Generic framingprocedure

ITU-T G.7041

White paper

Hussain QureshiStuart Ferguson

Colin Scotland

Electronic Products Solutions GroupTelecomms Networks Test Division

Scotland

2

Contents

Abstract ........................................................ 3

Drivers for Ethernetin the metropolitan network ..................... 4

Introduction to GFP ..................................... 5Frame mapped GFP .......................................... 5Transparent mapped GFP ................................. 5

Structure of GFP frame .............................. 7

GFP encapsulation techniques ................. 8Ethernet MAC encapsulation ........................... 8PPP/HDLC encapsulation ................................. 9FICON/ESCON encapsulation .......................... 9

Advantages of GFP .................................. 10

Conclusion ................................................ 11

Glossary of terms ..................................... 11

3

Abstract

Unlike synchronous SONET/SDH,Ethernet has a bursty trafficpattern. To compensate for itsidle times and bring it up to therequired data rate, encapsulation(or a wrapper) is needed toenable Ethernet traffic to betransported over the SONET/SDH network. One of the mostflexible encapsulationmechanisms is Generic FramingProcedure (GFP). This paperexamines the GFP framingstructure and its flexibilitytowards multi-serviceenvironments, demonstratingthat GFP has the potential ofbecoming the dominantencapsulation technique in next-generation SONET/SDHnetworks.

4

Drivers for Ethernet in themetropolitan network

less than traditional SONET/SDHinterfaces, and offers point-to-point and multipoint serviceoptions.

In addition, today’s networksneed to be flexible to support amulti-user environment wherethe usage pattern and usage timeof the end user can vary greatly.For example, bandwidthallocated for a remote officeconnection and for employeeInternet access can be reducedafter working hours, butbandwidth for connections to aWebsite host server must beavailable constantly. Ethernet, inassociation with evolvingtechnologies, will be able toprovide usage pattern flexibilityas well as ease of expansion.

Nevertheless, while Ethernet isused extensively in theenterprise network, the long-hauland metro networks use

Figure 1.From legacy to next-generationnetworks

predominantly SONET/SDH-based circuit technology. ForService Providers to takeadvantage of Ethernet flexibilityand cost efficiency, Ethernet overthe legacy network requires anadditional mechanism if theSONET/SDH networks are totransport packet-based trafficeffectively. And this is whereGFP plays a significant role. GFPencapsulates frame/packet-basedprotocols within SONET/SDH,acting as a rate-adapting bridgelayer between Ethernet andSONET/SDH. It also allowsindividual Ethernet streams tobe switched and groomed, andprovides header error correctionand channel identifiers for portmultiplexing. This gives ServiceProviders the flexibility andgranularity required to transportpacket-based traffic within thebasic SONET/SDH framingstructure.

Internet access and videoconferencing have transformedthe way companies across theglobe conduct business, bothwithin their organizations andwith their customers. Thetransformation is partly due tothe emergence of new networktechnologies, such as Ethernet,to provide the necessary supportservices and ensure sufficientquality and cost-effectiveness.Ethernet is the dominanttechnology in local area networks(LANs), and it is expected to findits way into metropolitan areanetworks (MANs) and wide areanetworks (WANs), along withprotocols such as Fiber Channel,enterprise system connection(ESCON) and fiber connection(FICON). The reason is that as auniversal service interface,Ethernet is capable of providingthe characteristics of a range ofstandard voice and data services,it can deliver broadband, it costs

5

Introduction to GFP

Frame-mapped GFP

This mode maps the entire clientframe into one GFP frame. It alsodescribes a single client framethat is mapped into a single GFPframe, for example, an Ethernetframe mapped into a GFP frame.

Transparent-mapped GFP

This mode facilitates thetransport of block-coded clientsignals (for example FiberChannel, ESCON or FICON) thatrequire very low transmissionlatency. The individual

GFP is defined as a genericmechanism to transport anyclient signal over fixed data-rateoptical channels. Any clientsignal like IP/PPP, Ethernet MAC,Fiber Channel, ESCON or FICONcan be mapped over thetransport network using GFP. Assuch, GFP provides a single,flexible mechanism to map anyclient signal into SONET/SDHand the optical transportnetwork (OTN).

GFP supports both point-to-pointand ring applications andeliminates the need for byte/bitstuffing. This avoids payloadspecific frame expansion, whichsaves bandwidth. GFP utilizes alength/HEC-based framedelineation mechanism that ismore robust than that used byHDLC (High-level Data LinkControl) which is single octet flagbased.

To cater for all mappingrequirements, two mappingmodes are currently defined forGFP:

� Frame-mapped GFP (GFP-F)� Transparent-mapped GFP

(GFP-T)

characters of a client signal arede-mapped from the client signaland then mapped into GFPframes. This process helps avoidbuffering an entire client framebefore it is mapped into the GFPframe.

In the transparent mappingmode, the GFP frame containsgroups of 8B/10B code-groupsmapped into a 64B/65B codewith a cyclic redundancy check(CRC). The GFP frame structureremains the same whether it isframe-mapped GFP ortransparent-mapped GFP.

Figure 2.The fields used todescribe the GFP frame

6

In general, there are two types ofGFP frames, GFP client framesand GFP control frames:

GFP client framesThese can be further classifiedinto two types:

� Client data frames, used totransport data from the clientsignal

� Client management frames,used to transport managementinformation about clientsignals between source andsink, for example Loss ofClient signal. These framescomprise the core header andpayload area only and havethe value of PTI as 100

GFP control frames (GFP idle frames)These frames are used for themanagement of the GFPconnection, and consist of onlythe Core Header field (nopayload area). These frames areused to compensate for the gapsbetween the client signal where

the transport medium has ahigher capacity than the clientsignal.

GFP frames from multiple portsor multiple client types aremultiplexed on a frame-by-framebasis. GFP idle frames areinserted if there are no GFPframes available fortransmission. This provides acontinuous stream of frames formapping.

In the case of a Client Signal Fail(CSF) condition, the GFP sourcegenerates a client managementframe every 100 ms. On receiptof a CSF indication, the GFP sinkdeclares a client signal failure.This condition is cleared eitherby receipt of a valid GFP frame,or when no CSF indications arereceived for 1000 ms. GFP idleframes are sent during the CSFcondition.

7

Figure 3.Basic GFP client frameconsists of a core headerand a payload area

Core Header This describes the GFP frame, independent of the content ofhigher layer Protocol Data Units (PDUs). The cHEC contains aCRC error control code for protection of Core Header contents.The scrambling of the Core Header provides additionalrobustness of the GFP frame delineation process.

Payload Header Payload Header supports data link management proceduresspecific to client signal.Type field identifies the format and presence of the ExtensionHeader and Payload FCS. This field also distinguishes betweenGFP frame types and between different services in a multi-service environment.Payload Type identifies the type of GFP client frames.Payload FCS Indicator identifies the presence or absence of thePayload FCS field.Extension Header Identifier is used to identify the GFP ExtensionHeader. (null, linear or ring).User Payload Identifier is used to convey the type of payloadcarried in the GFP Payload Information field.tHEC field protects the integrity of the Type fieldChannel Identification (CID) field is used to identify thecommunication channel at a GFP termination point.GFP Extension Header support the technology specific headers,for example Virtual Link Identifiers, source/destinationaddresses, port numbers and class of service.� Null Extension Header is intended for dedicated transport

paths of one client signal� Linear Extension Header is intended for several independent

links requiring aggregation onto a single transport path� Ring Extension Header is intended for use where a ring

transport path is shared among multiple client signalsExtension HEC field is used to protect the integrity of thecontents of Extension Header.

Payload In the case of Frame-mapped GFP, the GFP payload area containsInformation the framed PDU; in the case of Transparent GFP it contains aField group of client characters.

Payload FCS pFCS is used toprotect the contents of the GFP payload

Structure of a GFP frame

8

GFP encapsulation techniques

The relationship between anEthernet MAC frame and GFP isshown in Figure 4. There is aone-to-one mapping between ahigher layer PDU and a GFPPDU, with the boundariesaligned to each other.

Ethernet MAC encapsulation

In the case of Ethernet MAC’sencapsulation into GFP, thefields from Destination Addressto Frame Check Sequence (FCS)are placed in the GFP payloadarea. As the source adaptationprocess extracts the frame from

Figure 4.Ethernet MACencapsulation into GFPusing frame-mapped GFP

the client bit stream, the GFPsource adaptation processdeletes the gaps between thepackets, known as the InterPacket Gaps (IPGs). EthernetMAC is then forwarded forfurther encapsulation into theGFP frame. On the sink side, thesink adaptation process restoresthe IPGs, and Ethernet MAC isthen forwarded to the clientlayer for further processing.

Error handlingIn the source adaptation process,any PDUs received with errorsfrom the client side arediscarded before transmission.However, PDUs detected forerrors during transmission arepadded with appropriate bitsequence and FCS. This actionensures that the client end dropsthe errored PDU.

9

PPP/HDLC encapsulation

PPP/HDLC frame encapsulationinto GFP is similar to theEthernet MAC encapsulationprocedure and is shown inFigure 5.

PPP frames are firstencapsulated in an HDLC-likeframe. Similar to Ethernet MAC,the boundaries of the GFP PDUare aligned with boundaries ofthe framed PPP/HDLC PDUs. Thefields from Address to FCS(including the PPP informationfield padding) are placed in theGFP payload area. The GFPsource adaptation processremoves the flags and associatedescape characters and forwardsthe PPP/HDLC frame for furtherencapsulation in GFP. On thesink side, the PPP/HDLC frameis extracted from the GFP frameand forwarded to the client layerfor further processing, such asthe addition of flags and escapecharacters.

Figure 5.PPP/HDLC encapsulationinto GFP using transparent-mapped GFP

ESCON/FICON encapsulation

In the case of ESCON/FICONencapsulation, the physical layerof the client signal is decodedand the decoded characters aremapped into a 64B/65B blockcode. The block code is thenmapped into payload bytes of64B/65B code in the order inwhich they were received.

10

Advantages of GFP

Of the many advantages of GFPas an encapsulation mechanism,the most significant is that GFPminimizes the protocol-specificprocessing and protocoltranslation associated withPacket over SONET/SDH, andavoids the complex adaptationlayer processing of ATM.Traditional encapsulationtechniques such as Packet overSONET/SDH cannot guaranteebandwidth since the paddingused inflates the frames size.GFP frame headers, on the otherhand, are exactly the same sizeas the Preamble (which isdropped during encapsulation)which guarantees bandwidth.

GFP also provides one uniformmechanism to adapt any payloadtype to any transport media. Inother words, GFP results innetwork flexibility, efficiency androbustness.

11

Conclusion

The key issue for ServiceProviders looking to adoptEthernet over SONET/SDH as atransport mechanism is whetherit is capable of dealing with data-oriented applications andsynchronous traffic such asvoice. Using GFP resolves thisissue by providing an efficient,scalable and unified mode oftransport for both synchronousTDM traffic and dataapplications such as those usedin LANs, storage area networks(SANs), and the Internet. GFP isalso the most economical way ofadopting high-speed services inlegacy metro and long-haulSONET/SDH networks, and canprovide the basis for the evolvingresilient packet ring (RPR)technology. In short, GFP has thepotential to become thedominant encapsulationtechnique in the modernnetwork.

Glossary of terms

ASP Application service provider

cHEC Core header error correction

CID Channel identification

CMF Client management frame

CSF Client signal fail

eHEC Extension header error correction

ESCON Enterprise system connect

EXI Extension header indicator

FICON Fibre connection

GFP-F Generic framing procedure framed

GFP-T Generic framing procedure transparent

HDLC High-level Data Link Control

LAN Local area network

MAC Media access control

PDU Protocol data unit

pFCS Payload Frame Check Sequence

PFI Payload FCS indicator

PLI Payload length indicator

PTI Payload type identifier

RPR Resilient packet ring

SAN Storage area network

tHEC Type header error correction

UPI User payload identifier

VLAN Virtual LAN

By internet, phone, or fax, get assistance with all yourtest & measurement needs

Online assistance:

www.agilent.com/find/assist

Phone or FaxUnited States:(tel) 1 800 452 4844

Canada:(tel) 1 877 894 4414(fax) (905) 282 6495

China:(tel) 800 810 0189(fax) 1 0800 650 0121

Europe:(tel) (31 20) 547 2323(fax) (31 20) 547 2390

Japan:(tel) (81) 426 56 7832(fax) (81) 426 56 7840

Korea:(tel) (82 2) 2004 5004(fax) (82 2) 2004 5115

Latin America:(tel) (305) 269 7500(fax) (305) 269 7599

Taiwan:(tel) 080 004 7866(fax) (886 2) 2545 6723

Other Asia Pacific Countries:(tel) (65) 375 8100(fax) (65) 836 0252Email: tm_asia@agilent.com

Product specifications and descriptions in this document subjectto change without notice.

© Agilent Technologies, Inc. 2002Printed in USA, July 23, 20025988-7272EN

Agilent Technologies’ Test and MeasurementSupport, Services, and AssistanceAgilent Technologies aims to maximize the value youreceive, while minimizing your risk and problems. Westrive to ensure that you get the test and measurementcapabilities you paid for and obtain the support youneed. Our extensive support resources and services canhelp you choose the right Agilent products for yourapplications and apply them successfully. Everyinstrument and system we sell has a global warranty.Support is available for at least five years beyond theproduction life of the product. Two concepts underlieAgilent’s overall support policy: “Our Promise” and“Your Advantage.”

Our PromiseOur Promise means your Agilent test and measurementequipment will meet its advertised performance andfunctionality. When you are choosing new equipment,we will help you with product information, includingrealistic performance specifications and practicalrecommendations from experienced test engineers.When you use Agilent equipment, we can verify that itworks properly, help with product operation, andprovide basic measurement assistance for the use ofspecified capabilities, at no extra cost upon request.Many self-help tools are available.

Your AdvantageYour Advantage means that Agilent offers a wide rangeof additional expert test and measurement services,which you can purchase according to your uniquetechnical and business needs. Solve problemsefficiently and gain a competitive edge by contractingwith us for calibration, extra-cost upgrades, out-of-warranty repairs, and on-site education and training, aswell as design, system integration, projectmanagement, and other professional engineeringservices. Experienced Agilent engineers and techniciansworldwide can help you maximize your productivity,optimize the return on investment of your Agilentinstruments and systems, and obtain dependablemeasurement accuracy for the life of those products.

Agilent Email Updates

www.agilent.com/find/emailupdates

Get the latest information on the products and applications youselect.