ATT-TP76440-002 Ethernet Testing using the Acterna Test Set
Contents
1. Texas 76102 2004 2005 AT amp T Knowledge Ventures All rights reserved 3 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 2 Reasons for Reissue Updates to this document will be highlighted in this section 3 Turn Up Testing Procedures and Expected Results This Section contains procedures for initial verification of equipment related for the installation of AT amp T LOCAL EXCHANGE companies Ethernet services Use of these procedures will help ensure a successful service hand off and reduce repeat troubles This document also covers testing procedures that are features of a variety of test sets Test and Turn up procedures covered in this document are those utilizing two Acterna Test Set FST 2802 Test Pads The test device must have the capability of operating as an Ethernet switch port on a circuit Ethernet switches are capable of interpreting Ethernet frames data therefore you can use the test set to ensure physical and data layer integrity by transmitting and analyzing Ethernet frames The Acterna FST 2802 Test Pad is already being used by AT amp T LOCAL EXCHANGE companies personnel in connection with the Installation and Maintenance of Central Office Ethernet switches and the deployment of AT amp T s current Gig E offering GigaMAN The following test will determine the physical characteristics of the Ethernet element Testing the Ethernet Element from end to end will gi2. 4 Bytes 802 3 Frame Destination Source 802 2 Preamble Address Address Length LLC User Data CRC 8 Bytes 6 Bytes 6 Bytes 2 Bytes 3 Bytes 46 1500 Bytes 4 Bytes 2004 2005 AT amp T Knowledge Ventures All rights reserved 28 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 VLAN Tagging Protocols A VLAN is a grouping of Ethernet ports independent of physical location Grouped ports create a virtual workgroup A VLAN allows containment of broadcast traffic A VLAN has the same attributes as a physical LAN LAN networks are commonly divided into workgroups connected via common backbones to form virtual LAN VLAN topologies VLANs enable efficient traffic separation provide better bandwidth utilization and alleviate scaling issues by logically segmenting a LAN infrastructure into different sub networks Each VLAN becomes an individual broadcast domain within the LAN Switches or routers control access to various VLANs within a LAN Grouped ports create a virtual workgroup Frames are switched only between ports within the same VLAN Broadcast traffic is contained to the domain of the VLAN smaller collision domain There are two common tagging protocols ISL from Cisco amp IEEE 802 1q Addition 4 byte tag creates baby giants 802 1 has persuaded 802 3 to increase the maximum frame size four bytes from 1518 to1522 802 3ac Ethernet v2 0 Frame with
3. Acterna payload If this result is non zero it indicates that frames are being lost on the circuit under test VLAN Frames A count of VLAN frames as defined in IEEE 802 p q This result only appears for Test Pads with the VLAN Tagging option Unicast Frames The number of unicast frames received since the last test restart Multicast Frames The number of multicast frames received since the last test restart Broadcast Frames The number of broadcast frames received since the last test restart 64 Byte Frames A count of frames with a length of 64 bytes 65 127 Byte Frames A count of frames with lengths between 65 and 127 bytes inclusive 128 255 Byte Frames A count of frames with lengths between 128 and 255 bytes inclusive 256 511 Byte Frames 512 1023 Byte Frames A count of frames with lengths between 256 and 511 bytes inclusive Since we chose a frame size of 256 all the frames should be counted here If you chose a different frame size the counts will fall into the appropriate bins A count of frames with lengths between 512 and 1023 bytes inclusive 1024 1518 Byte Frames A count of frames with lengths between 1024 and 1518 bytes 2004 2005 AT amp T Knowledge Ventures All rights reserved 8 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 inclusive 3C 5 Error Stats Expectations Th
4. Byte Frames 25786950 Frame Size Max 64 M 65 127 Byte Frames 0 y Rx Bits Sec Cur 999950112 D 128 255 Byte Frames Loop Loop Single Reset Svc Up Down FCS Error Disruption erminate 1G Ethernet d Remote Loop Up Successful Unit FST 2802 in LLB Mode Figure 8 User Interface with All Results OK 2004 2005 AT amp T Knowledge Ventures All rights reserved 23 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 i UIM KEE MONJ TERM KE EE E SETUP Wl RU E EE Ca E WEE E Ww hakati BRAIDD AILEAN R ARBARA GAARAA AutoNeg Off Flow Control On Load Const Ethernet Summary Ethernet emm RESULTS RESULTS ALL ALL f SUMMARY SUMMARY OK OK M T Loop Loop T Start Fi Single Lue ji Reset Sve Up Down Traffic FCS Error Disruption erminate 1G Ethernet o l 0 2004 2005 AT amp T Knowledge Ventures All rights reserved 24 Ethernet Testing using the Acterna Test Set June 29 2005 Link Stats results AT amp T Services Inc ATT TP76440 002 Issue 1 The Link Stats category lists link statistics such as the average frame rate peak frame rate and the number of frames transmitted etc To view Link Stats results set the result category to Link Stats The following is a list and a general explanation of all the Link Stats results Bold type indicates elaboration as to the relevance to this specific test procedure Table 3 Link Stats
5. Flow Control Pause frame functionality is independent at both ends this test should be performed at each end of the circuit with the far end looped back Perform on one side at a time 2004 2005 AT amp T Knowledge Ventures All rights reserved 19 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 To set up for a Pause Ramp Test In order to perform Ramp test we will need to establish a loop at one end of the Ethernet network This can be done in one of two ways ch Local Loop with the FST 2802 Personnel operating the FST 2802 at the far end can tap the LLB button on the user interface This will cause the FST 2802 to return all of the error free traffic that it receives The near end FST 2802 should receive all the traffic that itsends Any missing or errored traffic will be noted The time delta for the round trip is calculated Note the LLB button on the user interface in Figure 5 Remote Loop with the FST 2802 Personnel at the near end FST 2802 can issue a Loop Up command to the far end FST 2802 When the loop up command is confirmed and traffic is started the near end FST 2802 should receive all the traffic that it sends Any missing or errored traffic will be noted The time delta for the round trip is calculated Upon completion the near end FST 2802 operator can tap the loop down button Note the Loop Up Down buttons on the user interface in Figure 5 Tap Setup Tap t
6. Services Inc 925 823 4616 E Mail we2583 camail att com Authors Wing Eng Area Manager Network Planning amp Engineering AT amp T Services Inc 925 823 4616 E Mail we2583 camail att com Bruce E Jones Associate Director Network Planning amp Engineering AT amp T Services Inc 817 338 6266 E Mail bj3246 txmail att com 2004 2005 AT amp T Knowledge Ventures All rights reserved Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 Table of Contents 1 COPYRIGHT PAGE EEN 3 2 REASONS FOR REISSUE ege eege ENEE ENEE EE 4 3 TURN UP TESTING PROCEDURES AND EXPECTED RESULTS c ccsccscesesssessssseeseensees 4 3A ETHERNET LINK CONNECTIVITY TEST bBpoCEpupE 4 SB FRAME GENERATION ra a aa a a E aaa EE aE A INA 5 3G SIGNAL THROUGHPUT WEE 5 BCT AGONSIANE Load RRE 6 3C 2 Constant Load Test Expectations 2 ccccsccecescecessecseneeeceneeesnaeeseneeceeeeecsssteneseseneessnntees 6 3C 3 Link Stats Exvpechatons neen nnenn reene enee 6 3C 4 Link Counts Exvpechations reene 8 3C 5 Error Stats Expectations 22 ccccccccssscccsceceeececeeeeeeaeceeneeeceanetsaaesseneeceeneessnsseseseceeneesenees 9 3D END TO END TESTING ILLUSTRATION cccccccceccccesceeeeeeecececeesseaueeseseseeseuseaueeseeeeseueeanenseseeeeees 9 3E NOTE ON SENDING DELIBERATE ERRORS sisssssssisisrsssesreuriirinrustetukirinetnnnnrinesne tnra nnnnnnun nunne 10 SF TEST SET INTEROPERABIL
7. The Acterna 2802 must be configured to run an All 1s or All Os test pattern and the MAC address of the Sunrise Telecom module needs to be entered as the Destination MAC address for the 2802 Auto negotiation should be set for Disabled 2004 2005 AT amp T Knowledge Ventures All rights reserved 11 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 The Sunrise Telecom MTT 28 10 100Base T module or the SSMTT 29 Gigabit Ethernet module must be configured for Layer 2 with an All 1s or All Os test pattern and the MAC address of the 2802 needs to be entered as the Destination MAC Address of the Sunrise Telecom module Auto negotiation should be set for Disabled 3F 2 B Loopback testing If the Sunrise Telecom Ethernet module is set in a manual loop back mode the Acterna 2802 module can run clean to itself The Sunrise Telecom Ethernet modules cannot run clean through a manual loop on the Acterna 2802 The Acterna 2802 must not be able to retransmit the Sunrise payload 3F 3 Cable Test Devices The Westek Cable Test Device has been approved by the AT amp T LOCAL EXCHANGE companies for use in the end to end testing of electrical 8 wire cable used for Ethernet and other electrical circuits This device can be used to test for electrical faults open short grounds from one RJ45 end point to the other RJ45 end point The master device attaches to one end with the remote attaching to the
8. VLAN TAG added PREAM pom e VLAN TAG USER DATA 6 6 4 4 Octets 8 2 46 1500 Getting more Assistance Acterna Technical Assistance Center Toll Free 1 866 ACTERNA 1 866 228 3762 www acterna com 2004 2005 AT amp T Knowledge Ventures All rights reserved 29 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 5 Non Supported Features and Expectations 5A Auto Negotiation Auto Negotiation allows two devices at either end of the circuit to advertise and negotiate the link operation mode such as the speed of the link and the duplex configuration of half or full duplex to the highest common denominator AT amp T does not support this feature all network elements with the feature shall be turned off due to the following reasons 1 Standards are not ubiquitous in the industry nor is their technical support for this parameter setting 2 Manufacturers meet the lowest common denominator on interoperability between one another 3 The flow through provisioning systems do not have the capability to provision auto negotiation settings in an automated or remote manner 4 Some manufacturers cards on SONET ADMs do not have the capability to flexibly provision auto negotiation of a port to on from off 5B Flow Control The Flow Control feature allows the provider to throttle back the customer traffic by sending Pause Frames when the
9. below depicts how the FST 2802 makes the existence of errors very obvious The following screens and accompanying discussion provide more depth to the analysis Here again we do not seek to provide an explanation for every result category that the FST 2802 provides but instead hone in on what is most relevant to this application Figure 6 User Interface with Summary and LED results displayed I UIM Biel E MON te SETUP THR au AutoNeg Off Flow Control On Load Const Ethernet Ethernet Signal Present Sync Acquired Link Active Frame Detect VLAN Frame Detect erminate 1G Ethernet SES Ke E The most relevant results from this test procedure reside in 3 categories Link Stats Link Counts and Error Stats 2004 2005 AT amp T Knowledge Ventures All rights reserved 22 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 Figure 7 User Interface with Link Stats and Link Counts results displayed MON EG THRU ta e AutoNeg Off Flow Control On Load Const Ethernet Link Stats Ethernet Link Counts Total Util Avg 100 00 A Received Frames 25786950 A Total Util Cur 100 00 P Transmitted Frames Total Util Peak 100 00 PAUSE Frames 0 Frame Rate Avg 1 49E 006 VLAN Frames 25786950 Frame Rate Cur 1488021 Unicast Frames 25786950 Frame Rate Min 1488020 Multicast Frames 0 Frame Rate Peak 1488023 Broadcast Frames 0 Frame Size Min 64 64
10. bit error across a circuit once BERT pattern synchronization has been achieved This assures the technician s that the proper end locations are the ones that are synchronized The parallel for Ethernet testing is to send a FCS error However one of the inherent traits of Ethernet is that error frames are discarded by Ethernet aware interfaces In the case of the 10 100 and Gigabit services over SONET the Network Element interfaces are Ethernet aware Frames with deliberate FCS errors sent by the test set 2004 2005 AT amp T Knowledge Ventures All rights reserved 10 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 will be dropped by the locally connected Ethernet port Instead of seeing an FCS what the far end test set will see is that a frame was lost In contrast the DWDM application would behave differently The ports on the DWDM device handle traffic with complete transparency Error frames are passed through along with valid frames The far end test set would count an FCS error However since the test set is behaving as an Ethernet element if it were in loop back it would discard an error frame just as a switch would In this way the technique so often used in the TDM world can still be used for most Ethernet Transport applications Caution Testing of cables jumpers and frame facilities need to be performed from end to end without the attachment of active Ethernet devices Pe
11. bytes for tagged frames Not all ADM manufacturers support the transport of jumbo frames today AT amp T will not support this feature in the initial Ethernet over SONET offerings 1 Not all Ethernet switches are capable of generating Jumbo Frames 2 AT amp T will not be able to guarantee transport of Jumbo Frames Customers will need to turn off this feature 5G Layer 2 Protection Aggregation Spanning tree protocol and link aggregation protocol are protocols that can be run on the Layer 2 Network to prevent the formation of transmission loops On Layer 1 it can be used to view several independent circuits between two customer sites as a single logical circuit Transport of spanning tree protocol or link aggregation control protocol frames through the EoS network is not guaranteed or supported Customers may not be able to run spanning tree protocol or link aggregation protocol through AT amp T s network 2004 2005 AT amp T Knowledge Ventures All rights reserved 32 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 6 References For further information or electronic copies of this document and related information visit the internal AT amp T LOCAL EXCHANGE companies web site http ebiz AT amp T com commonsystems or http apex sbc com Drawings may be viewed on the AT amp T LOCAL EXCHANGE companies Internal web site http woodduck standarddrawings sbc cbc index
12. htm N Issue amp Date ATT TP76200 Network Equipment Building Systems NEBS ATT TP76300 Installation Guide within the Central Office ATT TP76400 Detail Engineer Requirements for the C O Higher Excluding SONET Interfaces Specification ATT TP76440 002 AT amp T Ethernet testing using the Acterna Test Set June 2005 ATT TP76412 003 AT amp T Cable Test Device Dec 2003 ATT TP76450 Common Systems Standards for the AT amp T May 2003 Communications Network ATT TP76450 001 Common Systems Checklist May 2003 ATT TP76450 002 Common Systems Product Exception and Evaluation May 2003 Request ATT TP76440 001 AT amp T Ethernet Testing using the Sunrise Test Set June 2005 2004 2005 AT amp T Knowledge Ventures All rights reserved 33 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 7 Contacts 7A AT amp T Contacts Mike Yeilding Area Manager Network Centralized Support Drawings amp Ethernet Standards 925 823 4747 E Mail my1515 camail att com Bruce E Jones Associate Director Network Planning amp Engineering Common Systems 817 338 6266 E Mail bj3246 txmail att com Dave Smith Area Manager Network Planning amp Engineering AT amp T Services Inc 818 713 7308 E Mail ds2375 camail att com Wing Eng Area Manager Network Planning amp Engineering AT amp T Services Inc 925 823 4616 E Mail we2583 camail att com 7B Strategic Supplier Contac
13. initiates a pause in transmiting adequate to clear the buffer This is a function of line rate and processor speed Burst In order to emulate realistic customer traffic patterns the customer the test set can be programmed to transmit in burst mode This provides a more realistic network simulation for circuit provisioning About the FST 2802 The FST 2802 is the ideal telecommunications solution for testing and verifying Ethernet network elements and services It is an additional module for the widely deployed TTC Acterna TestPad platform which includes the 2209 2310 2510 and more It is designed for field use by transport technicians involved with installation commissioning service turn up and troubleshooting of TDM SONET and now Ethernet Transport Services Figure 3 FST 2802 TestPad This application note is not intended to provide an exhaustive training on every aspect of the FST 2802 features or menu structure Instead the following instructions will hone in on specific applications surrounding services built on the 15454 and the G1000 4 line card A complete overview of the TestPad and the FST 2802 Module can be gained from the user manual as well as from context sensitive help screens embedded in the unit Test Procedures 2004 2005 AT amp T Knowledge Ventures All rights reserved 15 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 This section provides step
14. microseconds You must transmit an Acterna payload to measure round trip delay RTD measurements are only performed when using a Constant traffic type Check the service contract or SLA for an average RTD parameter If there is none a benchmark for the particular network is still recommended In general this value should be under 100 ms Average delays in excess of 100 ms might impair the performance of delay sensitive applications Delay Avg The average round trip delay calculated in microseconds You must transmit an Acterna payload to measure round trip delay RTD measurements are only performed when using a Constant traffic type Svc Disruption The service disruption time maximum inter frame gap when service switches to a protect line calculated in milli seconds 3C 4 Link Counts Expectations Test Result Description Received Frames A count of frames received since the last test restart including errored frames Transmitted Frames A count of frames transmitted since the last test restart RX Acterna Frames A count of received frames with Acterna test packets PAUSE Frames Out of Seq Frames A count of PAUSE frames received from a remote Ethernet device This should be 0 until using the Ramp Test When the bandwidth approaches the provisioned limit this counter should increment by at least 1 A count of out of sequence frames Only valid while receiving frames with the
15. other This device will tell the technician at either end the correct test and wiring configuration end to end This device is suitable for all Ethernet pre wiring before circuit placements are made Circuit testing will be made with the Acterna and Sunrise Telecom test sets The Westek device also provides talk battery in order for a hand set to be placed at each Cable Test Device which permits the remote technicians to converse with one another The general availability of this product will be December 2003 2004 2005 AT amp T Knowledge Ventures All rights reserved 12 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 4 Testing Job Aids using the Acterna Test Set ACTERNA Tools Needed e Service description for the circuit under test e g the bandwidth to be verified e Acterna FST 2802 TestPad with correct GBICs quantity 2 e Appropriate fiber optic cables duplex SC multi mode or single mode Figure 1 Customer sites with EOS Handoffs The Ethernet traffic is mapped into SONET sub rates for efficient transport across the SONET or DWDM core This is the enabler of the Fractional services alluded to above Table 1 below describes the maximum guaranteed Ethernet throughput related to the appropriate SONET mapping This mapping flexibility allows for e Right sizing the backbone circuit for the customer needs e Conservation of core bandwidth Avoid reser
16. unit OO Joo ob A 3C 2 Constant Load Test Expectations On the left results pane select the Result Category button Select Link Stats and record stats Average Bandwidth Received 100 Current Bandwidth 100 Shows a continual growing number of frames received and frames transmitted On the right results pane select the Result Category Button Select the Error Stats and record stats Symbol Errors 0 FCS Errored Frames 0 Runts 0 Undersized Frames 0 Oversized Frames 0 Errored Frames 0 3C 3 Link Stats Expectations The Link Stats category lists link statistics such as the average frame rate peak frame rate and the number of frames transmitted etc To view Link Stats results set the result category to Link Stats The following is a list and a general explanation of all the Link Stats results Bold type indicates elaboration as to the relevance to this specific test procedure 2004 2005 AT amp T Knowledge Ventures All rights reserved 6 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 Link Statistics Stats Test Expectations Benchmark Test Result Expectations Description Total Util Average The average bandwidth received expressed as a percentage of the entire 1 Gbps of available bandwidth The average is calculated over the time period elapsed since the last test restart Total Util Cur The current bandwidth received expr
17. 1000 Mbps capable Ethernet port to connect to a smaller SONET sub rate such as an STS3c 150 Mbps For any fractional rate service it is imperative that Flow Control be enabled on the customer equipment Ethernet Testing Concepts There are three main test criteria for verifying proper operation of the equipment and associated provisioning for the Gigabit Ethernet Service There is also a fourth test related to the SONET backbone Link Connectivity This involves verifying that the ports on either end of the circuit operate properly by establishing a local Ethernet link with the attached test equipment The FST 2802 is 2004 2005 AT amp T Knowledge Ventures All rights reserved 14 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 being used to emulate the customer s end device Additionally connectivity also refers to the two desired end points test set to test set across the SONET network having communication Throughput This involves verifying that the contracted throughput Mbps is delivered error free by the circuit under test In the case of sub rate Gigabit services this also involves verifying proper configuration and operation of the flow control features Pause This is a measurement to ensure that a device that is responds when incoming traffic is coming at too high of arate The Pause frame is transmitted to the device which is overwhelming the receiver and
18. ITY c0cccccccccsceeeeecececceeesseeeeesuseseuaeaueeseceeeeeeuaueaseeeeesesauaaeeseeeeeeeees 11 3F 1 Same Type Test Sets 0 ccccccccceecccessecseneeeeeeneesnaecseneeceeneessaessaneeseenseseeneeseaaeseeeeessensesaes 11 SF 2 Different Type Test S t ossosa inaina ea EEAS AEA SE ERAR 11 3F 3 Cable Test DEVICES seniii aaiae en e ad a aadi daaa dee 12 4 TESTING JOB AIDS USING THE ACTERNA TEST SET cccesscsccsssssseesesesesesssesesesnanans 13 5 NON SUPPORTED FEATURES AND EXPECTATIONG 0 ccccesscsccsssssseeseeuseeeasssesseseeeeans 30 BA AUTO NEGOTIATION o c cccccccescececceecsceseeeseceeeeseeseaueessesseuaueeaeuseeeeeseeaueusaeeeseeuseauaageeeesseuuaueseesess 30 SBS FLOW GONTROL EE 30 jC FAULT PROPAGATION a a a EE 31 SRAI E e110 e U dee 31 E PERFORMANCE META OO a a aa a ended E EE A 31 jA JUMBO PRAN G deed dee De 32 5G LAYER 2 PROTECTION AGGREGATION ssssssssssssssssusiseeettuniinisstt tnit kinan trnki nran En EEEE nnan nann nn nannan nnna 32 6 REFERENCES aiiiar ein aereoa an ara arsaa aeaa ra aaa eaaa ia e aaa a aak anaa aaaea iaaa EENG 33 K el VE Te EE 34 TA AT amp T CONTACT 34 7B STRATEGIC SUPPLIER CONTACTS unana nun nn nnen an EAEE EEEE anann Ennan nennen nna 34 2004 2005 AT amp T Knowledge Ventures All rights reserved 2 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 1 Copyright Page Notice This document is an unpublished work prote
19. Results Test Result Description Total Util Avg The average bandwidth received expressed as a percentage of the entire 1 Gbps of available bandwidth The average is calculated over the time period elapsed since the last test restart Total Util Cur The current bandwidth received expressed as a percentage of the entire 1 Gbps of available bandwidth This result indicates what rate of traffic is making it across the link from the far end Ina Constant Rate test this should be equal to the customer contract rate For a Ramp Test this should ramp until it is at the contract rate and stop when a Pause Frame is seen Total Util Peak The peak bandwidth utilized by the received traffic expressed as a percentage of the entire 1 Gbps of available bandwidth since the last test restart Frame Rate Avg The average rate of received frames expressed in frames per second The average is calculated over the time period elapsed since the last test restart Frame Rate Cur The current rate of received frames expressed in frames per second This measurement is an average taken over the prior second of test time Frame Rate Min The minimum rate of received frames over a one second period expressed in frames per second Frame Rate Peak The maximum rate of received frames over a one second period expressed in frames per second Frame Size Min The size in bytes of the smallest frame seen
20. ST ATT TP76440 002 Ethernet Testing using the Acterna Test Set Abstract Presented in this document are the methods and procedures to test Ethernet Services using the Acterna Test Set Audience The primary audience for this document are AT amp T LOCAL EXCHANGE company personnel in the following disciplines Local Field organization LFO IN Special Services LFO OUT Information Technology Operations Switch Capacity Planner Engineer Transport Equipment Engineer TEE Facility Equipment Engineer FEE Digital Transport Engineer DTE Maintenance Engineer Space Planner Frame Planner Long Range Technical Planners Outside Plant Engineering Fundamental Network Planning Special Services I M and Local Field Operations Central Office LFO IN This document is to be used internally within AT amp T LOCAL EXCHANGE companies and their Authorized AT amp T Approved Vendors and has a limited distribution subject to the header footer information Effective Date June 29 2005 Date Issued Issue 1 June 29 2005 Expires On Reissue of this document Related Documents See Reference Section of this document and ATT TP76412 ATT TP76440 amp ATT 002 316 076 Canceled Superceded Doc N A Issuing Dept AT amp T Services Inc Network Planning amp Engineering Transport amp Common Systems and Outside Plant OSP Construction Business Unit Network Documents Coordinator Wing Eng Area Manager Network Planning amp Engineering AT amp T
21. by step instructions for performing Gigabit Ethernet tests using the TestPad End to End or Single Ended Testing Since Auto Negotiation of Ethernet link parameters and Flow Control performance are local events at each end of the circuit end to end testing with 2 FST 2802 s provides the most thorough verification of all aspects of the circuit operation Figure 4 depicts this scenario Figure 4 Set up for 2 Ended Gigabit Circuit Test Auto Negotiation Auto g gt e Pause Frames Pause Frames Connectivity Ethernet link initialization The recommended setting for customer equipment and therefore for the FST 2802 as it emulates customer equipment is for auto negotiation and flow control to both be turned on To specify link initialization parameters 1 Using the application buttons select a TERM gt 1G Ethernet for fiber or 10 100 Ethernet for copper 2004 2005 AT amp T Knowledge Ventures All rights reserved 16 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 YRU TERM Py AutoNeg 10 1 00 Ethernet Laser Loop Loop Single ENS Reset Svc CJ Up Down Traffic FCS Error OFF Disruption erminate 1G Ethernet Demo SES im l 0 AN 2 Tap SETUP A group of quick configuration buttons appears 3 Tap Link Init The Link Init tab appears 4 Tap the AutoNeg On Button 5 Tap the AutoNeg button that becomes available A window appears lis
22. cted by the United States copyright laws and is proprietary to AT amp T Properties Disclosure copying reproduction merger translation modification enhancement or use by anyone other than authorized employees or licensees of AT amp T Properties without the prior written consent of AT amp T Properties is prohibited 2004 2005 AT amp T Knowledge Ventures All rights reserved All rights reserved Trademarks Windows 95 97 98 2000 NT ME XP Excel Word for Windows PowerPoint Internet Explorer Office Professional 97 Visio 2002 Visio Professional 2002 Visio Enterprise Network Tools and Microsoft are trademarks of the Microsoft Corporation Netscape Navigator is a trademark of the Netscape Corporation AutoCAD 2000 2000i amp 2002 are trademarks of the Autodesk Corporation Adobe Acrobat 7 0 and PDF are trademarks of the Adobe Corporation Common Language SWITCH TIRKS CLEI CLLI and CLFI are registered trademarks of Telcordia Technologies Inc ANSI is a trademark of the American National Standards Institute TIA is a service mark of the Telecommunications Industry Association EIA is a service mark of the Electronic Industries Alliance IEEE is a trademark of the Institute of Electrical and Electronic Engineers All other trademarks belong to their respective owners Publisher Network Planning amp Engineering Common Systems amp Transport Standards AT amp T Services Inc 1116 Houston Street RM 840 Ft Worth
23. customer tries to send Ethernet frames that would overflow the buffer capacity on the Ethernet card AT amp T does not support this feature due to the following issues 1 Some Ethernet Cards do not have the capability to perform this feature 2 Some manufacturers products have the Flow Control feature tied to Auto Negotiation Turning on Flow Control also turns on Auto Negotiation which is not desired 3 Some manufacturers products require a complicated procedure and the aid of an external device to turn on Flow Control 4 Standards have not fully matured on this feature 2004 2005 AT amp T Knowledge Ventures All rights reserved 30 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 5C Fault Propagation Fault Propagation allows traffic sources and destinations to be notified if a fault occurs anywhere in the network AT amp T will not support this feature due to the following reasons 1 Some manufacturers Ethernet cards do not have Fault Propagation supported or available 2 Ifa customer has alternate paths set up for traffic reroutes in case of network failures the customer equipment may not be notified about the fault to perform a traffic reroute in a timely fashion 5D Throughput Customers should not expect to get the maximum throughput capacity and speeds While AT amp T supports the 99 999 reliability on circuit level traffic bursty Ethernet
24. d Frames A count of frames transmitted since the last test restart RX Acterna Frames A count of received frames with Acterna test packets PAUSE Frames A count of PAUSE frames received from a remote Ethernet device This should be 0 until using the Ramp Test When the bandwidth approaches the provisioned limit this counter should increment by at least 1 Out of Seq Frames A count of out of sequence frames Only valid while receiving frames with the Acterna payload If this result is non zero it indicates that frames are being lost on the circuit under test VLAN Frames A count of VLAN frames as defined in IEEE 802 p q This result only appears for TestPads with the VLAN Tagging option Unicast Frames The number of unicast frames received since the last test restart 2004 2005 AT amp T Knowledge Ventures All rights reserved 26 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 Multicast Frames The number of multicast frames received since the last test restart Broadcast Frames The number of broadcast frames received since the last test restart 64 Byte Frames A count of frames with a length of 64 bytes 65 127 Byte Frames A count of frames with lengths between 65 and 127 bytes inclusive 128 255 Byte Frames A count of frames with lengths between 128 and 255 bytes inclusive 256 511 Byte Frames A count of frames wi
25. d rate to make sure that no traffic over the contract will pass This test protects the provider from wasted core bandwidth due to incorrect provisioning GBIC Descriptions Table 2 lists the GBIC types with the associated cables and connectors Table 2 GBICs and cables Wavelength Mode GBIC Jacks Fiber Cable Connector 850 nm MM Black or beige Orange SC 1310 nm SM Blue Yellow SC It is a requirement to use matched GBICs and the appropriate fiber optic cable when connecting the network element to the customer device or to your FST 2802 TestPad Many Ethernet devices include support for Auto Negotiation This allows the element ports to automatically negotiate Ethernet parameters with the near end device such as the customer s switch or router or in this case the FST 2802 TestPad Settings required include half duplex or full duplex operation Full duplex means that each port can both transmit and receive simultaneously Half duplex mode means that if required by the attached customer device Some older Gigabit switches may require this mode A second selectable parameter that must be addressed is Flow Control Flow control refers to the ability to send and or respond to pause frames When enabled an Ethernet device will generate pause frames toward an attached device when there is too much ingress traffic This capability allows Ethernet devices to smooth out the bursty nature of Ethernet traffic This is what allows a
26. e Error Stats category lists error statistics such as the number of symbol errors FCS errored frames and runts In this test procedure any non zero value for these Error Stats constitutes a test failure Test Result Description Symbol Errors A count of invalid 10 bit code words received on the physical layer This result will not increment more than one time per frame FCS Errored Frames A count of frames containing Frame Check Sequence CRC errors Runts A count of frames under the required 64 byte frame length containing Frame Check Sequence CRC errors Undersized Frames A count of frames under the minimum 64 byte frame length Oversized Frames A count of frames over the maximum 1518 byte frame length Errored Frames A summed count of FCS Errored Frames Runts Undersized Frames and Oversized Frames 3D End to End Testing Illustration Eonnection made from EE Front of Medi Comerter T ait ah Ges Skt Ito Front of RI di i K gsgailt Connection mada from from Panel Port A Wen AL of Melk Coameter ikt J to Pront af KI dE Fanal Port 2d Far Connection mada from Front of Mei Converter Fbar H kt ite Fiar Connection mada from Front af Madi Converter Ebar Sht Jti Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 3E Note on Sending Deliberate Errors In the transport test world it is common to send a
27. eceived traffic expressed in bits per second This measurement is an average taken over the prior second of test time In a loopback test the Rx and TX Bits Sec Cur values should track together This is also true for an end to end test if the same profile settings are used Tx Bits Sec Cur The current bandwidth utilized by the transmitted traffic expressed in bits per second This measurement is an average taken over the prior second of test time This will tell us how fast we are transmitting In the Ramp Test once a pause frame is received we want to check this result to see that it settles in near the rate in the customer contract Delay Max The maximum Round Trip Delay measurement since the last test restart You must transmit an Acterna payload to measure round trip delay RTD measurements are only performed when using a Constant traffic type Check the service contract or SLA for a max RTD parameter If there is none a benchmark for the particular network is still recommended In general the Max RTD should not be vastly greater than the average A significant differential 2004 2005 AT amp T Knowledge Ventures All rights reserved 7 Ethernet Testing using the Acterna Test Set June 29 2005 AT amp T Services Inc ATT TP76440 002 Issue 1 Delay Min would indicate packet jitter is being imposed on the circuit by a switching or buffering device The minimum round trip delay calculated in
28. elect Constant and enter the bandwidth the customer is buying in either Mbps or by of a 1G or 10 100 M load as appropriate See the customer contract or work document 10 Tap OK 11 For a 2 ended test perform steps 2 through 11 at the far end 12 From the main user interface screen make sure the laser is on lower left gt flashing yellow and Signal Sync and Link Active are all still lit If there are any errors red LEDs tap Restart in the upper right Now tap Start Traffic You are now transmitting across the link 13 Desired Results The Summary Category should indicate All Summary Results OK The Link Counts category should show Out of Seq Frames equal to 0 14 Link Utilization should show full bandwidth as being transmitted by the far end See the combined results section at the end of the application note for detailed explanations Testing Pause Functionality For a full Gigabit service 1000Mbps the Constant rate traffic test we just performed above would be sufficient However we mentioned that we would use a Ramp traffic type to test the Pause functionality of the network elements What ramp allows us to do is to gradually increase the traffic rate until we approach and exceed the provisioned backbone bandwidth The result seen should be a pause frame issued by the network element port back towards the FST 2802 Also we should not be able to transmit over the contract rate without incurring Out of Sequence frames Since the
29. essed as a percentage of the entire 1 Gbps of available bandwidth This result indicates what rate of traffic is making it across the link from the far end Total Util Peak The peak bandwidth utilized by the received traffic expressed as a percentage of the entire 1 Gbps of available bandwidth since the last test restart Frame Rate Average The average rate of received frames expressed in frames per second The average is calculated over the time period elapsed since the last test restart Frame Rate Cur The current rate of received frames expressed in frames per second This measurement is an average taken over the prior second of test time Frame Rate Min The minimum rate of received frames over a one second period expressed in frames per second Frame Rate Peak The maximum rate of received frames over a one second period expressed in frames per second Frame Size Min The size in bytes of the smallest frame seen since last test restart Frame Size Max The size in bytes of the largest frame seen since last test restart Received Frames The number of frames received since the last test restart Transmitted Frames The number of frames transmitted since the last test restart Ina loop back test Received and Transmitted frames should be equal If they are not an error may have occurred Check error counters Rx Bits Sec Cur The current bandwidth utilized by the r
30. explained as we go along Before we do that we need to set up some Ethernet frames to actually send To define a frame profile 1 Tap Setup A menu appears listing the Quick Configuration buttons 2 Tap Tx Profiles The Tx Profiles tab appears Terminate 1G Ethernet Ki ID Summary Error Tx Profiles Rx Profiles Link Init Traffic Selections Settings Te Profile Profiles DA Type Dest Address 7P1 cP3 Profile 1 Unicast 008016280000 cP2 LP Frame EtherType DIX 0800 z Payload C Acterna VLAN Tag Non tagged sl User Defined Length en 3 Under Selections Tx Profile select the Profile P1 4 In the Settings fields specify the frame characteristics for the traffic transmitted using the profile 5 Under Profile select Profile 1 to set up the profile we chose to transmit 6 Under Length select 256 You can ignore DA Type Dest Address Frame and EtherType and VLAN Tag settings for this application as they are not relevant 7 Under Payload select Acterna 8 Now Tap on the Traffic tab 2004 2005 AT amp T Knowledge Ventures All rights reserved 18 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 Terminate 1G Ethernet Summary Error Tx Profiles Link Init Traffic Load Type Load Unit Const Load c Mbps Constant amp BW c Bursty Ramp cance 9 S
31. he Traffic Tab Oo CONDO Terminate 1G Ethernet Si Summary Error Tx Profiles Link Init Traffic 1D Load Type Time Step sec Stop Load Increment Constant Il 1 EI c Bursty V Errored Frames v Pause Frames Load Step XBW J i Dropped Frames z E en Select Ramp Select 5 0 seconds Time Step and 5 for Load Step All the 15454 sub rates are divisible by 5 Select all three of the Stop Load Increment criteria and set them at 1 Tap OK and return to the main user interface You can now turn on the Laser Tap restart to clear any errors and tap Start Ramp You are now transmitting traffic at increasing increments to determine at what bandwidth the circuit experiences Errored Frames Dropped Frames or Pause Frame Note at what rate any one of these events occurs The desired result is that there are no error or 2004 2005 AT amp T Knowledge Ventures All rights reserved 20 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 dropped frames and that pause frames are seen at the approximate BW that the customer is being sold 9 NOTE If errors begin to occur but no Pause Frame has been received verify that Flow Control Negotiation enabled on the network element 10 Desired Results A pause frame should be counted in the Link Counts category When it is the Ramp function will stop incrementing The Li
32. ly sends information when needed This is a connectionless protocol 10 Mb Ethernet The IEEE 802 3 standard configuration 10BaseT is the most popular LAN standard applied today The 10 is for 10Mbps the Base is for baseband signaling and the T is for UTP wiring Operates over twisted pair 10Base T and fiber 10Base FL Maximum distances are 100 meters 10BT and 2 kilometers 10BFL 100Mb Ethernet Data operates at 100 million bits per second 100 Mbps Manchester encoding for 200Mbps Baseband signaling method Operates over twisted pair 100Base TX and fiber 100Base FX Maximum distances are 100 meters 100Base TX and 2 kilometers 100Base FX SM Gigabit Ethernet IEEE 802 3z Merging of ANSI X3T11 Fiber Channel and 802 3 standards 1Gb Data Rate 8B 10B Encoding 1 25Gbps Baseband signaling method Operates over Long Wave LW laser over single mode and multimode fiber 1000Base LX and Short Wave SW laser over multimode fiber 1000Base SX Ethernet Framing Ethernet makes use of frames or packets to send information Ethernet frames contain addressing framing and error checking data as well as payload containing the user information There are two types of framing structures used within the industry the Ethernet Type II frame DIX and the IEEE 802 3 frame DIX Type II Frame Destination Source Payload Preamble Address Address Type User Data CRC 8 Bytes 6 Bytes 6 Bytes 2 Bytes 46 1500 Bytes
33. nk Stats category should show a Current Utilization equal to that which the customer is being provisioned e g fora STS12c customer 600Mbps a pause frame should have been issued when the test set made the jump from 600Mbps to 650Mbps Upon seeing the Pause frame the 2802 will down speed to 600Mbps or 60 utilization See the combined results section at the end of the application note for detailed explanations To set up for a Burst Test 1 Tap Setup 2 Tap the Traffic Tab Terminate 1G Ethernet Summary Error Tx Profiles Link Init Traffic Load Type Burst Config Load BW Frames Burst Burst Gap 100 Jl eg r Constant e ABW 4 L Bursty Frame Size No of Bursts Fixed c Fixed RAND Fixe Random Continuous Cancel Tap BW Tap Load 100 Tap Frame Sized Fixed Tap Frames Burst 16 Tap No of Bursts Continuous Tap OK to return to the main screen Complete steps 1 7 on the far end test device Start Traffic on both 2802 s 0 In Summary results ensure all results OK 2004 2005 AT amp T Knowledge Ventures All rights reserved 21 A OO JO Om P Gabi Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 Test Results Section Figure 5 above provides us with our desired results The left hand results screen is in the Summary category and indicates that there are no error conditions to report Figure 6
34. on has been achieved This assures the technician s that the proper end locations are the one s that are synchronized The parallel for Ethernet testing is to send an FCS error However one of the inherent traits of Ethernet is that errored frames are discarded by Ethernet interfaces In this application all frames with deliberate FCS errors sent by the FST 2802 will be dropped by the locally connected G1000 4 port Instead of seeing an FCS what the far end FST 2802 will see is that a frame was lost Similarly if the far end had a hard loopback and an FCS error was injected an Out of Sequence Lost Frame would be the result In this way the technique so often used in the TDM world can still be used for most Ethernet Transport applications 2004 2005 AT amp T Knowledge Ventures All rights reserved 27 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 Appendix B Brief Ethernet Basics This Appendix provides a very brief outline of the technology through some of it s terms Please contact Acterna for a more detailed explanation of Ethernet technology Ethernet Standards IEEE 802 3 Ethernet is a data transmission protocol as opposed to SONET which is optimized for telephony applications SONET continuously transmits data between devices in order to preserve synchronization Dummy bits fill gaps between user data SONET is a connection oriented protocol Ethernet on
35. ons select a TERM 10 100 Unit A Select SETUP A group of quick configuration buttons appears Select Link Init The Link Init tab appears Set the parameters to Auto Neg On Flow Control On Click OK to get back to the main user interface Repeat the same process at the far end test Unit B Connect each test set each end of circuit See Figure 1 and 2 Turn Signal ON for Unit A Now look for Signal from previous step Sync LEDs on the test set of Unit B If link is not established check the settings on both devices and try again 10 Turn Signal ON for Unit B 11 Now look for Signal from previous step Sync LEDs on the test set of Unit A If link is not established check the settings on both devices and try again 12 You now have an Ethernet connection between the two ends of the circuit This will be indicated by a Sync LED on both test sets at either end of the circuit OO OO Joo Go 3B Frame Generation 1 Monitor the LED s for Signal Sync Link Active and Frame Detect 2 All are lit indicating the Test Pad detects a signal obtains synchronization link is active and frames are detected 3 On the left results pane select the Result Category button 4 Select Link Stats and record stats Average Bandwidth Received 100 Current Bandwidth 100 Shows a continual growing number of frames received and frames transmitted 5 On the right results pane select the Result Category Button 6 Select the Error Stats and rec
36. ord stats Symbol Errors 0 FCS Errored Frames 0 Runts 0 Undersized Frames 0 Oversized Frames 0 Errored Frames 0 3C Signal Throughput The test set supports three traffic load types They are Constant Rate Burst or Ramp For this application we will use a Constant Load type at 10 100 The First step is to prepare the test Ethernet frames for transmission 2004 2005 AT amp T Knowledge Ventures All rights reserved Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 3C 1 Constant Load Test To choose a frame profile Select Setup A menu appears listing the Quick Configuration buttons Select Tx Profiles The Tx Profiles tab appears Under Selections Tx Profile select the Profile P1 There is no need to change anything in the Settings fields Repeat for the far end test set Unit B OO GON To set up for a Constant Load Test Select Setup if not already in set up Select the Traffic Tab Select Const Select BW and 100 Select OK and return to the main user interface Repeat steps 1 through 5 for the far end test set Unit B Select restart to clear any errors and press button to Start Traffic You are now transmitting traffic at 100 percent of the circuit capability Note if errors occur See the Test Results section for details on results categories and values Perform the 100 transmission test for 8 minutes or more 9 Repeat process for far end
37. re round trip delay RTD measurements are only performed when using a Constant traffic type Check the service contract or SLA for a max RTD parameter If there is none a benchmark for the particular network is still recommended In general the Max RTD should not be vastly greater than the average A significant differential would indicate packet jitter is being imposed on the circuit by a switching or buffering device Delay Min The minimum round trip delay calculated in microseconds You must transmit an Acterna payload to measure round trip delay RTD measurements are only performed when using a Constant traffic type Check the service contract or SLA for an avg RTD parameter If there is none a benchmark for the particular network is still recommended In general this value should be under 100 ms Average delays in excess of 100 ms may impair the performance of delay sensitive applications Delay Avg The average round trip delay calculated in microseconds You must transmit an Acterna payload to measure round trip delay RTD measurements are only performed when using a Constant traffic type Svc Disruption The service disruption time maximum inter frame gap when service switches to a protect line calculated in milli seconds Test Result Table 4 Link Counts results Description Received Frames A count of frames received since the last test restart including errored frames Transmitte
38. rform testing from test set to test set to validate Category 5t facilities 3F Test Set Interoperability 3F 1 Same Type Test Sets With the deployment of Ethernet type services it has been well known that there are idiosyncrasies and standards differences between products and manufacturers As a result it is most advantageous that when end to end testing is performed that the same type of test set and model from the same manufacturer is used for this testing The two test set manufacturers standardized within AT amp T Sunrise and Acterna work extremely effectively when the same test set is used on each end of the circuit The general availability of the Sunrise Telecom MTT 28 SSMTT 29 and Acterna 2802 Test Sets will be December 2003 3F 2 Different Type Test Sets Ethernet testing using different manufacturers products dictates special handing due to the soft standards found within the Ethernet architecture While AT amp T has selected two outstanding test sets from two different manufacturers Sunrise Telecom amp Acterna both which will test to each other in an end to end test The following information was ascertained through multiple interoperable tests When the following stipulations are met both test sets will function with one another If any other manufacturer test sets are used it will be necessary for interoperable lab tests be performed to insure their interoperability compliance as well 3F 2 A End to End testing
39. since last test restart Frame Size Max The size in bytes of the largest frame seen since last test restart Received Frames The number of frames received since the last test restart Transmitted Frames The number of frames transmitted since the last test restart Ina loop back test Received and Transmitted frames should be equal If they are not an error may have occurred Check error counters Rx Bits Sec Cur The current bandwidth utilized by the received traffic expressed in bits per second This measurement is an average taken over the prior second of test time In a loopback test the Rx and TX Bits Sec Cur values should track together This is also true for an end to end test if the same profile settings are used Tx Bits Sec Cur The current bandwidth utilized by the transmitted traffic expressed in bits per second This measurement is an average taken over the prior second of test time This will tell us how fast we are transmitting In the Ramp Test once a pause frame is received we want to check this result to see that it settles in near the rate in the 2004 2005 AT amp T Knowledge Ventures All rights reserved 25 Ethernet Testing using the Acterna Test Set June 29 2005 AT amp T Services Inc ATT TP76440 002 Issue 1 customer coniract Delay Max The maximum Round Trip Delay measurement since the last test restart You must transmit an Acterna payload to measu
40. th lengths between 256 and 511 bytes inclusive Since we chose a frame size of 256 all the frames should be counted here If you chose a different frame size the counts will fall into the appropriate bins 512 1023 Byte Frames A count of frames with lengths between 512 and 1023 bytes inclusive 1024 1518 Byte Frames Acount of frames with lengths between 1024 and 1518 bytes inclusive Error Stats Results The Error Stats category lists error statistics such as the number of symbol errors FCS errored frames and runts In this test procedure any non zero value for these Error Stats constitutes a test failure Table 5 Error Stats results Test Result Description Symbol Errors A count of invalid 10 bit code words received on the physical layer This result will not increment more than one time per frame FCS Errored Frames A count of frames containing Frame Check Sequence CRC errors Runts A count of frames under the required 64 byte frame length containing Frame Check Sequence CRC errors Undersized Frames A count of frames under the minimum 64 byte frame length Oversized Frames A count of frames over the maximum 1518 byte frame length Errored Frames A summed count of FCS Errored Frames Runts Undersized Frames and Oversized Frames Note on Sending Deliberate Errors In the transport test world it is common to send a bit error across a circuit once BERT pattern synchronizati
41. ting the auto negotiation capabilities For Pause Capable select the following Rx Only This will allow the FST 2802 to properly pause traffic when it receives Pause Frames For Full Duplex FDX and Half Duplex HDX select Yes on both will support both 6 Click OK on both screens to get back to the main user interface 7 Connect the FST 2802 to the network element using the proper cable Fiber or Copper as appropriate For Fiber tap the gray flashing Laser OFF button on the lower left of the screen to activate the FST 2802 laser The button will illuminate yellow and say Laser ON The 2 Gigabit ports should now perform auto negotiation and establish the Ethernet link 8 Since this is a 2 ended test perform the same process at the far end FST 2802 9 Look for Signal Sync and Link Active LEDs on the right side of the FST 2802 If you do not get a Link make sure that the technician at the far end has the 2802 set up through step 7 of this procedure 10 You now have an Ethernet connection between the locally connected G1000 4 port and the test device 2004 2005 AT amp T Knowledge Ventures All rights reserved 17 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 Throughput Testing The FST 2802 supports three traffic load types They are Constant Rate Burst or Ramp For this application we will use both Constant Rate and Ramp The reasons well use these two will be
42. traffic are not guaranteed speeds or performance reliability factors Just a few of the many reasons are listed 1 Some Ethernet cards do not deliver the ideal throughput due to Ethernet processing issues or clock synchronization components 2 Customers could sometimes see effective throughput even more than the expected maximum speeds This could happen because while Ethernet is mapped over SONET the inter frame gap in Ethernet is removed and re inserted only when Ethernet is de encapsulated from SONET 5E Performance Metrics Almost all Ethernet cards have issues with performance metrics Some of the PM counts are incorrectly calculated on these cards Some important PM counts are missing on these cards Standards for Performance Metrics are considered as optional in many manufacturers products As aresult AT amp T will not offer or support Performance Metrics for Ethernet 1 Customers may see packet losses or degraded performance due to certain traffic patterns 2 AT amp T will not be able to easily verify customer complaints on this issue 3 Customers should not expect AT amp T to respond to throughput or packet loss problems 2004 2005 AT amp T Knowledge Ventures All rights reserved 31 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 5F Jumbo Framing Jumbo Frames are non standard Ethernet Frames that are longer than 1518 bytes for untagged frames and 1522
43. ts ADC Telecommunications Inc Carter Gaddis Field Sales Engineer ADC Telecommunications Inc 972 878 2066 E Mail Carter Gaddis adc com Dennis Powell Field Sales Engineer ADC Telecommunications Inc 847 452 3468 E Mail dennis powell adc com Sunrise Telecom Jeff Harmon Regional Sales Engineer 972 529 6073 E Mail jharmon sunrisetelecom com Acterna Todd Watkins AT amp T Account Team 972 692 3781 todd watkins acterna com Westek 800 526 2673 2004 2005 AT amp T Knowledge Ventures All rights reserved 34
44. ve a true test of all segments of the circuit The same procedure can be of use testing the different segments when needing to isolate a defect for repair purposes The three tests performed are essential for the successful deployment of Ethernet services These tests are to determine e Connectivity e Throughput e QoS Ethernet Elements The Ethernet elements that the 10 100BaseT signals passed through to complete the end to testing are the following e ADC FMT Fiber Management Tray e ADC RJ 45 to RJ 45 EIA TIA T568 B specification e ADC Media Converter Module 10 100BaseT 100Base LX RuJ 45 1310nm FP Laser SC e 10ft Cat 5e 5t patch cords e 10ft Fiber Optic SC UPC patch cords 3A Ethernet Link Connectivity Test Procedure 1 Connectivity Signal Light 2 Synchronization Bit Stream from far end 3 Link Establishment with far end test set Both ends communicating 2004 2005 AT amp T Knowledge Ventures All rights reserved 4 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 Before you transmit and receive traffic Ethernet frames over a circuit you must initialize an Ethernet link At a minimum initializing a link involves connecting compatible test sets that emulates Ethernet switch ports to a circuit and allows the test sets to transmit idle traffic Connectivity Link Establishment To specify link initialization parameters Using the application butt
45. ving full STS24c channels for lesser loads e Hitless remote upgrades as customer purchases greater bandwidth to meet growing needs no truck roll no line card swaps as with OC3 to OC12 jump Table 1 Relationship between SONET mapping and maximum throughput Transport Circuit Size GE Interface Speed Bandwidth in Mbps STS1 5 50 STS3c 15 150 STS6c 30 300 STS9c 45 450 STS12c 60 600 STS24c 100 1000 STS48c 100 1000 2004 2005 AT amp T Knowledge Ventures All rights reserved 13 Ethernet Testing using the Acterna Test Set AT amp T Services Inc June 29 2005 ATT TP76440 002 Issue 1 NOTE These values are based on average EOS throughput test results with various frame sizes Results can vary based on test equipment frame size and line conditions In general the SONET circuit can transport more than the Bandwidth listed in the table but published values are rounded rather than listed in traditional Sonet TDM sizes For example actual throughput on an OC12c might approach 620 Mbps with certain frame sizes but the circuit is operating correctly as long as it operates error free to 600 Mbps The bandwidth sold will be defined in the customer s contract and should be part of the work document for the circuit turn up The testing should verify error free operation of the circuit up to that bandwidth In order to check overall provisioning of the circuit traffic should be analyzed above the contracte
Download Pdf Manuals
Related Search