Ethernet White Paper

8/8/2019 Ethernet White Paper

1/14

The Field Engineers Challenge - rom LAN to Carrier Ethernet

Carrier EthernetFrom its modest beginning in the early 1970s as a way to connect computers in a

data center, Ethernet has evolved over the years into the dominant networking

technology. Today it is used in a wide range o environments, rom home networks

connecting a ew computers to large enterprise networks connecting tens o

thousands o nodes. While it was originally designed as a local area networking

technology, it is now increasingly used by telecommunications carriers as an

alternative to traditional circuit switched networks over long distances. Carrier

Ethernet reers to the extensions necessary in order to allow telecommunications

carriers to deploy and provide Ethernet services to their customers.

The ollowing are the main reasons behind the adoption o Ethernet in

carrier networks:

Ethernet is everywhere. Retail as well as corporate customers are already using

Ethernet. By adopting Ethernet in the backbone, the same technology can be used

rom end-to-end, thus eliminating the need or interace or rame conversion.

Cost eectiveness. Due to economies o scale, Ethernet components are inexpensive,

thus lowering both the capital and operational expenses.

Increased fexibility. Unlike traditional point-to-point circuits that come in discretespeeds (T1/E1, T3/E3, OC-3/STM-1), Ethernet allows subscribers to add bandwidth

more incrementally and more quickly. Typically, bandwidth upgrades only involve

confguration changes and can take eect in days rather than weeks.

To promote the use o carrier Ethernet and to increase interoperability between

vendors, the Metro Ethernet Forum (MEF), an industry consortium, produces technical

specifcations or Ethernet services in metropolitan areas as well as in the wide area.

In addition to extending Ethernet service to the customers, carriers and service

providers are also building and converting their own core networks to Ethernet.

For example, more and more mobile phone operators and DSL providers are

using Ethernet as the back-haul technology instead o the traditional TDM orATM network[1].

QoSOver the years, Ethernet has evolved rom its original best-eort, CSMA/CD based

data communications technology to a QoS (Quality o Service) enabled service

capable o supporting multi-media trafc in a converged network. As an example,

MEF uses the concept o a bandwidth profle to describe the subscribers trafc.

It specifes the average rate o committed and excess trafc that a subscriber can

302 Enzo Drive

San Jose, CA 95138

Tel +1 408-363-8000

Fax +1 408-363-8313

www.sunrisetelecom.com


2/14

2


generate into the providers network, and the associated perormance objectives in terms

o the delay, rame loss and availability. Thus each trafc profle consists o the ollowing

parameters:

CIR(CommittedInformationRate)

CBS(CommittedBurstSize)

EIR(ExcessInformationRate)

EIS(ExcessInformationSize)

Service rames up to the committed rate are considered in-profle and are delivered

as per the perormance objectives. Those sent up to the excess rate are allowed into the

providers network but are considered out-o-profle and are delivered without any service

perormance objectives. Trafc sent above the excess rate is discarded. Thus subscribers

who have dierent bandwidth profles will be accorded a dierent level o service. [2]

Field Testing RequirementsIn an increasingly competitive environment, carriers and service providers must be able to

provision circuits quickly and cost-eectively, while at the same time ensuring that the

service meets or exceeds the Service Level Agreement (SLA). This requires engineers and

technicians to be able to accurately and efciently test circuits in the feld. Field testing

generally alls into a ew categories basic connectivity testing and service verifcation;

perormance testing and QoS testing.

To acilitate feld testing, portability o the test equipment becomes an important

consideration. Thus instead o chassis based units, hands-held devices are commonly

used. In recent years, these units have become widely available and aordable.

The SunLite GigE rom Sunrise Telecom is one such example and will be used to illustrate

the tests described in this white paper.

Basic Connectivity Testing and Service Verifcation

Whether it is a new customer or an existing customer ordering a new circuit, the feldtechnician must ensure that the circuit is operational beore it is turned-up. This type

o testing usually consists o the ollowing:

Linkcheckconrmthelinkisactiveandsuccessfullynegotiatedaspeedand

duplex setting

DHCPconrmthattheCPEcanobtainavalidIPaddress

PINGconrmthatthereisIPconnectivity

TracerouteconrmthatIPtrafcistakingthedesiredroute

These tests are also commonly used to troubleshoot an Ethernet network. (See sidebar on

Ethernet Troubleshooting)AdditionaltestssuchasloadingawebpageusingHTTPor

downloadingaleusingFTParealsooftenperformedinordertoverifythatInternetconnectivity is established.

PerformanceTestingBasic connectivity testing is generally sufcient or best-eort service such as residential

Internet access which has no implicit perormance guarantees. For corporate customers

who require services with specifc perormance objectives, it is common to employ the

RFC 2544 tests.


3/14

3


1In addition to these our, RFC 2544 also defnes the system recovery test and the reset test which measure thespeed at which a DUT recovers rom an overload condition and a device reset. These are not used as much mainlybecause these metrics are typically not part o the carrier SLA. Furthermore, these two tests are also harder toimplement in an automated test.

RFC 2544 [3], published in 1999 by the IETF, defnes the Benchmarking Methodology or

Network Interconnect Devices. It was originally designed to allow the standardized testing

and benchmarking o a single interconnect device such as a router or a switch (known as

the DUT or Device Under Test). This methodology has become the de acto standard

perormed routinely in QA labs and verifcation labs in order to quantiy the perormance

o network devices. The ollowing diagram shows the classical test setup.

In this setup, a test system is responsible or injecting trafc into the DUT and then

analyzing the rames coming out. By correlating the input and output, the test system can

measure the transmission characteristics o the DUT.

The RFC 2544 benchmarking methodology defnes a series o perormance measurements.The terminology and metrics used in the test suite is defned in [4]. The ollowing are the

our benchmark tests that are most commonly implemented in an automated test suite1 :

Throughput This test measures the highest data rate at which none o the rames are

dropped by the DUT. The ollowing rame sizes in bytes should be used during the test 64,

128, 256, 512, 1024, 1280 and 1518.

Latency Once the throughput has been determined, this test measures the one-way delay

through the DUT at the throughput rate.

Frame loss This test measures the rame loss rate throughout the entire range o input

data rates and rame sizes.

Back-to-back This test measures the longest rames burst (rames transmitted with

Figure 1 - Standard RFC 2544 Setup or Device Testing

Frame Loss =(InputCount-OutputCount) 100%

InputCount

09CL-00074B


4/14

4


the minimum inter-rame gaps in between) that the DUT can handle without the loss o

any rames.

In recent years, the RFC 2544 test suite has been adapted to measure the perormance o

systems and networks. Thus instead o measuring the ability o a switch or router to pass

rames, we can now use the same methodology to quantiy the transmission characteristics

o a virtual circuit or a network. The ollowing shows two examples. The frst one shows an

example o what MEF defnes as the Ethernet Line (E-Line) service, which is essentially anemulation o a virtual circuit. The second example is an Ethernet LAN (E-LAN) service, which

emulates a multipoint network.

09CL-00075B

Figure 2 - Testing an Ethernet Virtual Connection

09CL-00076B

Figure 3 - Testing an Emulated LAN Service


5/14

5


Adapting RFC 2544 rom testing a DUT in the lab to a production network in the feld

requires a ew changes:

Thetestsystemusedinthetestlabconnectsportsthatarephysicallyclose

to each other. Typically, the same test system transmits trafc rom one

set o ports and receives trafc rom another set o ports. In a network,

the two measurement points are geographically separated. Thus two or

more test units are used to measure the input and the output. In otherwords, one unit generates the trafc and the other unit receives the trafc.

Inthelab,thetestsystemisstationary.Intheeld,thetestunitmustbe

portable and can be deployed quickly in various locations.

RFCs2544and1242denelatencyintermsoftheone-waydelay.Ina

network where two or more test units are required to perorm the test,

round-trip delays are measured. Measuring one-way delay would require the

test units to be time-synchronized.

The ollowing are some sample results o a RFC2544 throughput test o a Gigabit switch:

TESTMODE:THROUGHPUTTEST

The automated test stepped through all the prescribed rame sizes testing each at the line rate.

In this test, the switch was able to pass through wire rate trafc or all rame sizes.

SPEEDMODE:1000FULL

FLOWCONTROL : OFF

FLOWCONTROL : OFF

PREAMBLESIZE:8

DATAPATTERN:RANDOM

TX RATE(INIT) : 60%

TX RATE(FINAL) : 100%

TX RATE(DELTA) : 10%

TXPACKETS:1000064 BYTES :

MAX RATE : 100%

FRAME LOSS : 0

PASS/FAIL:PASS

128 BYTES :

MAX RATE : 100%

FRAME LOSS : 0

PASS/FAIL:PASS

256 BYTES :

MAX RATE : 100%

FRAME LOSS : 0

PASS/FAIL:PASS

512 BYTES :

MAX RATE : 100%

FRAME LOSS : 0

PASS/FAIL:PASS

768 BYTES :

MAX RATE : 100%

FRAME LOSS : 0

PASS/FAIL:PASS

1024 BYTES :MAX RATE : 100%

FRAME LOSS : 0

PASS/FAIL:PASS

1280 BYTES :

MAX RATE : 100%

FRAME LOSS : 0

PASS/FAIL:PASS

1518 BYTES :

MAX RATE : 100%

FRAME LOSS : 0

PASS/FAIL:PASS

------- END -------


6/14

6


09CL-00077B

09CL-00078B

The ollowing shows the two test conigurations or measuring the downstream and

upstream perormance.

Figure 4 - Testing a DSL Access Network

Figure 5 - Measuring Asymmetric Throughput

Asymmetric TestingIn feld testing, another common test scenario is the broadband access network with

asymmetric bandwidth. In ADSL, or example, it is important to test the two directions

(upstream and downstream) separately. This is illustrated in the ollowing diagram.

Furthermore, by inserting the test units at various points o the upstream network, we candetermine the throughput o the regional as well as access networks.

Field Testing o QoS Enabled NetworkTodays multi-service networks must support a variety o trafc such as data, voice

and video. Corporate clients also have SLAs with stringent perormance requirements.

For example, some carriers oer a gold, silver and bronze service, each with a dierent

set o service objectives. Thus increasingly, the network inrastructure must be able

to dierentiate and prioritize trafc. This is usually done using one or more o the

ollowing mechanisms.


7/14

7


2The 802.1p header is an 802.1Q header with the VLAN ID set to 0.

PhysicalPort. Trafc can be given dierent priorities simply based on the originating port.

In a private network, the physical port may correspond to, say, the voice VLAN where all the

IPphonesare.Inacarriernetwork,eachportmaycorrespondtoadifferentclientwitha

dierent SLA.

802.1p. Within a switched network, Ethernet rames can be tagged with an additional

802.1p header.2 This provides QoS at the MAC level. There are three priority bits (reerred

toasPriorityCodePointorPCP)whichindicatethepriorityoftheframe.Thisissometimesreerred to the Class o Service (CoS).

IPTOSorDiffserv.ThenetworkcanalsoprovideQoSattheIPlayerusingtheTOS(Typeof

Service) or Diserv (Dierentiated Services). Unlike the 802.1p bits which are usually set by

the end system, the TOS and Diserv bits are usually assigned by the router based on either

theIPaddressofthesourceorthetypeoftrafc(TCPorUDPportnumbers).

VLAN. In an aggregation network, all the subscribers within the same service class can be

assigned to a unique VLAN. For example, in a DSL network, a DSLAM can assign all the Gold

service subscribers to the same VLAN. The backbone network can then prioritize trafc based

on the VLAN ID. This VLAN tag is usually reerred to as the S-VID or the provider VLAN. In

order to preserve the customers own VLAN designation (C-VID), Q-in-Q is oten used.

TestingPrioritizationRegardless o the prioritization scheme, one must then be able to confrm that indeed the

network is prioritizing trafc in accordance with the service level objectives. Testing QoS

requires the ability to inject streams o trafc that will be accorded dierent treatment by

the network and then measure their respective perormance. The ollowing diagrams

illustrate a ew scenarios.

09CL-00079B

Figure 6 - Prioritization Based on Physical Port

In this scenario, prioritization is done based on the physical port. For example, all trafc

originating rom port 1 o the ingress switch might be given a higher priority (Gold

service). On the receiving side, by monitoring the two trafc streams, we can confrm that

when there is congestion (e.g., when both streams are transmitting at over 50% o the line

rate), more packets rom stream 2 will be dropped compared to stream 1.


8/14

8


3These ew diagrams show a separate port and a separate test unit used to generate the background trafc. Thisis only necessary i the ingress and egress ports have the same speed and thus additional trafc is needed toartifcially create a congestion condition.

In this test setup, 3 streams o trafc are injected into the network each with a dierent

PCP(0beingthelowestand7thehighestpriority).Onthereceivingside,thetestset

monitors each stream to look or impairments such as packet loss. In order to veriy that

trafc is being prioritized, there must be congestion on the output port, thus orcing the

network to decide which packets to drop. This congestion can be simulated i necessary by

injecting some additional background trafc rom another port.3

09CL-00080B

Figure 7 - Prioritization Based on 802.1p

ThenexttwoscenariosillustratesimilarsetupsforverifyingIPprioritizationaswellas

prioritization based on VLAN.

09CL-00081B

Figure 8 - Prioritization Based on TOS or DiServ


9/14

9


09CL-00082B

Figure 9 - Prioritization Based on VLAN

The last example shows an example o Q-in-Q. Each test packet has two 802.1Q tags, the

outer tag is the provider tag and the inner tag is the customer tag. In this example, aprovider tag o 100 might indicate a Gold service and 101 might indicate a Silver service.

Thus the test will determine i each provider tag indeed accords each class with a dierent

service level.

Using the SunLite GigE, one can construct multiple simultaneous packet streams and

perorm a BERT Throughput test. In the example below, our streams are enabled that

correspond to our dierent bandwidth profles. Each packet is confgured with 2 VLAN tags

(Q-in-Q). The outer tag (S-VID) has a VLAN ID o 100 whereas the inner tag (C-VID) has a

VLAN ID o 200.

Troubleshooting EthernetWhat to do when the network does

not work? Here is where trouble-

shooting tools come in. In Ethernet,

there are a number o common

problems that can be identifed

quickly with the right tools. These

include the ollowing:

Cablingerror.TheLANcableisconnected to the wrong port.

Cablefault.Thereisaproblemwiththe connector resulting in an open

or short circuit.

Speedorduplexmismatch.TheNIC isconguredforaspeed/duplex

setting that is incompatible with

that o the switch port.

DHCPerror.Stationsareunable toobtainanIPaddressfromthe DHCPserver.

DuplicateIPaddress.Multiple devicesareconguredstatically withthesameIPaddress.

Routingproblem.Stationsareunable to reach destinations

outsidethelocalVLANeitherbecause the deault gateway is

down or misconfgured.

To diagnose these problems quickly, the feld personnel needs a number o troubleshooting tools that are now packaged in hand-held test units.

The ollowing are some o these common tools:

CabletesterThisfunctionchecksforanycablefaultssuchasopenandshortcircuits.UsingTDR(TimeDomainReectometry),thetestercanalsodetermine the length o the connected cable.

PortidentierThisfunctionactivatesanddeactivatestheEthernetportthuscausingtheportLEDontheswitchtoturnonandoffperiodically.Thisisusefulinidentifyingtheswitchportconnectedtoadevice.

LinkstatuscheckWhenthetestunitisconnectedtoaswitchport,thisfunctiondisplaysthelinkstatussuchasthespeed,duplexandowcontrol.

DHCPThisallowsatestunittoconrmthatanIPaddresscanbeacquiredsuccessfullyfromtheDHCPserver.

DevicescanThisfunctionscansarangeofIPaddresseslookingforactivedevices.

PINGThisstandardfunctionprobesanIPaddressusingICMP.Itreturnstheround-tripdelaybetweenthetestunitandthetargetdevice.

TracerouteThisstandardfunctiontracestheIPpathtoatargetaddressidentifyingalltheintermediaterouters.


10/14

10


Figure 10 - Enable our packet streams. Figure 11 - Each packet has a MAC, VLAN and IP header.

This stream will transmit at 10% line rate.

Figure 13 - Every transmitted rame was echoed back by the

responder at the other end - thereore no rame loss.

Figure 14 - This test also measures the

rame delay statistics.

Figure 12 - There is an outer tag and inner tag.

Ater confguring the streams, trafc was turned on. The ollowing screens show the test results:


11/14

11


JitterorPacketDelayVariationReal-timetrafcsuchasIPtelephonyorIPvideoissensitivenotonlytotheamountof

delay but also whether the amount o delay is relatively constant or varying. This measure

is reerred to as the packet jitter or packet delay variation4. To complicate matters, there are

many ways o calculating delay variations and they can yield dierent results. The ollowing

are a ew commonly used defnitions:

InRFC3393,theone-wayIPDV(IPdelayvariation)isdenedasthedifference

between the delays experienced by two consecutive packets rom the

same stream.5

InRFC4689,jitterisdenedastheabsolutevalueoftheIPDV.6

MEFdenesframejitterasthedifferencebetweenthemaximumframedelay

and the minimum rame delay within a group o samples.7

InITUY.1540,the2-pointPacketDelayVariationisdenedasthedifference

betweenthepackettransferdelayofthepacketandadenedreferenceIP

packet transer delay.8

TriplePlayFor real-time applications such as voice and video, network transmission attributes such

as packet delay, jitter and packet loss can adversely aect the quality o the reception. For

example,forIPtelephony,thefollowingarecommonlyrecommendedvaluesinorderto

maintain acceptable voice quality:

Packetdelay:Lessthan150msone-way(lessthan100msone-wayfortoll-

quality voice).

Jitter:Lessthan50ms

Packetloss:Lessthan1%

For triple play, one must be able to generate multiple data streams and veriy that voice

and video trafc is transmitted with the required perormance. In this test, the dierent

streamscanrepresentdata,voiceandvideo.Howthenetworktreatseachstreamdependsonthenetworkconguration.Insomenetworks,allIPphonesareassignedtoaseparate

voiceVLAN.Inothers,theIPphonesareprogrammedtotagalltrafcwithhigherpriority

using 802.1p. Thus in designing a test, one must take into consideration the confguration

o the network inrastructure in order to eectively simulate the trafc mix. The ollowing

diagram illustrates a typical setup.

4Packetjitterordelayvariationmeasuresthechangeinpacketdelay.Thisisnottobeconfusedwithvideojitterwhich happens when the picture jumps. Video jitter can be the result o packet loss and excessive packet jitter.

09CL-00083B

Figure 15 - Testing Triple Play


12/14

12


Future Requirements - Ethernet Service OAMAs carrier Ethernet becomes more widely deployed, an area which is evolving rapidly is

Ethernet Service OAM (Operation, Administration and Maintenance). The objective is to

provide carriers with a comprehensive set o tools to monitor the availability and

perormance o Ethernet services. There are a number o related eorts in progress:

IEEE802.3-2005(previously802.3ah)EthernetLinkOAM(Ethernetin

the First Mile)

IEEE802.1agEthernetServiceOAM

ITU-TY.1731OAMFunctionsandMechanismsforEthernetBasedNetworks

MEF17ServiceOAMRequirementsandFramework

It is expected that as these standards mature, the prescribed tests (such as connectivity

check, loopback, link trace and so on) will be implemented in the feld test sets.


13/14

13


ConclusionsEthernet has become the ubiquitous networking technology. Carriers and service

providers are under competitive pressure to provision service quickly and cost-eectively.

One o the challenges o the feld technician is to veriy the integrity o the service which

isbecomingmorecomplex.BasictestingusessimpleIPtoolssuchasDHCP,PINGand

traceroute to veriy connectivity. Increasingly, Ethernet service must also be measured

against the service level. RFC 2544 is the test suite that can be used to determine the

throughput, latency, rame loss and back-to-back. In a QoS enabled network, trafc may

be prioritized depending on the physical port, 802.1p, TOS or DiServ, or VLAN tag. To veriy

proper QoS handling, the feld engineer must be able to generate multiple packet streams

with dierent characteristics and measure the perormance o each stream separately.

For example, in a network that uses Q-in-Q, the packets must be confgured with at leasttwo VLAN headers to allow the aggregation switches to prioritize trafc using the outer tag.

Finally, to ensure that the network can handle real-time trafc properly, the packet delay

variation (which is not part o the RFC 2544 test) must also be included.


14/14

14


Reerences1 TR-101, Migration to Ethernet Based DSL Aggregation, Broadband Forum, April 2006.

2MEF10.1,EthernetServicesAttributesPhase2,MetroEthernetForum,November2006.

3 Bradner, S. and McQuaid, J., RFC 2544 Benchmarking Methodology or Network Inter-

connect Devices, IETF, March 1999.

4 Bradner, S., RFC 1242 Benchmarking Terminology or Network Interconnection Devices,

IETF, July 1991.

5Demichelis,C.andChimento,P.,RFC3393-IPPacketDelayVariationMetricforIPPer-formanceMetrics(IPPM),IETF,November2002.

6Poretsky,S.,Perser,J.,Erramilli,S.,andKhurana,S.,RFC4689TerminologyforBench-

marking Network-layer Trafc Control Mechanisms, IETF, October 2006.

7MEF6.1,EthernetServicesDenitionPhase2,MEF,April2008.

8ITU-TY.1540InternetProtocolDataCommunicationServiceIPPacketTransferand

AvailabilityPerformanceParameters,December2002.

For more inormation, please call:

1-800-701-5208 or 1-408-363-8000 or visit www.sunrisetelecom.com

Our e-mail: [email protected]

Documents

Ethernet White Paper