ng.sdh.slides.e

Presentation extracted from the book:“Migration to Next Generation SDH”

ISBN 84-609-4420-4see at wwww.trendcomms.com

Migration to Next Generation SDH

by José M. Caballero [email protected]

Section

SDH classic

3/58© 2005 Trend Communications - www.trendcomms.com

Reasons to define SDH

•••• The Antitrust law in the US followed by Bell being broken into smaller companies

•••• It was necessary to interconnect new PTTs: SONET definition

•••• B-ISDN specification to integrate any traf-fic: SDH and ATM standardisation

•••• Advanced management needs: comput-ers and telecom must work together

•••• Requirement for having new infrastruc-tures manage any type traffic: data, voice, multimedia


bytes vs. bits

Standardised since 1988 when the G707, G708, G709 CCITT recommendations appeared

•••• SDH is byte oriented, it means that a byte is the unit for mapping and multiplexing

•••• STM-N is the name for the transport frames. They have always a period of 125µs

•••• An important consequence is that in SDH 1 byte represents a 64 Kbit/s channel

125 µs

0 n1 byte

rate =8 bits

125·10-6seg.= 64Kbit/s

0

125 µsframe 1 frame 2


SDH is a flexible architecture •••• direct internet working between equipment

•••• scalability up to 10 Gbit/s

•••• direct add & drop for low speed tributaries

•••• Rich in OAM functions

•••• Support to fit any application

•••• remote and centralised management

•••• Fault tolerance: fast traffic routing in case of faults

•••• SDH is highly compatible with SONET

•••• very efficient in managing circuits

•••• fast circuit definition from a centralised point

•••• advanced facilities for quality monitoring


Classic SDH: Circuit provisioning

SDH provides efficient, reliable and flexible transport for circuits

multiplexing, transport, routing, management, reliability

Internetservices Frame Relay ATM GSMPOTS

transport networkSDH

transmission media cable/fibre/radio


Classic SDH Network elements

STM-NSTM-NREG

SDHMUX

STM-MSTM-N

STM-N

M >N

Regenerators Multiplexers

STM-M STM-M

STM-N, PDH

West East

Add & Drop Multiplexer (ADM) Digital Cross-Connect (DXC)

STM-N

STM-N

STM-N

STM-N


Transport design

National Backbone

Primary Network

Access Network

STM-16

STM-4

STM-1 or PDH


Nothing compares with SDH resilience

protection ringservice ring

ADM

ADM

AD

MADM

ADM

ADM

service &

ADM

ADM

AD

M

MUX

1:1

high priority

1:N

1+1MUX

MUXMUX

MUXMUX

protection rings

low priority


The SDH multiplexing map


The STM-1 frame

•••• STM-1 = AUG + RSOH + MSOH

•••• RSOH and MSOH sections allow the traffic management

•••• The VC4 is floating inside the STM-1: VC4 is asynchronous STM1 is synchronous

•••• The AU pointer always points to the position where the VC4 starts and follows possible fluc-tuations

RSOH

MSOH

AU4 ptr

1

VC4

27010

STM-1

HP POH

TU pointers

VCn VCn

LP POHLower Order Path Overhead

500 µs375 µs

250 µs125 µs

MUX

STM-1

3:1

STS-1STS-1STS-1

G1F2H4

J1B3C2

F3K3N1

ptr

V5

V5

ptr

K4N2

J2V5

9

1

270101 9


OAM: Detected events at the end of a RS

LOS

LOF

MS-AIS (A1, A2 OK; the rest all 1s)

AU-AIS(All 1s)

TU-AIS(All 1s)

PDH AIS(All 1s)

B1 w ith errors

MS-RDI(K2=xxxxx110)

HO-RDI(G1=xxxx1xxx)

LO-RDI(V5=xxxxxxx1)

M U X REGLO-PTE HO-PTE

2M

34M

140M

STM-1 STM-NLO-PTEH O-PTE

2M

34M

140M

STM-1

HO-PTE LO-PTEM U X

MULTIPLEXER

SECTION

SECTION

HIGH ORDERPATH

LOW ORDERPATH

REGENERATORSECTION

REGENERATOR

Section

Next Generation SDH

NG SDH = classic SDH + [GFP+VCAT+LCAS]


Streaming forces

New services and applications based on internet, mobile, multimedia, DVB, SAN, Ethernet or VPN, are demanding long hault transport. State of the art:

1. Ethernet, are in an early stage of development for efficient optical transport

2. Most have their transport infrastructure entirely based on SDH/SONET.

3. There is a lot of experience in managing SDH/SONET.

No other technology than SDH/SONET has this maturity grade at the optical physical layer.

TVVoIP

VPN

3G

InternetCircuit

xDSL

Mobile VPLSFRL VoD

ISDN

TVVoIP

VPN

3G

InternetTele

Circuit

Mobile MPLSFRL Gaming

ISDN

phone

VoIPTriplePlay

ISDNVoD


Next Generation SDH: Packet Friendly

SDH/SONET has also evolved to more efficiently adapt statistical multiplexing traffic based on data packets.

TVVoIP

VPN

3G

Internet

SANCircuit

xDSL

Mobile VPLSFRL

VoD

ISDN

TV

Fibre ChannelESCON

LCAS

VoIP

VPN

3G

Internet

GFP-F

SDH/Sonet NG

PDH

Tele

WDM - Dark Fibre - Coax - Wireless - OTN

VLAN

Contiguous Concatenation Virtual Concatenation

MPLS

10/100/1000 Ethernet

FICON

DVB

GFP-T

SANCircuit

ATM

HDLC/PPP/LAPS

Mobile MPLSFRL

3play

ISDN

IPIP

Services

Transport Network

Transmission Media

phone


Legacy and Next Generation SDH

The Next Generation of SDH/SONET is offering:•••• data-packet interfaces

•••• TDM interfaces

•••• new functionalities

The objective is:

•••• to support efficiently any type of traffic (including of data packets)

•••• get the best of Legacy SDH including:

- resiliency,

- reliability,

- scalability,

- centralised management

- rerouting


Key NG SDH/SONET

GFP (Generic Framing Protocol)is robust and standardized encapsulation procedure for the transport of packetised data on SDH/SONET. In principal, GFP performs bit rate adaptation managing features such as priorities, discard eligibility, transport channels selection, and submultiplexing.

VCAT (Virtual Concatenation)is a mechanism which assigns granular bandwidth sizes rather than the exponential provision of the Contiguous Concatenation. Therefore VCAT is more flexible and efficient.

LCAS (Link Capacity Adjustment Scheme)LCAS can modify dynamically the allocated VCAT bandwidth by adding/removing members of a pipe in use. LCAS is being used also to implement diversity for traffic resilience.

LCAS

GFP-F NG SDH/SONET

Contiguous Concatenation Virtual Concatenation

GFP-T


Next Gen. SDH Network elements

Legacy NE: Reg, Mux, ADM, DXC Multiservice Provisioning Platform (MSPP)

Multiservice Transport Platform (MSTP)

EthernetVPNDVBSAN

PDH

SDH

DWDM

Multiservice Switching Platform (MSSP)

SDHDWDMSDH

DWDMSDH

(classic ADM + packet interfaces)

(DXC-like)(MSPP + DWDM)

VPNDVBSAN

PDHEthernet

SDH

Client interfaces

SDH


Multiservice Provisioning Platform (MSPP)

A Multiservice Provisioning Platform (MSPP) is basically the result of the evolution of legacy ADM and TDM interfaces and optical interfaces, to a type of access node that includes a set of:

•••• legacy TDM interfaces

•••• data interfaces, such as Ethernet, GigE, Fiber Channel, or DVB

•••• NG SDH/SONET functionalities such as GFP, VCAT and LCAS

•••• optical interfaces from STM-0/STS-1 to STM-64/OC-192

EthernetVPNDVBSAN

PDH

SDH

(classic ADM + packet interfaces)


Multiservice Transport Platform (MSTP)

A Multiservice Transport Platform (MSTP) bassically is a MSPP with DWDM functions to drop selected wavelengths at a site that will provide higher aggregated capacity to multiplex and to transport client signals.

MSTP allows to integrate SDH/SONET, TDM and data services, with efficient WDM transport and wavelength switching.

Typically, MSTPs are installed in the metro core network.

DWDMSDH

(MSPP + DWDM)

VPNDVBSAN

PDHEthernet


Multiservice Switching Platform

A Multiservice Switching Platform (MSSP) is the NG equivalent for cross-connect, performing efficient traffic grooming and switching at STM-N/OC-M levels but also at VC level.

MSSPs should support more than just data service mapping, namely true data services multiplexing and switching.

MSSP is still emerging as a NG Network element, while MSPP and MSTP are quite mature.

DWDMSDH

DWDMSDH

(DXC-like)


Generic Framing Protocol (GFP)

•••• GFP, defined in ITU-T G.7041, provides data rate adaption and frame delineation

•••• There are two mapping service for data protocols:

1. GFP-T (Transparent) is a layer 1 encapsulation in constant sized frames. Optimised for traf-fic based on 8B/10B codification such as 1000BASE-T, Fibre Channel, and ESCON.

2. GFP-F (Framed) is a layer 2 encapsulation in variable sized frames. Optimised for data packet protocols such as DVD, PPP and Ethernet.

.

.

.

.

.

.

Ethernet SDH/Sonet Ethernet

GFP GFP

Source Sink

STM-n/OC-m

Encapsulating

T r a n s m i s s i o n

MappingMultiplexing Demultiplexing

DemappingDe-encapsulating

.

.

.

Tx

Port 1Port 2

Port n

.

.

.

Rx

Packets

Port 1Port 2

Port n

MSPP

queues demapper

MSPP

mapper Queues


Frame-Mapped GFP (GFP-F)

•••• The entire client packet is dropped into a GFP frame

•••• Data Client signals such as Ethernet, PPP and DVB are queued waiting to be mapped

•••• Some codes can be removed to minimize the transmission size

•••• GFP-F supports submultiplexing onto a single channel for low-rate sources

GFP-F results in a more efficient transport, however, the encapsulation processes described above increase latency, making GFP-F inappropriate for time-sensitive protocols.

Sink

Source Sink

CIDSubmultiplexing

STM-n/OC-m

Encapsulation Mapping Multiplexing Demultiplexing Demapping Dencapsulation

......

RxPacketsqueues demappermapper

1 34

GFP

-F

GFP

-F

... ...

SA

N,

Fib

er

Ch

ann

el

SA

N,

Fib

er

Ch

an

ne

l


Transparent GFP (GFP-T)

•••• GFP-T client signals are mapped into fixed-length GFP framesand transmitted immediately without waiting for the entire client data packet to be received.

•••• GFP-T encapsulates any protocol as long as they are based on 8B/10B line coding, which is why it is often called protocol-agnostic.

•••• ALL the client characters, without exception, are transported to the far end.

•••• GF-T is very good for isocronic protocols and SAN, such as ESCON or FICON. This is be-cause it is not necessary to process client frames or to wait for arrival of the complete frame

Source Sink

Encapsulation Mapping Multiplexing Demultiplexing Demapping Dencapsulation

Eth

ern

et, P

PP, D

VB

Eth

ern

et, P

PP, D

VB

...

demapper

STM-n/OC-m

8B/10B mapper

...

Reassembly

GFP

-T

GFP

-TIndependent Streams

......


GFP mapping of client signals

Depending on the type of GFP in question, the mapping function can drop the whole signal (in the case of GFP-T), or can throw away certain delineation fields (GFP-F).

SoFDest Add

Source AddLength

LLC DataPadFCS

FlagAddressControlType

PayloadPadFCS

Extension

FlagPad

Preamble

Ethernet

HDSLC/PPPP

BBWSoF

HeaderDataCRCEOF

LLC/SNAP

Fiber Channel

Core Header

Payload Header

Extension Header

Payload

Check Sum

thrown awaytransported

GFP

GFP-F

GFP-F

GFP-F


GFP frame formats and protocols

1

3

5

7

9

+1

+n+4

PTI

byte

PFI EXI typeUPI

PLI

cHEC (CRC-16)

tHEC (CRC-16)

EXI

eHEC (CRC-16)

pFCS (CRC-32)

Payload

+n

66

Core Header

Payload Header

Extention Header(optional)

Payload

Check Sum(optional)

GFP frame

PLI: PDU Length IndicatorcHEC: Core HEC protectionPTI: Payload type Identifier

000: client data100: client management

PFI: Payload FCS Indicator1: presence of FCS0: absence

EXI type: Extension Header Identifier0000: Null0001: Linear0010: Ring

UPI: User Payload Identifier (PTI=0)

Payload: Space for the framed PDUpFCS: Payload FCS

2

4

6

8

+2

01x: Ethernet (GFP-F)02x: PPP (GFP-F)03x: Fiber Channel (GFP-T)04x: FICON (GFP-T)05x: ESCON (GFP-T)06x: Gigabit Ethernet (GFP-T)08x: MAPOS (GFP-F)09x: DVB (GFP-T)0Ax: RPR (GFP-F)

0Cx: Async Fiber Channel (GFP-T)

scra

mbl

ed b

ytes

(opt

iona

l)

0Bx: Fiber Channel (GFP-F)

eHEC: Extension HEC protection

EXI: Extension Header Identifier

4 bytes

4 bytes

0-60 bytes

n bytes

0-4 bytes

tHEC: Type HEC protection

CIDSpare

Line

ar E

XI t

ype

Nul

l E

XI

tHEC: Type HEC protectionCID: Channel ID for submultiplex

tHEC

TypeType

tHEC

eHEC


GFP-F vs. GFP-T

Feature GFP-F GFP-T

Protocol transparency low high

Efficiency high low

Isocronic protocols no yes

Encapsulation protocol level Layer 2 (Frames) Layer 1 (Physical)

Optimized for Ethernet SAN, DVB

LCAS protection likely poor

Statistical submultiplexing yes no

SAN transport no yes

Ethernet transport optimum possible


Concatenation

Concatenation is the process of summing the bandwidth of X containers of the same type into a larger container.

There are two concatenation methods:

•••• Contiguous concatenation, which creates big containers that cannot split into smaller pieces during transmission. For this, each NE must have a concatenation functionality.

•••• Virtual concatenation, which transports the individual VCs and aggregates them at the end point of the transmission path. For this, concatenation functionality is only needed at the path termination equipment.

STM-1

STM-64

STM-16

STM-4

STM-256OC-768

OC-192

OC-48

OC-12

OC-3/STS-3VC-4

VC4-4c

VC4-16c

AU-4

AU4-4c

AU4-16c

STS-3c SPE

VC4-64c

VC4-256c

AU4-64c

AU4-256cAUG-256

STS-12c SPE

STS-48c SPE

STS-192c SPE

STS-768c SPE

STS-3c

STS-12c

STS-48c

STS-192c

STS-768c STS-768

AUG-64STS-192

AUG-16STS-48

AUG-4STS-12

AUG-1STS-3

x4

x4

x4

x4

40 Gbps

10 Gbps

2.5 Gbps

622 Mbps

155 Mbps

x1

x1

x1

x1

x1

x1

x1

x1

x1

x1

C-4-256c

C-4-64c

C-4-16c

C-4-4c

C-4


Contiguous and Virtual Concatenation

G1F2H4

J1B3C2

F3K3N1

G1F2H4

J1B3C2

F3K3N1

Bandwidth requirement

G1F2H4

J1B3C2

F3K3N1

SDH

G1F2H4

J1B3C2

F3K3N1

G1F2H4

J1B3C2

F3K3N1

G1F2H4

J1B3C2

F3K3N1

G1F2H4

J1B3C2

F3K3N1

G1F2H4

J1B3C2

F3K3N1

VC4-4c VC4-3v or STS-9v

Bandwidth delivery

One Path Several paths

(430 Mbps)

21

3 III

4

II

IV

622 Mbps

3 x155 Mbps

Contiguous concatenation Virtual concatenation

(3 in SDH, or9 in SONET)

One VCG

3 VC members


Contiguous Concatenation (VCAT)

Contiguous concatenation: Pointers and containers. A VC-4-Xc (X = 1, 4, 16, 64, 256) structure, where X represents the level. The increment/decrement unit (justification) is 3 X, as it depends on the level: AU-4=3 bytes, AU-4-256c=768 bytes.

VirtualContainer

VCtype

ContiguousConcatenation

Numberof VCs

G1F2H4

J1B3C2

F3K3N1

R

VC-4-Xc / STS-3Xc SPE

Fixed columns

261X

C-4-Xc

260X1

H1 Y Y H2 1 1 H3 H3 H3

H3

H1 Y Y....... H2 1 1....... H3 H3 H3.......

H2H1 1 3

1 9

1 36

3

12 24

H1 Y Y....... H2 1 1....... H3 H3 H3.......1 14448 96

AU-3 ptr

AU-4 ptr

AU-4-4c ptr

AU-4-16c ptr

H1 Y Y....... H2 1 1....... H3 H3 H3.......1 9X3X 6X

AU-4-Xc ptr

6

Y: 1001SS11

1: All 1s byte Justification unit = 3 X bytes size = X-1

(STM-0/STS-1)

(STM-1/OC-3)

(STM-4/OC-12)

(STM-16/OC-48)

(STM-n/OC-m)

STM / STS pointers

or SPE


Virtual Concatenation (VCAT)

Packet-oriented, statistically multiplexed technologies, such as IP or Ethernet, do not match well the bandwidth granularity provided by contiguous concatenation.

•••• VCAT is an inverse multiplexing technique that allows granular increments of bandwidth in single VC-n units.

•••• At the source node VCAT creates a continuous payload equivalent to X times the VC-n

•••• The set of X containers is known as a Virtual Container Group (VCG), and each individual VC is a member of the VCG.

•••• Lower-Order Virtual Concatenation (LO-VCAT) uses X times VC11, VC12, or VC2 contain-ers (VC11/12/2-Xv, X = 1... 64).

•••• Higher-Order Virtual Concatenation (HO-VCAT) uses X times VC3 or VC4 containers (VC3/4-Xv, X = 1... 256), providing a payload capacity of X times 48 384 or 149 760 kbit/s.

VirtualContainer

VCtype

VirtualConcatenation

Numberof VCs


Virtual Concatenation Operation

•••• MFI: Multiframe Indicator, VCG: Virtual Container group, SEQ: Sequence Number

•••• Virtual concatenation is required only at edge nodes

•••• Sink node must compensate for the different delays

VC3-4vVC3-4v

A B

FE

X YG

50%

25%VC3-4vVC3-4v

MFI=k+1

MFI=k

SQ=0..3

MFI=i

MFI=i-1

MFI=iSQ=0..3 t1

t1-125µst0+125µs

t0

Rate Adaption,VC3

555

555

Source Sink node

VC GroupContiguous PayloadsVC3 with SEQ=2 and SEQ=3

SDH/SONET

Delay adjust

VC Group Contiguous Payloads

SEQ=0

25%

MSSP

VC3

VC3

#0

VC3

#2

VC3#3

VC3#3

VC3#1

VC3#1

ReassemblyMapping, Segmentation

VC3#2

VC3#0

MSSP


Virtual Concatenation Graphically

G1

J1B3C2

1

G1F2H4

J1B3C2

F3K3N1

VC-3

871

VC-3G1F2H4

J1B3C2

F3K3N1

125 µs

AU3 ptrsRSOH

MSOH

MFI=0

STM-n

SEQ=0...X

9

11 n·270

AU3 ptrsRSOH

MSOH

MFI=1

STM-n

SEQ=0...X

2·n·270

AU3 ptrsRSOH

MSOH

MFI=2

STM-n

SEQ=0...X

3·n·270

AU3 ptrsRSOH

MSOH

MFI=4095

STM-n

SEQ=0...X

4096·n·270

375 µs 512 ms MFI=0

STM-n. . .

VC-3G1F2H4

J1B3C2

F3K3N1

SEQ=0MFI=0

SEQ=1

SEQ=3MFI=0.

..

9

1

VC3-4v

4·841

X -Segments

X times VC3

VC-3

SEQ=3MFI=1.

..

9

1

VC3-4v

87

4·841

X -Segments

X times VC3

G1F2

J1B3C2

SEQ=0

VC-3

MFI=4095

SEQ=1

SEQ=3MFI=4095.

..

9

1

VC3-4v

871

4·841

X -Segments

X times VC3

. . . . . .

. . . . . .

(VC-3-Xv)

t

. . .

.

. . .

.

. . .

.

. . .

.

. . .

.

. . .

.

t

SEQ=0...X

EOS EOS EOS

Con

tiguo

us p

aylo

ad

VCGSEQ=0MFI=1

SEQ=1F2H4

VC-3G1F2H4

J1B3C2

F3K3N1

VC-3G1F2H4

J1B3C2

F3K3N1

VC-3G1F2H4

J1B3C2

F3K3N1

VC-3G1F2H4

J1B3C2

F3K3N1


Virtual vs. Contiguous Concatenation

SDH SONET X times Capacity Justification Transport

VC-4 STS3c-SPE 1 149,760 Kbps 3 bytes STM-1/OC-3

VC-4-4c STS12c-SPE 4 599,040 Kbps 12 bytes STM-4/OC-12

VC-4-16c STS48c-SPE 16 2,396,160 Kbps 48 bytes STM-16/OC-48



SDH SONET X times Capacity Virtual Capacity

VC-11-Xv VT.15-Xv 1 to 64 1,600 Kbps 1,600 to 102,400 Kbps

VC-12-Xv VT2-Xv 1 to 64 2,176 Kbps 2,176 to 139,264 Kbps

VC-2-Xv VT6-Xv 1 to 64 6,784 Kbps 6,784 to 434,176 Kbps

VC-3-Xv STS-1-Xv 1 to 256 48,384 Kbps 48,384 to 12,386 Kbps

VC-4-Xv STS-3c-Xv 1 to 256 149,760 Kbps 149,760 to 38,338,560 Kbps


Virtual vs. Contiguous Concatenation

Service Bit Rate Contiguous Concat. Virtual Concatenation

Ethernet 10 Mbit/s VC-3 (20%) VC-11-7v (89%)

Fast Ethernet 100 Mbit/s VC-4 (67%) VC-3-2v (99%)

Gigabit Ethernet 1000 Mbit/s VC-4-16c (42%) VC-4-7v (95%)

Fiber Channel 1700 Mbit/s VC-4-16c (42%) VC-4-12v (90%)

ATM 25 Mbit/s VC-3 (50%) VC-11-16c (98%)

DVB 270 Mbit/s VC-4-4c (37%) VC-3-6v (93%)

ESCON 160 Mbit/s VC-4-4c (26%) VC-3-4v (83%)


K4 multiframe (VCAT & LCAS codification)

•••• K4 is part of the LO-PO overhead and is repeated every 500 ms

•••• 32 bits are sent in a complete multiframe which takes 16ms to repeat. (500 x 32=16 ms)

•••• The bit-2 superframe is made up of 32 multiframes and takes 512 ms to repeat.

K4 multiframe

MFI: Multiframe Count Indicator [0...31] SQ: Sequence Indicator int he VCG [0...

1 2 4 6 8 10 12 143 5 7 9 11 13 15 16frame

0 16ms

CTR

L

17 18 20 22 24 26 28 3019 21 23 25 27 29 31 32

RS-

Ack

V5Lower Order Path Overhead

125 µs250 µs

375 µs500 µs

(1bit per .5ms)K4 bit2

bit2

0000

100

010

0010

0

0011

0

0100

0

0101

0

0110

0

0111

0

0001

1

0010

1

0011

1

0100

1

0101

1

0110

1

0111

1

0000

0

1 2 4 6 8 10 12 143 5 7 9 11 13 15 16k4 superframe

MFI

512ms

1000

110

010

1010

0

1011

0

1100

0

1101

0

1110

0

1111

0

1001

1

1010

1

1011

1

1100

1

1101

1

1110

1

1111

1

1000

0

17 18 20 22 8 26 28 3019 21 23 25 27 29 31 32

0

SQCTRL...

CRC-3

J2 N2K4

MFI CTRLSQ MST CRC-30 0 0 0


H4 multiframe (VCAT & LCAS codification)

•••• H4 is part of the HO-PO overhead

•••• H4 is repeated every 125 ms

•••• 16-byte multiframes takes 16 ms

•••• A complete multiframe of 4096 bytes takes 512 ms to repeat (125x4096=512 ms)

0001

0010

0100

0110

1000

1010

1100

1110

0011

0101

0111

1001

1011

1101

1111

0000

0001

0010 4

1110

0011

1111

0000

1 2 256

0011

CTR

L

0 0

0 0

CR

C8

MST

0 0

0 RSA

ck

0 0

0 0

SQ

0 0

0 G

ID

0 0

0 0

0 0

0 0

0 0

0 0

MFI

1

CTR

L

MFI

2

0 0

0 G

ID

SQ

0 0

0 0

1 2 4 6 8 10 12 143 5 7 9 11 13 15 16 409617 18 19 20 21

5-8bits

1-4bits

frame

MFI1

0ms 2ms 512mstime

G1F2H4

J1B3C2

F3K3N1

multiframe multiframe

Hig

er O

rder

Pat

h O

verh

ead

MFI2: Multiframe indicator part 2 [0...255] MFI: Combination of MFI2-MFI1 [0...4095]MFI1: Multiframe indicator part 1 [0...255]

SQ: Sequence Indicator int he VCG [0...255]


Ethernet service provided by VCAT/LCAS

The link between node A and node Z transports Ethernet frames using a Virtual Concatenation Group of three members. Three separate LCAS protocols constantly monitor each peer connection: LCAS-a of node R talks with LCAS-a of node Z, LCAS-b(R) with LCAS-b(Z), ... LCAS-n (R) with LCAS-n (Z)

GFP

Ethernet

Ethernet frame

VCAT

Traffic

LCAS

S T

W

X

Payload Segregation

TrafficControlAdaption

MSSP - A

Ethernet

NG SDH

Legacy SDH MSSP - BAccess Access

Payload Aggregation

Flow Control

LCAS

VCAT member

Node RNode Z

Flow Control

Ethernet service

Y

R Z

CP1

CP2

ab

c-d

ab

c-d


Link Capacity Adjustment Scheme (LCAS)ITU-T as G.7042, designed to manage the bandwidth allocation of a VCAT path. LCAS can add and remove members of a VCG that control a VCAT channel. LCAS cannot adapt the size of the VCAT channel according to the traffic pattern.

Source to Sink messages:

•••• Multi-Frame Indicator (MFI) keeps the multiframe sequence.

•••• Sequence Indicator (SQ) indicates member’s sequence to reassemble correctly the client signal that was split and sent through several paths.

•••• Control (CTRL) protocol messages which can be fixed, add, norm, eos, idle, and dnu.

•••• Group Identification (GID) is a constant value for all members of a VCG.

Sink to Source include:

•••• Member Status (MST), which indicates to source each member status: fail or OK.

•••• Re-Sequence Acknowledge (RS-Ack) is an ack of renumbering after a new eos member.


LCAS protocol

fixed: Indicates fixed bandwidth and no LCAS supportadd: Member to be added to the VCGnorm: Normal transmissioneos: End of sequence, the member has the highest VCG seq number, Normal transmissionidle: Member is part of the VCG or to be removeddnu: Do Not Use, receive side reported MST FAIL status

ack: Re-Sequence Acknoledge after eos msg (RS-Ack msg)

ok: active member, no failure condition detected (MST msg)fail: failure condition detected in member (MST msg)

Source to Sink(CNTRL message)

Sink to Source(MST & RS messages)

Sinkfixed, add, norm, eos, idle, dnu

ok, fail, ackSource So Sk


Simplified LCAS Source and Sink State Machine

Start

MADD

MREMOVE

IDLE

ADD NORM

REMOVE

DNU

add

norm, eos, ack

failok

ack, fail idle

idle

ok

idle

MREMOVE

dnu

OK

IDLE

FAIL

eos, norm,

MADD

fail

ok, dnu, ack

Startfail

others

MREMOVE

idle, fixed, fail

add

fail

Source

IDLE: Not provisioned memberADD: In process of being added to the VCGNORM: Active and provisioned member, good pathDNU (Do Not Use): Provisioned but its path has failedREMOVE: In process of being removed of the VCG

Sink

IDLE: Not provisioned memberOK: Provisioned and active memberFAIL: Provisioned member, failed path

Source States Sink States

states in transition

dnu, idle

So Sk

MADD, MADD,MREMOVEMREMOVE

MADD: add one or more members of the VCGMREMOVE: remove one member of the VCG


VCAT Channel managed with LCAS

LCAS helps network operators to efficiently control NG SDH connections established at VCAT sites. The use of LCAS is not compulsory, but improves VCAT management

Sink

Source

LCAS

LCAS

VCGTx Rx

Source

TxRx

Sink

VCAT channel

LCAS

LCAS

Node A

fixed, add, norm, eos, idle, dnuok, fail, ack

IDLE IDLEADD FAIL

NORM OKDNU FAIL

REMOVE OK

abcd

hijk

fixed, add, norm, eos, idle, dnuok, fail, ack

MADDMREMOVE

Node B

NMS

MADD

MREMOVE NMS

MADDMREMOVE

NMS

MADDMREMOVE

NMS

NMS: Network Management System

SinkSource

Member States


LCAS protocol: MADD and Path error

a b c d a b c d

IDLE IDLEADD FAIL

NORM OKDNU FAIL

REMOVE OK

add (b)

ack

eos (b)norm (d)

fail (b)

ok

dnu (b)eos (d)

ack

ok (b)eos (b)

norm (d)

MADD (b)

ack

MADD (b)

SinkSource

LCAS sample

.

.

.path error in (b)

no error

Member States

Sink Member StatesSource Member States


LCAS protocol: MREMOVE

LCAS is a two-way handshake protocol resident in H4 and K4 and executed permanently between source and sink as many times as VCAT members.

IDLE IDLEADD FAIL

NORM OKDNU FAIL

REMOVE OK

SinkSource

LCAS sample

Member States

MREMOVE (d)

idle (d)fail (d)

idle (b)fail (b)

eos (a)

.

.

.

ack

MREMOVE (b)

a b c d a b c dSink Member StatesSource Member States


Sink messages are fault tolerant

Sink to Source messages (MST, RS-Ack) are redundant while Source to Sink are specific to each member. This means that Sink messages are repeated as many times as members in the group. It also means that the origin of sink messages is irrelevant, because all the members are sending the same information in a multiframe.

1a2a

3a

1b2b

3b

1a2a

3a

1b2b

3b

Source-A

Sink-B

Sink-A

Source-B

LCAS

Si-A to So-A messages

So-B to Si-B Ctrl. messages

Si-B to So-B messages

So-A to Si-A Ctrl. messages

Rx

Tx

VCG

Rx

Tx


H4 multiframe sequence

0255

160159 9596

128 127192191

22422331

32

40953072

0MFI 63

642048

1024

bytes

H4

16 bytes

multiframe (225)

2 ms

0 21 3member

8 9

4 65 7

255

.....

status (0-7)

021

3

64ms

256 -

MST MST

member MST status32 multi frames

8 9CTRL 0 CRC8 MST RSA 0 SQ

GID 0

00

MFI

1

8 9

16MFI

512ms

4096 bytes

256 multiframes

ok / fail

225

multiframe (31)

multiframe (0)

member status

ok / fail

ok / fail

ok / fail

ok / failok / failok / fail

ok / fail

MSTMST

MST MST

MSTMST

255.....

02 13

MSTMST

2 ms


LCAS applications

•••• VCAT bandwidth allocation, LCAS enables the resizing of the VCAT pipe in use when it receives an order from the NMS to increase or decrease the size.

•••• Network Resilience, In the case of a partial failure of one path, LCAS reconfigures the con-nection using the members still up and able to continue carrying traffic.

•••• Asymmetric Configurations, LCAS is a unidirectional protocol allowing the provision of asymmetric bandwidth between two MSSP nodes to configure asymmetric links

•••• Cross-Domain Operation, because LCAS resides only at edge nodes it is not necessary to coordinate more than one configuration centre

Path1 = 25%

Path3 = 50%

A B

FE

X YG

A B

FE

X YG

Path1= 35%

Path3 = 65%

Normal operation After breakdown

VCATVCAT VCAT VCAT

LCASLCASMSSPMSSP

Path225%

Path20%

Section

Conclusions


SDH is future proof

•••• SDH cannot be considered anymore as a “legacy technology”

•••• The evolution to Next Generation SDH is the future

•••• Future proof at least for the next 10 years

•••• NG SDH can deliver packet and TDM services

•••• Carriers are already migration from static, circuit-oriented SDH tomultiservice packet-friendly NG SDH


Migration to Next Generation

It is only necessary to migrate the edges to get a full Next Generation SDH network

IT IS TODAY’S BEST COMBINATION FOR DATA AND CIRCUIT TRANSPORT

SDHLCASVCAT Virtual Containers Transport

Bandwidth management

Paths, Sections

GFC Mapping in Frames

Laye

rs

SDH SDH SDHLCASVCATGFC

Paths, Sections

EthernetVPNDVBSAN

PDHEthernetVPNDVBSAN

PDHMSSP MSSP

Existing SDH/Sonet

Next Generation SDH

Clients Clients

SDH NGSDH classic


NG SDH means unification and standardisation

Next Generation SDH unifies and standardises transport infrastructure for any type of client network packet or circuit oriented network such as Ethernet, PDH, Frame Relay, UMTS, SAN...

ATM

Telephone

UMTS

SDH

SDH

SDHnode

Frame Relay

STM-NGFrame Relay

MSSP


Old and New services as well

•••• NG SDH helps to provide the right SLA to Ethernet services

•••• Transport for high definition audio and video

•••• High speed data for Internet or other networks

•••• Fast bandwidth management to satisfy requirements

•••• Integration under the same architecture circuit and packet networks

ServicesTelephony

Frame RelayUMTS

GSM

InfrastructuresSDH DWDM

Ethernet

ATM access

FrameRelay

IWU

Host

Superserver

SAN

Wireless

VoIP


Cost effective

•••• Universal standard: multivendor

•••• Unifies infrastructures under a central architecture avoiding a new overlap network

•••• Reduces the number of network elements needed to provide advanced services

•••• New technologies are making NG SDH more competitive and cheaper

•••• With just a few network elements is possible to configure a network

•••• Simplifies the management because of centralised configurations

Next Gen SDH network

SDH accessSDH ring2,5 Gbit/s

Ethernet access

Data storagePDH access


Telecom basis in the next 10 years?

•••• “SDH will be the dominant technology in the next 10 years” Pioneers Optical Edge Net-works Boston (MA) June 2000

•••• “SDH is being used in wireless, we think there is a market at least for the next 10 years” Eli Lotan Global Telephony March 01

•••• “Ethernet demand lights up but SDH will remain at the core network” Ken Weiland Tele-communications International April 03

•••• “Ethernet and Optic networks have to match SDH resilience and management to chal-lenge SDH at the metro and core network” JM Caballero Mundo Electr. April 04

•••• “Installed base is huge to simply walk away from” S. Clavena Heaving Reading Nov. 03

•••• “Ethernet, of course, but over SDH” Network Planning and Operations US


Ethernet challenge or SDH complement?

•••• Just a few carriers and operators around the world are setting up Core Networks removing SDH and relying only on Ethernet and/or DWDM

•••• It has always been difficult to justify in economic, technical and management terms a com-pletely new separate network. And that is still the case.

•••• Ethernet reduces leased access costs while NG SDH enhances QoS

•••• Native Ethernet services, and best effort technologies in general, do not scale well

•••• SDH resilience is critical to deliver 24/365 Ethernet services

•••• Ethernet is by nature a best effort technology, carrier-class is already under way

•••• Ethernet does not match today the OEM functions provided by SDH

•••• A pure packet network, i.e. Ethernet + MPLS, has to emulate circuits: it’s a risky solution!


All-optical challenge or SDH complement?

•••• DWDM alone does not have the diversity or management functions of NG SDH

•••• The new MSTP nodes combines reconfigurable OADM with NG SDH features

•••• Provides new interfaces (packet + circuit oriented) while offering wider managing services

•••• Automate wavelength provisioning and lower wavelength delivery costs

•••• Ability to reduce network elements in the circuit path

Conclusion: Optical layer integration is a must for NG SDH using MSPP that will provide higher flexibility and Lower Capital Expense (Capex) operational expense (Opex)


New data services

CPE Access Metro Core

Physical

Transport

Service

ClientGigE, Circuit, ATM

FRL, SAN, DVB

MSPP

MSTP

MSSP

ROADM

ADM

CXC


NG SDH is S·T·R·A·T·E·G·I·C•••• Is being adopted not only by incumbent operators and carriers but also by new ones.

•••• It is interesting to consider the adoption of RPR and/or MPLS to enhance the feature of the Ethernet + NG SDH tandem to layer 2 protection, QoS and multipoint access.

•••• Video transport and Storage over NG SDH is just starting to impact in today’s networking

•••• Automatic layer 1 reconfiguration complements Layer 2 RPR reconfiguration

•••• High reliability equivalent similar in core and metro

•••• Centralised management of events, bandwidth provisioning

•••• Next Generation network (such as MSPP, MSSP) elements will be as fundamental to tele-com networks in the coming decade as routers were to the Internet of the 90s

ng.sdh.slides.e

Documents

au3 ptrs msoh

multiservice

generic framing

failure condition

multiservice

multiservice

sequence indicator

virtual container