SONET/SDH

Smita Rai, Anpeng Huang, Suman

Sarkar

04/30/2006 Page 1

Outline
• Introduction
• SONET Technology
– Architecture
– Frames
• SONET Networks
– Point-to-point
– Rings
• Next Generation SONET
– Why NG-SONET?
– Virtual Concatenation (VCAT)
– Link Capacity Adjustment Scheme (LCAS)
– Generic Framing Procedure (GFP)

04/30/2006 Page 2

1
 Introduction
• SONET: Synchronous Optical NETwork

• Standard for optical telecommunications

transport

• SONET defines optical carrier (OC) levels

and electrically equivalent synchronous
transport signals (STS) for fiber-optic—
based transmission hierarchy

04/30/2006 Page 3

Introduction
• Developed originally by Bellcore

• Standardized by Exchange Carriers

Standards Association (ECSA) for
American National Standard Institute
(ANSI T1.105)

• Outside US, maintained by International

Telecommunication Union (ITU)

• European equivalent to SONET: SDH

(Synchronous Digital Hierarchy)

04/30/2006 Page 4

2
 Introduction
• SONET’s current place in optical network
– Widely deployed

– Ideal for back haul core networks, regional

networks, and metro networks

– Interconnects switches and routers for transport

purposes

– Rarely used within customer premises/access

networks (Ethernet, PON, DSL are more
economic)

04/30/2006 Page 5

Advantages of SONET
• Increased bandwidth over traditional
telecomm system

• Increased configuration flexibility that

supports future applications, with variety
of transmission rates

• Reduction in equipment requirement

• An increase in network reliability

• And many more …

04/30/2006 Page 6

3
 SONET Protocol

04/30/2006 Page 7

SONET: A layered Architecture

• SONET’s layered architecture includes a physical (optical)
layer, section layer, a line layer, and a path layer.
Path Layer

Line Layer

Section layer

Path Frames Fiber, FSO Path

SONET Line Line
Line
Layers
Section Section Section Section
Physical Physical Physical Physical Physical
ADM, Optical Regenerator ADM Optical ADM, Optical
Switch or (pass through) Amplifier Switch or
Cross Connect Cross Connect
Path origin Intervening network Path Termination

SONET is a four layer architecture

04/30/2006 Page 8

4
 Defining Digital Signal ‘x’
• These signals are often multiplexed onto SONET

• DS0 is the base for the digital signal X series

• DS1, used as the signal in the T1 carrier, is 24 DS0

(64 Kbps) signals transmitted using pulse-code
modulation (PCM) and time-division multiplexing
(TDM)

• DS2 is four DS1 signals multiplexed together to

produce a rate of 6.312 Mbps

• DS3, the signal in the T3 carrier, carries a multiple

of 28 DS1 signals or 672 DS0s or 44.736 Mbps and
is the basis for SONET STS1
04/30/2006 Page 9

SONET Protocol
• Path Layer: Mapping of DS0, DS1, DS3,
FDDI to Payload

• Line Layer: Frame synchronization and

data multiplexing into a SONET frame

• Section Layer: Framing, error monitoring

and bit-scrambling

• Photonic Layer: Fiber characteristics,

encoding, transmitter and receiver types

04/30/2006 Page 10

5
 SONET Equipment
• In a simple SONET linear network
• PTE (path-terminating equipment) sits at each end,
takes sub-rate signals and multiplexes them into a
SONET frame
• Between PTE devices are several LTE (line-terminating
equipment) devices: an ADM (add/drop multiplexer)
– Instead of demultiplexing entire signal, as PTE does, it
pulls out only sub-rate services it needs and inserts
different signals
• If network is long enough, STE (section-terminating
equipment) will be in fiber. These repeaters take
incoming signal from one side and rebroadcast it out on
other side.
– Placement of STE in the network depends on strength of
signal being transmitted down the fiber

04/30/2006 Page 11

SONET Equipments
• PTE: Path Terminating Equipment
• LTE: Line Terminating Equipment
• STE: Section Terminating Equipment

04/30/2006 Page 12

6
 Basic SONET transport signals
• SONET super-rate transport signal hierarchy:

OC Level Line Rate SDH Transport SPE

Term Overhead Size* The large percentage
OC-1 51.840 Mbps N/A 1.728 Mbps 50.112 Mbps (3.3%) of bandwidth
Not All values devoted to overhead is
Of N allowed OC-3 155.520 Mbps STM-1 5.184 Mbps 150.336 Mbps referred to as the
OC-12 622.080 Mbps STM-4 20.736 Mbps 601.344 Mbps “SONET overhead tax”
OC-24 1.244 Gbps N/A 41.472 Mbps 1.203 Gbps
OC-48 2.488 Gbps STM-16 82.944 Mbps 2.405 Gbps
OC-24 OC-192 9.963 Gbps STM-64 331.776 Mbps 9.622 Gbps
Rarely used
OC-768 39.813 Gbps STM-256 1.327 Gbps 38.486 Gbps

• All OC-N, N>1, are multiples of OC-1 (51 Mbps)

• An STS-N is the electrical equivalent of an OC.
*Gross capacity, includes path overhead and any fixed stuff bytes

SONET Transport Overhead consumes an

appreciable amount of bandwidth

04/30/2006 Page 13

Basic SONET transport signals

• SONET sub-STS-1 (Virtual Tributary) transport signal

hierarchy:
SDH SPE VTs per
VT Level Line Rate
Term Columns VTG*
VT1.5 1.728 Mbps TU-11 3 4
VT2 2.304 Mbps TU-12 4 3
VT3 2.456 Mbps N/A 6 2
VT6 6.912 Mbps TU-2 12 1

• The VT hierarchy was based on the TDM hierarchy and

therefore not friendly to transporting Ethernet-based client
signals. However, next generation SONET adds features that
overcome this limitation.

SONET Virtual Tributaries are primarily sized

To accommodate the TDM hierarchy

04/30/2006 Page 14

7
 Digital Hierarchy
North America International
Bit Rate (Mbps) Name Bit Rate (Mbps) Container Transport Name Bit Rate (Mbps) Container Transport

40000 STS/OC-768 STM-256

10000 STS/OC-192 STM-64
2500 STS/OC-48 STM-16
622 STS/OC-12 STM-4

155 STS/OC-3 STM-1

140 E4 140 VC4

51

45 DS-3/T3 45 STS/OC-1 STS-1 SPE

34 E3 34 VC3

8 E2 8

6 DS-2/T2 6
E1 2 VC12
2

1.5 DS-1/T1 1.5

VT1.5
0.064 0.064 E0 0.064

STS-1: Payload 49.536 Mbps + Overhead 2.304 Mbps

(4.5 %)
Total = 51.84 Mbps

04/30/2006 Page 15

SONET: A robust, layered optical transport

architecture and technology

ANSI T1X1.5
Optical Hierarchical interfaces

T1.105 T1.105.01 T1.105.02 T1.105.03 T1.105.04

SONET SONET APS SONET Jitter at DCC
Basic Payload Network Protocols
Description Mappings Interfaces & Arch.

T1.105.05 T1.105.06 T1.105.07 T1.105.08 T1.105.09

Tandem Physical Sub-STS-1 In-band Timing
Connection Layer Interface FEC And
Maintenance Specification Rate and Specification Synchronization
Formats
APS, Automatic Protection Switching
DCC, Data Communications Channel
FEC, Forward Error Correction

04/30/2006 Page 16

8
 SONET Frame Structure
• Basic SONET frame structure
• Signal known as Synchronous Transport
Signal-1 (STS-1)

04/30/2006 Page 17

SONET Frame: STS-1

• It consists of 9 rows of 90 bytes i.e. total
of 810 bytes
• It is transmitted from left to right and top
to bottom
• Transmission takes place serially
• Frame length is 125μs (i.e. 8000 frames
per second)
• STS-1 has a bit rate of 51.84Mbps

04/30/2006 Page 18

9
 STS-1 Frame
• First three columns of SONET
frame are called Transport
Overhead (TOH)
• The remaining 87 columns are
called Synchronous Payload
Envelope (SPE): data
• The first column of SPE is
called Payload Overhead
(POH)
• Note: Every SONET frame
repeats every 125μs no
matter how fast line speed
gets. As line rate goes up
frame gets bigger, to keep
frame rate at 8000 frames per
second
04/30/2006 Page 19

Byte Functions
• Section Overhead
– A1 and A2: Framing bytes
– E1: used for a voice channel between two technicians
as they install and test an optical link
• Line Overhead
– H1 and H2: Pointers. The SPE can be floating in a
SONET frame. It can start in one frame and end in the
next frame.
• These two bytes are allocated to a pointer that indicates
the offset in bytes between the pointer and the first byte
of the STS SPE
– K1 and K2: Automatic Protection Switching. These
bytes used to indicate defects, alarms detected at
receiving terminal back to corresponding transmitting
terminal through protection channels
04/30/2006 Page 20

10
 SONET Interleaving
• If N STS-1s are byte interleaved, the resulting
frame is called STS-N.
• For example for STS-3, byte interleaving means
1st byte of 1st STS-1 (called A1) is transmitted,
then A1 byte of 2nd STS-1, then A1 byte of 3rd
STS-1 ...

04/30/2006 Page 21

More on STS Frame

• Data can come into a device slower or faster
than it is transmitted out at the other side
• SPE can be floating and the pointers give the
position from where it is going to start
• H1 and H2 bytes are the pointer bytes

• Bits 1 to 4 are New Data Flag (NDF) bits

represented by Ns. It is set to 0110 for
normal operation. Bits 5 and 6 are currently
undefined in SONET

04/30/2006 Page 22

11
 Floating SPE
• The value of the bits 7 to 16 can vary from
0 to 782
– A value of 0 indicates that SPE starts at the first
byte immediately after H3 byte
– If the pointer value is 1 the payload starts at
the second byte after H3 etc.

04/30/2006 Page 23

SONET Networks
• Types:
– Linear
• Easy to design and install
• Do not offer redundancy as in rings

– Ring
• Working fiber
• Protection fiber

– Combination

04/30/2006 Page 24

12
 SONET Rings
• Unidirectional Path
Switched Ring (UPSR)
– Working and protection fiber
– Working traffic travels in one
direction
– Protection traffic in another
direction
– Signal from A to B would travel on
working fiber from A to B
– From B to A, signal would travel
from B to C to D to A
– If a unidirectional ring fails, it will
switch transmission from the
working fiber to the protection
fiber and it will travel in opposite
direction, so transmission from A
to B would be A to D to C to B

04/30/2006 Page 25

BLSR
• Bi-directional Line
Switched Rings (BLSR)
– Working and protection
traffic can travel in both
directions
– Signal from Node 0 to Node 1
will travel clockwise
– Signal from Node 1 to Node 0
will travel anti-clockwise
– In both clockwise and anti-
clockwise fiber, half of
capacity is reserved for
protection traffic and other
half for working

04/30/2006 Page 26

13
 Protection Switching BLSR

Normal operation: Traffic After failure: Traffic between Node 0 to

between Node 0 to Node 2 Node 2 is carried to Node 3 on
protection fiber and switched back to
working at Node 3 (Link restoration)

04/30/2006 Page 27

Next Generation SONET

ƒ New development of SONET/SDH:
Next generation SONET/SDH (NGS)

ƒ Next-generation SONET is evolution and enhancement of

existing SONET for network efficiency while increasing
broadband service potential.

ƒ Three next generation SONET capabilities are featured:

– VCAT (Virtual Concatenation)
• Provides efficient matching of SONET payload
bandwidth to service requirements.
• Only required at service ingress and egress points.
– LCAS (Link Capacity Adjustment Scheme)
• Technique to dynamically adjust bandwidth
provided for the service.
– GFP (Generic Framing Procedure)
• Updates SONET protocol to provide mapping for
virtually any kind of service into SONET. Only
required at service ingress and egress points.

04/30/2006 Page 28

14
 SONET
Virtual Concatenation
(VCAT)

04/30/2006 Page 29

SONET Channelization and

Concatenation
• SONET provides three options for payload granularity
– Channelized, individual STS-1s and Virtual Tributaries (VTs)
• Each STS-1 is 51.840 Mbps, with 50.112 Mbps of data capacity and 1.728
Mbps of overhead.
• VTs come in 4 sizes: VT1.5 (1.728 Mbps), VT2 (2.304 Mbps), VT3 (~3.456
Mbps), and VT6.0 (~6.912 Mbps).
• Channelized SONET supports the traditional TDM hierarchy well.

– Contiguously Concatenated (CCAT) STS-1s

• Only five standardized sizes defined (STS-3c, 12c, 48c, 192c, 768c)
• The contiguously concatenated STS-1s must be adjacent.
• STS-Nc must begin at designated STS #s (alignment constraints).

– Virtual Concatenation, a new SONET Next Generation Feature

• relatively new (2001), very flexible, defined for either STSs or VTs.
• STSs/VTs in a virtual concatenated SPE need not be adjacent.
• The size, in terms of the number of STSs is not restricted, it may be any
number of STSs, up to including 192.

04/30/2006 Page 30

15
 Channelization and Concatenation
Comparison
• An example comparison of channelized, contiguously concatenated, and
virtual concatenated SONET OC-12s.
A STS-1-Xv instance is known an a
Virtual Concatenation Group (VCG)

STS-1 #12 STS-1 #12

STS-1 #11 STS-3c #4 STS-1 #11
STS-1 #10 STS-1 #10 An STS-1
STS-1-5v
STS-1 #9 STS-3c must STS-1 #9 that is a
Start on specific member of
STS-1 #8 STS-3c #3 STS-1 #8
STS boundaries an STS-1-Xv
STS-1 #7 STS-1 #7 need not be
OC-12 (1, 4, 7, 10) STS-1-4v
STS-1 #6 STS-1 #6 contiguous
(1, 4, 6, 11)
nor must it
STS-1 #5 STS-3c #2 STS-1 #5 start on a
STS-1 #4 STS-1-3v STS-1 #4 designated
STS-1 #3 STS boundary
Within each STS-1 #3
STS-1 #2 STS-3c, STS-3c #1 STS-1 #2
STSs must be
STS-1 #1 STS-1 #1
contiguous
(A) (B) (C)
Fully Contiguously Virtual
Channelized Concatenated Concatenated
OC-12 OC-12c (VCAT) OC-12
(example)

04/30/2006 Page 31

CCAT, VCAT: impact on NEs

• Member STS-1s of a CCAT signal must travel together through the network,
requiring some functionality at every NE the CCAT group traverses in the
network.
• VCAT only requires functionality at the source and destination NE.
All STS-1s
Physical of the STS-N
Link (OC-N) must follow the CCAT
CCAT
same path CCAT
INE INE
CCAT OC-N CCAT Destination
Every NE NE
Source STS-Nc
Must have CCAT
NE INE
functionality
A) CCAT in the Network
No VCAT
functionality STS-N-3v STS #1
required in INE INE
INE INE
VCAT
STS-N-3v STS #2
VCAT INE
OC-N Destination
INE
STS-N NE
Source -3v S The STS-1s
NE TS #3
INE of the STS-N-Xv
may take different
B) VCAT in the Network paths
INE, Intermediate Network Element

04/30/2006 Page 32

16
 Why VCAT?
• VCAT facilitates efficient transport of many non-TDM
client signals that contiguous concatenation doesn’t
transport well.
Client Signal Contiguously Concatenated Virtually Concatenated
Best fit Container % Best fit Container %
Name Rate container Rate Efficiency container Rate Efficiency
Fast Ethernet 100 Mbps STS-3C 155 Mbps 64.5% STS-1-2v 104 Mbps 96.2%

ESCON
200 Mbps STS-12C 622 Mbps 32.1% STS-1-4v 207 Mbps 96.6%
Fibre Channel
New Format 274 Mbps STS-12C 622 Mbps 44.1% STS-3c-2v 311 Mbps 88.1%
New Format 500 Mbps STS-12C 622 Mbps 80.4% STS-1-10v 518 Mbps 96.5%

Fibre Channel 2.488 1.089

1 Gbps STS-48C 40.2% STS-3c-7v 91.8%
Gigabit Ethernet Gbps Gbps

New Format 3 Gbps STS-192C 30.1% STS-3c-20v 3.1 Gbps 96.8%

9.953
New Format STS-192C Gbps 60.3% STS-3c-39v 6.1 Gbps 98.4%
6 Gbps

Compare how much

% Transport Efficiency = client rate / container rate * 100 More efficient
VCAT is

04/30/2006 Page 33

Why VCAT - CCAT Inefficiency Example

• The inefficiency of CCAT; an example using four Gigabit

Ethernet client signals transported using STS-48c signals.

60% unused
Client CCAT Transport Transport
Signals Container Signal 192
10 Gbps
STS-48C
#4
1 Gbps #1 #1 STS-48C 144
Unused
1 Gbps #2 CCAT-Based Payload
STS-48C
#2 #3 STS-48C OC-192 #3
1 Gbps #3 96
4 Gbps STS-48C
1 Gbps #4 #4 STS-48C 1 Gbps #4
3 Gbps #2
1 Gbps #3 48
2 Gbps STS-48C
1 Gbps #2
1 Gbps #1
1 Gbps #1
Payload STS-1s

Only 40% of OC-192 total payload utilized

but all 192 STS-1s are expended

04/30/2006 Page 34

17
 Why VCAT?
• The efficiency of VCAT; an example using Gigabit Ethernet
• Using the same size (OC-192) contained as the preceding CCAT
example, much more can be transported used VCAT.

Unused
Payload 192
Client VCAT Transport 2 x STS-1-2v
2 x 100 + STS-1-20v
Signals Container Signal 2 x 200 1 Gbps #9 +
STS-1-20v
1 Gbps #8 2 STS-1-4v
STS-1-2v STS-1-20v
1 Gbps #7 120
2 x 100Mbps STS-1-20v
1 Gbps #6 100
#9 #2 STS-1-20v STS-1-20v
1 Gbps #1 VCAT-Based 1 Gbps #5 80
#8 #1 #3 STS-1-20v 1 Gbps #4 STS-1-20v
OC-192 60
#7 #4 STS-1-20v 1 Gbps #3 STS-1-20v
40
1 Gbps #9 #6 #5 STS-1-20v 1 Gbps #2 STS-1-20v
20
2 x 200Mbps 1 Gbps #1 STS-1-20v
STS-1-4v Payload STS-1s

All 192 STS-1s expended,

96% of OC-192 total payload utilized

04/30/2006 Page 35

Why VCAT – CCAT vs. VCAT

Efficiency Comparison

• A side-by-side comparison of the CCAT and VCAT

examples 60% unused 4% unused
Unused
Payload 192
10 Gbps 192
STS-48C 2 x 100 + 2 x STS-1-2v
STS-1-20v
#4 1 Gbps #9
Unused 144 2 x 200 STS-1-20v +
STS-48C 1 Gbps #8
Payload STS-1-20v 2 STS-1-4v
#3 1 Gbps #7 120
96 1 Gbps #6 STS-1-20v
STS-48C 100
4 Gbps 1 Gbps #5 STS-1-20v
1 Gbps #4 #2 80
3 Gbps 1 Gbps #4 STS-1-20v
1 Gbps #3 48 60
2 Gbps 1 Gbps #2 STS-48C 1 Gbps #3 STS-1-20v
1 Gbps 40
#1 1 Gbps #2 STS-1-20v
1 Gbps #1 20
Payload STS-1s 1 Gbps #1 STS-1-20v
Payload STS-1s

4 GbE signals only 9 GbE signals

+
2 x 200 Mbps
+
1 x 100 Mbps

04/30/2006 Page 36

18
 Why VCAT - Example Summary
• The example shown in the preceding two slides articulates
the benefits of VCAT as compared to CCAT. In the same
amount of Bandwidth (OC-192) VCAT:
– Carries more client signals
– Has higher transport efficiency
Parameter CCAT VCAT
Gigabit Ethernets Transported 4 9
Total SPE bandwidth used by client signals 4 Gbps 9.6 Gbps
SPE bandwidth unused 6 Gbps 0.4 Gbps
Transport Efficiency 40% 96%
STS-1s Utilized 192 192

VCAT Is Superior To CCAT For

Non-TDM Client Signal Transport

04/30/2006 Page 37

Virtual Concatenation, VCAT

• Flexible containers for SONET, SDH, and OTN

ITU-T SG15

G.vcm G.707 11.2, 11.4 G.709 18.1

Virtual Concatenation (VCAT for SDH) (VCAT for OTN)
Management

4Q2005

ANSI T1X1.5
Optical Hierarchical interfaces

T1.105
7.3.2, 7.3.3 OTN, Optical Transport Network
Virtual
SDH, Synchronous Digital Hierarchy
Concatenation
VCAT, Virtual Concatenation
(For SONET)

04/30/2006 Page 38

19
 Virtual Concatenation
No Concatenation
1050 Mbps
2.5 Gbps OC-
Gig Ethernet Data 48/STM-16 low
(1.0/1.2 Gbps) efficiency

196 Mbps 622 Mbps OC -

ESCON 23/STM-4 low
SAN efficiency
(160/200 Mbps)
150 Mbps
155 Mbps STM-1
OC-3/STM-1 TDM high efficiency
(155 Mbps)

VCAT
STS-3c-7v/VC-4-7v
Gig Ethernet (1050 Mbps)
Data
(1.0/1.2 Gbps)
STS-1-4v/VC-3-4v 2.5 Gbps OC-
ESCON (196 Mbps) 48/STM-16 packed
(160/200 Mbps) SAN at nearly 88%
efficiency
STS-3/STM-1
OC-3/STM-1 (150 Mbps)
(155 Mbps) TDM

04/30/2006 Page 39

SONET VCAT Summary

• Pro
– VCAT is considerably more efficient in transporting datacom
client signals than channelized or contiguously concatenated
SONET.
– VCAT flexibility (non-contiguous STSs, any STS starting
boundary, and variable bandwidth) is a big plus, especially in
networks where proprietary or non-telecom client signals must
be transported.
– Also works well for TDM hierarchy.
• Con
– Relatively new, availability in COTS products, test equipment,
and components not widespread (but growing fast).
– Must compensate at the terminating Network Element for the
differential delay due to the different paths taken by the
individual STS-1 or -3 members of a STS-1 or 3 –Xv signal.

04/30/2006 Page 40

20
 Link Capacity Adjustment
Scheme (LCAS)

04/30/2006 Page 41

Link Capacity Adjustment Scheme

• Link Capacity Adjustment Scheme (LCAS) is a next

generation SONET feature that allows VCAT Groups
(VCGs) to be resized dynamically.
• This allows network providers to more quickly respond to
customer requests for service bandwidth changes.
• LCAS provides a control mechanism for both changes (VCG
size increase or decrease) on demand, and in response to
failures of individual STS-1s.
• This latter feature allows the network provider to quickly
substitute an unused STS-1 for a failed STS-1, sustaining
the SLA-contracted service in the face of the failure.
• Basically, LCAS is a set of commands to be implemented
within SONET or OTN Network Elements that support VCAT

04/30/2006 Page 42

21
 LCAS Architecture

• LCAS Commands are unidirectional and exchanged

between the VCG source (So) NE and sink (Sk) NE

Sk NE for So NE for
VCG X VCG X
Member P Member P
VCG X Member P

NE NE

VCG A Member N
Sk NE for
So NE for VCG A
VCG A Member N
Member N

04/30/2006 Page 43

LCAS States & Control Messages

• LCAS So and Sk NEs implement a standards-defined state
machine and exchange five different command types
Control Messages:
• FIDLE: not a group member, no adds pending
Sk NE for • FADD: request to add member to a group So NE for
• FDNU: request to delete member from group
VCG X VCG X
• FEOS: Member has highest sequence # in group.
Member P • FNORM: member is part of group Member P

VCG X Member P
NE NE

VCG A Member N
So NE for Sk NE for
VCG A VCG A
Member N Member N

So States: Sk States:
• IDLE (not provisioned as part of VCG) IDLE (not provisioned as part of VCG)
• NORM (provisioned, good path to Sk ) OK (incoming signal to Sk is good )
• DNU (provisioned, bad path to Sk) FAIL (Incoming signal to Sk is bad)
• ADD (In process of being added to VCG)
• REMOVE: (Being removed from VCG)

04/30/2006 Page 44

22
 LCAS Parameters

• LCAS uses three parameters to describe a VCG

– The XMAX parameter indicates the maximum size of a VCG in
terms of number of members.
– The XPROV parameter indicates the number of provisioned
members in a VCG
• Each ADD command increments XPROV by 1
• Each REMOVE command decrements XPROV by 1
• 0 ≤ XPROV ≤ XMAX
– The X parameter indicates the actual number of VCG members.
• 0 ≤ X ≤ XPROV ≤ XMAX

04/30/2006 Page 45

LCAS – What it doesn’t do

• Because LCAS is a control mechanism operating

between So and Sk NEs, a Network Management
System (NMS) is needed to:
– Create the VCG
– Assign (provision) a value for the VCG XMAX parameter.
• XMAX value should include “spare” members for use as
substitutes when paths fail.
• XMAX value should also include spares for operator initiated
VCG capacity changes.
– Initiate VCG capacity changes not resulting from a
member failure.

04/30/2006 Page 46

23
 Link Capacity Adjustment Scheme, LCAS

• Dynamic adjustment of transport bandwidth

ITU-T SG15

G.7042 G.707 G.709 18.3

LCAS 11.2.1, 11.4.1 LCAS for OTN
LCAS for SDH

ANSI T1X1.5
Optical Hierarchical interfaces

T1.105 7.3.4 LCAS, Link Capacity Adjustment Scheme

OTN, Optical Transport Network
LCAS For SDH, Synchronous Digital Hierarchy
SONET SONET, Synchronous Optical Network

04/30/2006 Page 47

Generic Framing Procedure

(GFP)

04/30/2006 Page 48

24
 GFP Introduction

• All optical network client signals must be mapped in some way to an

underlying optical transport payload container in order to transport the
client through the network.
• Typically, past mapping methods have been either fixed mappings or
very narrow in scope
– For example, Packet On SONET (POS), maps only IP packets over
PPP in HDLC like framing.
• GFP is more efficient and effective; its two modes can accommodate
many different client signals
• GFP can be considered to be a shim layer between the data link and
physical layers of the OSI reference model (as can any mapping
mechanism).
• GFP is not a protocol. There are no peer-to-peer interactions between
network elements.

04/30/2006 Page 49

GFP Flexibility

• GFP performs adaptation for both constant bit rate and PDU-oriented
client signals. These two modes, GFP-T and GFP-F make GFP flexible
from a client transport perspective.
• GFP can be mapped to a number of transport mechanisms. This flexibility
gives customers a straight-forward and standards-compliant migration
path from SONET to G.709.
• GFP’s payload area can accommodate 4 to 65,535 octets, this payload
size flexibility is an advantage when adapting different client signals.
• GFP’s header extension provides a field for Channel IDs, allowing
identification of up to 256 separate channels.
• When combined with SONET virtual concatenation (VCAT), the bandwidth
offered to clients can be any multiple of VTs or STSs.

04/30/2006 Page 50

25
 GFP Introduction

• Generic Framing Procedure (GFP) is an International

Telecommunications Union (ITU-T) recommendation. Its current version
is G.7041 (12/2003).
• GFP delineates octet-aligned variable length payloads from higher-level
client signals for mapping into octet synchronous paths such as
SONET/SDH (T1.105/G.707) and OTN ODUk paths (G.709).
• Client signals may be PDU-oriented, such as Ethernet, IP/PPP, or block-
code oriented constant bit rate (CBR) streams such as Fiber Channel or
Escon.

Other Client
Ethernet IP/PPP
Signals
Clients
GFP - Client-Specific Aspects GFP is Flexible,
(payload dependent)
GFP Serve many different clients
GFP - Common Aspects Maps to many transport paths
(payload independent)
VCAT
Other octet- OTN ODUk
SONET Synchronous paths Path (g.709) Transport
GFP architecture and relationship to client signals and transport paths*
*From ITU-T G.7041

04/30/2006 Page 51

GFP Client Signal Adaptation

• GFP supports two client signal adaptation modes:
– Frame-Mapped GFP (GFP-F), is a PDU-oriented adaptation mode.
– Transparent GFP (GFP-T), is a block-code oriented adaptation mode.
• GFP-F requires visibility of the client PDU and operates at the data link
layer (or higher).
• GFP-T operates on a coded character stream. For GFP-T, the same
PHY-X interface must exist at both ends of the path (A, A’)

Application Layer
Bridge, switch, or router Bridge, switch, or router

A GFP-T GFP-F GFP-F GFP-F GFP-F GFP-T A’

PHY-T PHY-X PHY-X PHY-T

10BASE ESCON 10BASE

ESCON Transport Network
100BASE FICON 100BASE
FICON
1000BASE Fibre Channel 1000BASE
Fibre Channel
10GBASE 1000BASE 10GBASE
1000BASE
GFP Functional Architecture *
*From ITU-T G.7041

04/30/2006 Page 52

26
 GFP Frame Format
• GFP frames consist of a core header and a payload area.
• GFP idle frames do not have a payload area.
16 bit Payload
Length Indicator

cHEC
(CRC-16)

1 Flexible Header
Core Header 4 Format is a
GFP advantage
Payload
Headers
Payload (4-64 bytes)
4-65535
Area
Flexible payload FCS helps with
size is a GFP End-to-end
advantage Payload Performance Monitoring
n
Area And Troubleshooting
Octet 1 8
Bit
Optional
Payload FCS 4 Octets
(CRC-32)

GFP Frame Format* *From ITU-T G.7041

04/30/2006 Page 53

GFP Client Management Frame

• A GFP client management frame (CFM) may be used to

transport client signal management and control
information.
• Illustrated below is an example of a CMF that carries a
sequence number field and a count of user frames between
marker field.
• A CFM could be used, for example, to transport client-
specific signaling.

04/30/2006 Page 54

27
 Adding GFP On SONET
• An example illustrating where GFP exists within the SONET
architecture.
• In this case, GFP-F is being used for a PDU-oriented client signal, and
is transporting this client over a SONET transport mechanism.

Service Termination Network Service

Termination

Client GFP, Client

Protocol ITU-T G.7041 Protocol
Stack Stack
GFP-mapped SPE, end-to-end
GFP-F GFP-F
Path Path

SONET Line Line Line

Layers Section Section Section Section Section
Physical Physical Physical Physical Physical

MSPP, ADM or Regenerator MSPP, ADM or Regenerator MSPP, ADM or

Cross Connect Cross Connect Cross Connect
(Origination) (Pass Through) (Termination)

04/30/2006 Page 55

Generic Framing Procedure, GFP

• A mapping procedure for optical networking

ITU-T SG15

G.7041 G.707 10.6 G.709 18.2.4

GFP GFP for SDH GFP for OTN

ANSI T1X1.5
Optical Hierarchical interfaces

T1.105.02 6.2.3 STS-1 Level Mapping of GFP Frames

6.3.1.4 STS-1 Mapping of GFP Frames
SONET 6.3.2.3 STS-N Mapping of GFP Frames
Payload 6.3.3.3 STS-1-Xv and STS-3c-XV Mapping of GFP Frames (i.e. to VCAT)
Mappings
GFP: Generic Framing Procedure; OTN: Optical Transport Network; SDH: Synchronous Digital Hierarchy

04/30/2006 Page 56

28
Differentiation Among Metro Aggregation

Legacy Emerging
Multi- Service VP
Capabilities Ring Proprietary
VC
SONET MSPP
RPR
GFP

GbE
POS
Bandwidth
Efficiency
Removing Stat. Mux. Stat. Mux.
stranded over indiv. for multi-
capacity channels end points

04/30/2006 Page 57

29