would provide enough instruction for the completion of the objective. Consultations with project supervisors are hoped to yield advice on this matter. Ideas for further study: Data encoding schemes and the power requirements of each. o o o Non-return to zero (NRI) encoding and its variants Manchester encoding and its variants Implementations of Pulse Amplitude Modulation (i.e. PAM-5)

Page 28

3.3 Basic Objective 3:
“Observe and measure the power consumption of devices in a typical network, both under load and whilst idle.”

3.3.1 Experiment 1: Measuring Power Usage of Various S-states
Justification The desktop computer is one of the most ubiquitous elements of an organisation’s network. Indeed, the computer networks were originally developed to facilitate the sharing of information between computers. As the computer can be considered as a network device itself (referred to as “building blocks” [31] of a network), an investigation of the power consumed by an average machine is of interest when gauging an entire organisation’s power usage. Several studies [1] [12] have made noted that computer systems do not draw a steady level of power over time; instead it has been shown to fluctuate depending on the power-mode of the system. As such, the following experiment aims to measure the power drawn by an average PC over time with an aim to show that S-modes† draw progressively less power as they are increased.

Figure 3.4: Setup of Experiment 1

†

Of the ACPI model (discussed in section 2.1.1)

Page 29

Methodology The following computer system’s S-state was altered and its components remained constant throughout the experiment. A specification sheet for the Optiplex 775 (Mini-tower) is included as Appendix 1. The main item of interest to this experiment is the output of the PSU, rated at 305W. The device used to measure the power draw of the system was a Maplin “Plug-In Mains Power & Energy Monitor” (hereafter referred to as the “power monitor”). It features a power measurement mode which updates every second with readings in Watts. The following S-states were altered: S0 (active-processing): Achieved by putting as much load on the CPU as possible. Ideally a load of 95-100% would be used to gather measurements from. This was achieved by running several high-strain tasks on the system at once†. S0 (active-idle) – Achieved by ensuring that the system had fully booted up and was performing nothing more than regular housekeeping tasks. Ideally a CPU load of 0-5% would be used to gather measurements from. S3 (standby) – Achieved by using the “sleep” function available within the operating system. S5 (soft-off) – Achieved by using the “shutdown” function available within the operating system. The system used had no facility to measure the S4 (hibernate) state, and the “S6”/G3 (mechanical off) state was omitted due to the fact its results would always equate to 0W.

Procedure 1 The power monitor was positioned between the plug of the computer system and mains power supply and set to the Watts monitoring mode. 2 The system was placed into the appropriate S-mode, and sixty seconds were allowed to elapse to allow readings to stabilise and ensure that transitions between S-modes were not still underway. 3 Measurements were taken from the power monitor at intervals of 30 seconds for a total of two minutes.

†

25 instances of simultaneous high resolution video/audio playback were used.

Page 30

Results Power Mode
S0 (active-idle) S3 (standby) S5 (soft-off)

t
85W 44W 4W 3W

0 84W 46W 3W 3W

30 88W 44W 3W 2W

60 86W 45W 4W 2W

90 87W 45W 3W 2W

120

S0 (active-processing)

Table 3.2: Results of Experiment 1

Figure 3.5: Results of Experiment 1

Discussion The results of the experiment confirm what was presumed about S-levels. Powering the system down into S5 and S3 expectedly consumed less power than leaving a machine in its S5 state. The experiment also proved the presence of the “phantom load” phenomena. In S5 mode, a small amount of power was still drawn from the supply despite the fact that the device’s power button had been pressed and was presumed to be off. It was also shown that the amount of power a system can draw whilst under load can be from 183% to 200% more than whilst idle.

Page 31

Perhaps most surprisingly, the experiment showed that the difference between S3 and S5 state was miniscule, often only being a single Watt higher. This suggests that the S3 sleep mode provided with modern operating systems is indeed a viable alternative to placing the machine in S5 state.

Conclusions The results of the experiment show that S3 and S5 S-state offer large power saving opportunities for network devices. Enterprises stand to conserve considerable amounts of energy by placing devices into higher S-states whenever their use isn’t required. The implementation of power management policies on individual machines (for example, powering down into S3 state after 30 minutes of inactivity) and the use of technologies which could place devices into these states remotely could aid in providing maximum savings. The results also suggest that the differences between leaving a computer in S3 state and S5 state are negligible.

Further Notes The system used for preliminary measurements cannot be considered as representative of an average computer. The specifications on the test machine are far in excess of most other machines available in the University and as such would be expected to draw far more power than more common lab machines. Also of note was a lack of readings for the S4 power state. In a future repetition of this experiment it would be desirable to locate a machine which could achieve this. The large time intervals at which power readings were taken do not give a high enough resolution to see the true fluctuation of power draw from the system. In future, it is proposed to take measurements at shorter intervals. Possible values for this could be five seconds, or even the power monitor’s maximum refresh rate of one second.

Page 32

3.3.2 Experiment 2: Data Load’s Impact on Power Consumption of a Network Switch
Justification In order to enable communication between large numbers of IT devices, the use of Layer 2 and Layer 3 hubs and switches is required. However as noted by Coffman & Odlyzko [23], Ethernet links find themselves sitting idle most of the time.

Methodology

Figure 3.6: Setup of Experiment 2

Four computer systems identical to the Optiplex 755 were connected to a Netgear WGR614 home router with its wireless access capability turned off. This resulted in the device basically operating like a small four-port Ethernet switch, configurable via a web browser on a host machine. A data sheet for the router is included as Appendix 2. Two of the computer systems would be designated “servers” and would host a large DVD image file on their hard drives. The other two systems would be designated clients machines.

Page 33

The variable factor in the experiment would be the “load state” of the router. Two different states exist: Idle: The four hosts connected to the switch would be performing no communications with the switch or to each other. Under Load: The two client machines are instructed to download the image files from the server machines. Large amounts of data are transferred through the switch, maxing out the bandwidth pool of 100Mbps.

Procedure 1 The power monitor is placed between the plug of the switch and the mains power and set to the Watts monitoring mode. 2 The router is placed in its desired “load state”. For the “Under Load” state, thirty seconds are allowed to elapse to ensure the file transfer is fully under way. 3 Measurements were taken from the power monitor at intervals of 30 seconds for a total of two minutes. Results Load State Idle Under Load

t
1W 1W

0 1W 1W

30 1W 1W

60 1W 1W

90 1W 1W

120

Table 3.3: Results of Experiment 2

Discussion Sadly, the results of this experiment prove that the equipment sourced was inappropriate to draw any meaningful conclusions from. The Netgear WGR614, a small, inexpensive home router proved far too underpowered to gauge whether excessive data transmission actually had any effect on its power usage. In addition, the power meter proved inadequate for this task as it only measured power to the nearest Watt.

Page 34

For the next steps towards completing this objective, this experiment will have to be re-imagined. Either a switch or router which consumes more power will have to be located (such as a CISCO enterprise level switch), or a different measurement (such as kW-h consumption over an extended period of time) will have to be taken from this setup. In the latter case, challenges of maintaining an “under load” network state for a meaningful amount of time will become apparent. Even in this case, the suitability of such a low powered device is questionable. Further notes In addition to the above issues, similar concerns about the time interval measured at are raised. A smaller interval will be required to gain more meaningful results. However the main priority is to acquire a network device that will provide significant power readings to analyse.

3.4 Advanced Objectives
No progress has been completed on the project’s advanced objectives as the prerequisite basic objectives have not been completed.

Page 35

4

Summary & Completion Plan

Summary
Literature Review: Objective 1: Complete

Methodologies researched, resulting in decision to use Roth’s. Meetings with University Representatives were useful, but not all required data is available through them. Narrowing of scope may be a consideration.

Objective 2:

Very little progress completed. The location of appropriate texts is a high priority. Advice from supervisors will be sought.

Objective 3:

Two experiments have been preliminarily performed. Experiment 1: Minor refinements and higher resolution results required. Experiment 2: Suitable network device must be located for measurements.

Advanced Objectives: Awaiting completion of prerequisite basic objectives.

Completion Plan
The third management meeting for this project is scheduled to take place on Wednesday the 16th of December. At this meeting the concerns regarding certain objectives will be voiced, and hopefully suitable advice will be given that would help the project overcome these obstacles. The project is on schedule, with large amounts of work being completed on several of its objectives. If this rate of work is maintained then there should be nothing barring its completion for the due date of Week 12 of the second trimester.

Page 36

5
[1]

References

Roth et al. (2002)"Energy Consumption by Office and Telecommunications Equipment in Commercial Buildings Volume I: Energy Consumption Baseline" Arthur D. Little, Inc.
[2]

Energy Information Administration (2007) "Annual Electric Power Industry Report" (Table 7.4) Energy Information Administration
[3]

UK Department for Business Innovation & Skills (2005) "Trends in energy prices between 2003 and 2010" berr.gov.uk [Accessed 20:00, December 15th 2009 at http://www.berr.gov.uk/files/file16806.pdf]
[4]

Bessen, J. (2002) "Technology Adoption Costs and Productivity Growth: The Transition to Information Technology" (Fig. 1) Review of Economic Dynamics Volume 5 Issue 2 (2002) p.443-469
[5]

Hogbin, E.J. (2004) "ACPI: Advanced Configuration and Power Interface, Section 4: About ACPI" Linux.org [Accessed throughout Oct-Dec 2009 at: http://www.linux.org/docs/ldp/howto/ACPI-HOWTO/aboutacpi.html]

[6]

Roth & McKenney (2007) "Energy Consumption by Consumer Electronics in U.S Residences" (p.70) TIAX LLC.
[7] [8]

Visionsoft (2009) "Windows Shutdown and Wake On LAN Primer" Visionsoft

Hewlett-Packard Corporation et al. (2009) "Advanced Configuration and Power Interface Specification: Revision 4.0" acpi.info [Accessed 20:00, October 24th 2009 at http://www.acpi.info/DOWNLOADS/ACPIspec40.pdf]
[9]

Meier & Siderius (2006) "Regulating Standby" (p.9-225) ACEEE Summer Study on Energy Efficiency in Buildings

[10]

Harris & Cahill (2005) "Power Management for Stationary Machines in a Pervasive Computing Environment" (p.2) Proceedings of the 38th Hawaii International Conference on System Sciences (2005)
[11]

Lippis, N.J. (2009) "Controlling Corporate Energy Consumption via the Enterprise Network" (p.2-7, Table 1) Lippis Enterprises, Inc.
[12]

Roberson et al. (2002) "Energy Use and Power Levels in New Monitors and Personal Computers" (p.2, Table 9) University of California
[13] [14]

EnergyStar (2006) "2006 Annual Report" (Table 10, Table 1, Figure 1) United States Environmental Protection Agency EnergyStar (2008) "ENERGY STAR Program Requirements for Computers: Version 5.0" (p.9) energystar.gov

[Accessed 16:00, 11th November 2009 at: http://www.energystar.gov/ia/partners/prod_development/revisions/downloads/computer/Version5.0_Computer_Spec.pdf]
[15]

McWhinney et al. (2005) "ENERGY STAR product specification development framework: using data and analysis to make program decisions" (p.1623) Energy Policy 33 (2005) 1613–1625
[16]

McCabe, K. (2009) "Amendment to IEEE 802.3 Standard Enhances Power Management and Increases Available Power" ieee.org [Accessed 17:00 November 16th 2009 at: http://standards.ieee.org/announcements/stdbd_approves_ieee802.3at.html]
[17] [18]

CISCO Systems (2008) "Power over Ethernet (PoE) Power Requirements FAQ, Document ID: 97869" (p.2) CISCO Systems

3Com Corporation (2002) "Power Over Ethernet: Current State of the Technology and the IEEE Standard" (p.3-4) 3Com Corporation
[19] [20]

Morgan, T (2006) "Power-Over-Ethernet: The Reality of Designing A Powered Device" (p.3) AMD (1995) "Magic Packet Technology" (p.1-2) AMD Publication #: 20213

Page 37

[21]

Berger, M (2008) "Mobile Manageability in Low-Power and Operating-System-Absent States" Intel Technology Journal, Volume 12, Issue 4 (2008)
[22] [23] [24]

Bennett et al. (2006) "Improving the Energy Efficiency of Ethernet: Adaptive Link Rate Proposal" ethernet alliance Coffman & Odlyzko (2001) "Internet growth: Is there a “Moore’s Law” for data traffic?" (p.34) AT&T Labs

Nedevschi et al. (2008) "Reducing Network Energy Consumption via Sleeping and Rate-Adaptation" (p.1) University of California
[25]

Blanquicet, F. (2008) "PAUSE Power Cycle: A New Backwards Compatible Method to Reduce Energy Use of Ethernet Switches" (p.2-11, Figure 1) ethernet alliance
[26]

Nordman et al. (2007) "Improving the Energy Efficiency of Ethernet-Connected: A Proposal for Proxying" (p.4-8, Figure 2) ethernet alliance
[27] [28] [29] [30]

EnergyStar (2008) "ENERGY STAR Program Requirements for Computers: Version 5.0 - DRAFT 2" (p.9) EnergyStar Silva, E. (1998), “1998 LAN Hub Forecast, 1997-2002: The Switch Migration” IDC Report #17673. Dahlquist & Borovick (2000) “Worldwide LAN Switch Market Forecast and Analysis, 2000-2004” IDC Report #22925.

Sun & Lee (2005) "Case study of data centers’ energy performance" (p.522-531) Energy and Buildings 38 (2006) p.522–533
[31]

CISCO Systems (2005) "CCNA 1 and 2: Companion Guide (Cisco Networking Academy Program): Third Edition" (p.6) CISCO Press

Page 38

Appendix 1: Dell Optiplex 755 Datasheet

Page 39

Appendix 2: Netgear WGR614v9 Datasheet

Page 40