All rights reserved. Printed in the USA. November 1998.
The information in this document is subject to change without notice. The statements, configurations, technical data,
and recommendations in this document are believed to be accurate and reliable, but are presented without express or
implied warranty. Users must t ake full responsibility for th eir a pplic atio ns o f a ny products specif i ed in th is d ocume nt .
The information in this document is proprietary to Bay Networks, Inc.
The software described in this document is furnished under a license agreement and may only be used in accordance
with the terms of that license. A summary of the Software License is included in this document.
Trademarks
BCN, BLN, BN, FRE, and Bay Networks are registered trademarks and BayRS, BCC, Connection Management
System, System 5000, and the Bay Networks logo are trademarks of Bay Networks, Inc.
All other trademarks and registered trademarks are t he property of their respective owners.
Restricted Rights Legend
Use, duplication, or disclosure by the United States Government is subject to restrictions as set forth in subparagraph
(c)(1)(ii) of the Rights in Technical Data and Computer So ftware clause at DFARS 252.227-7013.
Notwithstanding any other license agreement that may pertain to, or accompany the delivery of, this computer
software, the rights of the United States Government regarding its use, reproduction, and disclosure are as set forth in
the Commercial Computer Software-Restricted Rights cl ause at FAR 52.227-19.
Statement of Conditions
In the interest of improving internal design, operational function, and/or reliability, Bay Networks, Inc. reserves the
right to make changes to the pr oducts described in this document without notice.
Bay Networks, Inc. does not assume any liability that may occur due to the use or application of the product(s) or
circuit layout(s) described herein.
SUCH PORTIONS OF THE SOFTWARE ARE PROVIDED “AS IS” AND WITHOUT ANY EXPRESS OR
IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.
In addition, the program and information contained herein are licensed only pursuant to a license agreement that
contains restrictions on use and disclosure (that may incorporate by reference certain limitations and notices imposed
by third parties).
ii
117374-C Rev 00
Bay Networks, Inc. Software License Agreement
NOTICE: Please carefully read this license agre ement before copying or using the accompanying software or
installing the hardware unit with pre-enabled software (each of which is referred to as “Software” in this Agreement).
BY COPYING OR USING THE SOFTWARE, YOU ACCEPT ALL OF THE TERMS AND CONDITIONS OF
THIS LICENSE AGREEMENT. THE TERMS EXPRESSED IN THIS AGREEMENT ARE THE ONLY TERMS
UNDER WHICH BAY NETWORKS WILL PERMIT YOU TO USE THE SOFTWARE. If you do not accept these
terms and conditions, retu rn the product, unused and in the original shipping container, within 30 days of purchase to
obtain a credit for the full purchase price.
1. License Grant. Bay Networks, Inc. (“Bay Networks”) grants the end user of the Software (“Licensee”) a personal,
nonexclusive, nontransferable license: a) to use the Software either on a single computer or, if applicable, on a single
authorized device identified by host ID, for which it was originally acquired; b) to copy the Software solely for backup
purposes in support of authorized use of the Software; and c) to use and copy the associated user manual solely i n
support of authorized use of the Software by Licensee. This license applies to the Software only and does not extend
to Bay Networks Agent software or other Bay Networks software products. Bay Networks Agent software or other
Bay Networks software products are licensed for use under the terms of the applicable Bay Networks, Inc. Software
License Agreement that accomp anies such software and upon payment by the end user of the applicable license fees
for such software.
2. Restrictions on use; reservation of rights. The Software and user manuals are protected under copyright laws.
Bay Networks and/or it s licensors retain all title and ownership in both the Software and user manuals, including any
revisions made by Bay Networks or its licensors. The copyright notice must be reproduced and included with any
copy of any portion of the Software or user manuals. Licensee may not modify, translate, decompile, disassemble, use
for any competitiv e analysis, re v erse engineer , distrib ute, or create deriv ati ve works from the Softwa re or user manuals
or any copy, in whole or in part. Except as expressly provided i n t hi s Agreement, Licensee may not copy or transfer
the Software or user manuals, in whole or in part. The Soft ware and user manuals embody Bay Networks’ and its
licensors’ confidential and proprietary intellectual property. Licensee shall not sublicense, assign, or otherwise
disclose to any third party the Software, or any information about the operation, design, performance, or
implementation of the Software and user manuals that is confidential to Bay Networks and its licensors; however,
Licensee may grant permission to its consultants, subcontractors, a nd agents to use the Softw are at Licensee’s facility,
provided they have agreed to use the Software only in accordance with the terms of this license.
3. Limited warranty. Bay Netw o r ks wa r ra nts ea c h item of So ft ware, as delivered by Bay Net w o rk s an d pr o per ly
installed and operated on Bay Networks hardware or other equipment it is originally licensed for, to function
substantially as described in its accompanying user m anual during its warranty period , which begins on the date
Software is first shipped to Licensee. If an y item of S oftware f ails to so function d uring its w arranty period, as the sole
remedy Bay Networks will at its discretion provide a suitable fix, patch, or workaround for the problem that may be
included in a future Software release. Bay Network s fur ther w arra nts to Licen see that the medi a on which the
Software is provided will be free from defec ts in materials and wo rkman ship under no rmal use for a peri od of 90 da ys
from the date Software is first shipped to Licensee. Bay Networks will replace defective media at no charge if it is
returned to Bay Netw orks during the warran ty perio d alon g with proof of the date of shipment . This war ranty do es not
apply if the media has been dam aged as a resul t of acci dent, misuse , or ab use. The Licen see assumes all re sponsibilit y
for selection of the Software to achieve Licensee’s intended results and for the installation, use, and results obtained
from the Software. Bay Networks does not warrant a) that the functions contained in the software will meet the
Licensee’ s requireme nts, b) that the Software will operate in the hardware or software combinations that th e Licen see
may select, c) that the operation of the Softw a re will be uninterru pte d or error free, or d) that all defec ts in the
operation of the Software will be corrected. Bay Networks is not obligated to remedy any Software defect that cannot
be reproduced with the latest Software release. These warranties do not apply to the So ftw are if i t has been (i) altered,
except by Bay Networks or in accordance with its instructions; (ii) used in conjunction with another vendor’s product,
resulting in the defect; or (iii) damaged by improper environment, abuse, misuse, accident, or negligence. THE
FOREGOING WARRANTIES AND LIMITATIONS ARE EXCLUSIVE REMEDIES AND ARE IN LIEU OF ALL
OTHER WARRANTIES EXPRESS OR IMPLIED, INCLUDING WITHOUT L IMITATION ANY WARRANTY OF
MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Licensee is responsible for the security of
117374-C Rev 00
iii
its own data and information and for maintaining adequate procedures apart from the Software to reconstruct lost or
altered files, data, or programs.
4. Limitation of liability. IN NO EVENT WILL BAY NETWORKS OR ITS LICENSORS BE LIABLE FOR ANY
COST OF SUBSTITUTE PROCUREMENT; SPECIAL, INDIRECT, INCIDENTAL, OR CONSEQUENTIAL
DAMAGES; OR ANY DAMAGES RESULTING FROM INACCURATE OR LOST DATA OR LOSS OF USE OR
PROFITS ARISING OUT OF OR IN CONNECTION WITH THE PERFORMANCE OF THE SOFTWARE, EVEN
IF BAY NETWORKS HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. IN NO EVENT
SHALL THE LIABILITY OF BAY NETWORKS RELATING TO THE SOFTWARE OR THIS AGREEMENT
EXCEED THE PRICE PAID TO BAY NETWORKS FOR THE SOFTWARE LICENSE.
5. Government Licensees. This provision applies to a ll Softwa re and docum entation acquired d irectly or i ndirectly by
or on behalf of the United States Government. The Software and documentation are commercial products, licensed on
the open market at market prices, and were developed entirely at private expense and without th e use of any U.S.
Government funds. The license to the U.S. Government is granted only with restricted rights, and use, duplication, or
disclosure by the U.S. Government is subject to the restrictions set forth in subparagraph (c)(1) of the Commercial
Computer Software––Restricte d Rig hts cla u se o f FAR 52.227-19 and the lim itatio ns se t o ut in this license for civilian
agencies, and subparagraph (c)(1)(ii) of the Rights in Technical Data and Computer Software clause of DFARS
252.227-7013, for agencies of t he Department of Defense or their successors, whichever is applicable.
6. Use of Software in the European Community. This provision applies to all Software acquired for use within the
European Community. If Licensee uses the Software within a country in the European Community, the Software
Directive enacted by the Council of European Communities Directive dated 14 May, 1991, will apply to the
examination of the Software to facilitate interoperability. Licensee agrees to notify Bay Networks of any such
intended examination of the Software and may procure support and assistance from Bay Networks.
7. Term and termination. This license is effecti ve until terminated; however, all of the restrictions with respect to
Bay Networks’ copyright in the Software and user manuals will cease being effective at the date of expiration of the
Bay Networks copyright; those restrictions relating to use and disclosure of Bay Networks’ confidential information
shall continue in effect. Licensee may terminate this license at any time. The license will automatically terminate if
Licensee fails to comply with any of the terms and conditions of the license. Upon termination for any reason,
Licensee will immediately destroy or return to Bay Networks the Software, user manuals, and all copies. Bay
Networks is not liable to Licensee for damages in any form solely by reason of the termination of this license.
8. Export and Re-export. Licensee agrees not to export, directly or indirectly, the Software or related technical data
or information without first obtaining any required export licenses or other governmental approvals. Without limiting
the foregoing, Licensee, on behalf of itself and its subsidiaries and affiliates, agrees that it will not, without first
obtaining all export licenses and approvals required by the U.S. Government: (i) export, re-export, transfer, or divert
any such Software or technical data, or any direct product thereof, to any country to which such exports or re-exports
are restricte d or em b argoed under United States export control laws and regulatio ns, or to any national or resident of
such restricted or embargoed countries; or (ii) provide the Software or related technical data or information to any
military end user or for any military end use, including the design, development, or production of any ch emical,
nuclear, or biological weapons.
9. General. If any provision of this Agreement is held to be invalid or unenf orceable by a court of competent
jurisdiction, the remainder of the provisions of this Agreement shall remain in full force and effect. This Agreement
will be governed by the laws of the state of California.
Should you have any questions concerning this Agreement, contact Bay Networks, Inc., 4401 Great America Parkway,
P.O. Box 58185, Santa Clara, California 95054-8 185.
LICENSEE ACKNOWLEDGES THAT LICENSEE HAS READ THIS AGREEMENT, UNDERSTANDS IT, AND
AGREES TO BE BOUND BY ITS TERMS AND CONDITIONS. LICENSEE FURTHER AGREES THAT THIS
AGREEMENT IS THE ENTIRE AND EXCLUSIVE AGREEMENT BETWEEN BAY NETWORKS AND
LICENSEE, WHICH SUPERSEDES ALL PRIOR ORAL AND WRITTEN AGREEMENTS AND
COMMUNICATIONS BETWEEN THE PARTIES PERTAINING TO THE SUBJECT MATTER OF THIS
AGREEMENT. NO DIFFERENT OR ADDITIONAL TERMS WILL BE ENFORCEABLE AGAINST BAY
NETWORKS UNLESS BAY NETWORKS GIVES ITS EXPRESS WRITTEN CONSENT, INCLUDING AN
EXPRESS WAIVER OF THE TERMS OF THIS AGREEMENT.
iv
117374-C Rev 00
Contents
Preface
Before You Begin ...........................................................................................................xxiii
Text Conventions ...........................................................................................................xxiv
This guide describ es as ynchro nous transf er mode (ATM) and Mult ipro tocol Lab el
Switching (MPLS) and what you do to star t and cus tomize t hese serv ices on a Bay
®
Networks
You can use the Bay Command Console (BCC
ATM on a router. BCC supports some ATM features. However, some features are
not supported and must be configured using Site Manager. In this guide, you will
find instructions for using both the BCC and Site Manager.
To configure MPLS, you must use Site Manager.
router.
™
) or Site Manager to configure
Before You Begin
Before using this guide, you must complete the following procedures. For a new
router:
•Install the router (see the installation guide that came with your router).
•Connect the router to the network and create a pilot configuration file (see
Quick-Starting Routers).
Make sure that you are running the latest version of Bay Networks BayRS
Site Manager software. For information about upgrading BayRS and Site
Manager, see the upgrading guide for your version of BayRS.
117374-C Rev 00
™
and
xxiii
Configuring ATM and MPLS Services
Text Conventions
This guide uses the following text conventions:
angle brackets (< >)Indicate that you choose the text to enter based on the
description inside the brackets. Do not type the
brackets when entering the command.
Example: If the command syntax is:
ping
<ip_address>
ping 192.32.10.12
, you enter:
bold text
Indicates text that you need to enter and command
names and options.
Example: Enter
show ip {alerts | routes
Example: Use the
dinfo
command.
}
braces ({})Indicate required elements in syntax descriptions
where there is more than one option. You must choose
only one of the options. Do not type the braces when
entering the command.
Example: If the command syntax is:
show ip {alerts | routes
show ip alerts or show ip routes
, you must enter either:
}
.
brackets ([ ])Indicate optional elements in syntax descriptions. Do
not type the brackets when entering the command.
Example: If the command syntax is:
show ip interfaces [-alerts
show ip interfaces
or
, you can enter either:
]
show ip interfaces -alerts
.
xxiv
117374-C Rev 00
Preface
ellipsis points (. . . )Indicate that you repeat the last element of the
command as needed.
Example: If the command syntax is:
ethernet/2/1 [<
ethernet/2/1 and as many parameter-value pairs as
parameter> <value>
] . . ., you enter
needed.
italic textIndicates file and directory names, new terms, book
titles, and variables in command syntax descriptions.
Where a variable is two or more words, the words are
connected by an underscore.
Example: If the command syntax is:
show at <
valid_route
valid_route>
is one variable and you substitute one value
for it.
screen textIndicates system output, for example, prompts and
system messages.
Example:
Set Bay Networks Trap Monitor Filters
separator ( > )Shows menu paths.
Example: Protocols > I P ide nti fies the IP option on the
Protocols menu.
vertical line (
)Separates choices for command keywords and
|
arguments. Enter only one of the choices. Do not type
the vertical line when entering the command.
Example: If the command syntax is:
117374-C Rev 00
show ip {alerts | routes}, you enter either:
show ip alerts or show ip routes, but not both.
xxv
Configuring ATM and MPLS Services
Acronyms
AALATM adaptation layer
ABRavailable bit rate
AFIauthority and format identifier
AISalarm indication signal
ALCadaptation layer controller
AREATM Routing Engine
ARPAddress Resolution Protocol
ATMasynchronous transfer mode
BFEBlacker front-end encryption
B-ISDNBroadband Integrated Services Digital Network
BUSbroadcast and unknown server
CLPcell loss p riority
CPCScommon part convergence sublayer
xxvi
CSconvergence sublayer
CSUchannel service unit
DCEdata communication equipment
DDNDefense Data Network
DSUdata service unit
DTEdata terminal equipment
ELANemulated local area network
ERerror recovery
FIBforwarding information base
HECheader error control
IETFInternet Engineering Task Force
ILIIntelligent Link Interface
ILMIInterim L ocal Management Interface
IPInternet Protocol
117374-C Rev 00
Preface
IPXInternetwork Packet Exchange
ITU-TInternational Telecommunication Union - Telecommunication
Standardization Sector
LANElocal area network emulation
LDPlabel distribution protocol
LELAN emulation
LECLAN emulation client
LECSLAN emulation configuration server
LERlabel edge router
LESLAN emulat ion server
LISlogical IP subnet
LLCLogical Link Control
LUNILAN emulation UNI
MACmedia access control
MBSmaximum burst size
MCRminimum cell rate
MCSmulticast server
MIBmanagement inform ation base
MPCMulti-Protocol over ATM client
MPLSMultiprotocol Label Switching
MPOAMulti-Protocol over ATM
MPSMPOA server
MTUmaximum transmission unit
NHRPNext Hop Resolution Prot ocol
NMLNative Mode LAN
NMSnetwork management station
NNInetwork-to-network inte rface
OAMOperations and Management
OAM&POperations, Administration, Maintenance and Provisioning
117374-C Rev 00
xxvii
Configuring ATM and MPLS Services
OC-3Optical Carrier-level 3
OSIOpen Systems Interconnection
OSPFOpen Shortest Path First
PCRpeak cell rate
PDpoll data
PDNPublic Data Network
PDUprotocol data unit
PHYphysical [layer]
PMDphysical medium dependent
PTpayload type
PVCpermanent virtual circuit
RDIremote defect indication
RIPRouting Information Protocol
RSresynchronization
SAALsignaling AAL
xxviii
SAPservice access point
SARsegmentation and reassembly
SCRsustainable cell rate
SDsequenced data
SDUservice data unit
SMDSSwit ched Multimegabit Data Service
SNAPSubnetwork Access Protocol
SNMPSimple Network Management Protocol
SONET/SDHSynchronous Optical Network/Synchronous Digital Hierarchy
SPEsynchronous payload envelope
SRMSystem Resource Module
SSCOPService Specific Connection Oriented Protocol
SSCSservice specific converge nce sublayer
STPshielded twisted pair
You can now print Bay Networks technical manuals and release notes free,
directly from the Internet. Go to support.baynetwork s.com/libr ary/ tpubs/ . Fi nd the
Bay Networks product for which you need documentation. Then locate the
specific category and model or version for your hardware or software product.
Using Adobe Acrobat Re ader, you can open the manuals and r eleas e note s, sea rch
for the sections you need, and print them on most standard printers. You can
download Acrobat Reader free from the Adobe Systems Web site,
www.adobe.com.
You can purchase Bay Networks documentation sets, CDs, and selected technical
publications through the Bay Networks Collateral Catalog. The catalog is located
on the World Wide Web at support.baynetworks.com/catalog.html and is divided
into sections arranged alphabetically:
•The “CD ROMs” section lists available CDs.
•The “Guides/Books” section lists books on technical topics.
•The “Technical Manuals” section lists available printed documentation sets.
xxix
Configuring ATM and MPLS Services
Make a note of the part numbers and prices of the items that you want to order.
Use the “Marketing Collateral Catalog description” link to place an order and to
print the order form.
How to Get Help
For product assi stance, support contracts, information about educational services,
and the telephone numbers of our gl obal supp ort offices, go to the following URL :
http://www.baynetworks.com/corpor ate/ conta cts/
In the United States and Canada, you can dial 800-2LANWAN for assistance.
xxx
117374-C Rev 00
Chapter 1
Understanding ATM, MPO A,
ATM Router Redundancy, and OAM
This chapter describes the concepts underlying ATM and, where appropriate, the
specific ways Bay Networks implements these concepts on its routers. It contains
the following information:
TopicPage
ATM General Information1-2
Classical IP over ATM Concepts1-28
ATM LAN Emulation Concepts1-33
Multi-Protocol ov er ATM Concepts1-39
ATM Router Redundancy Concepts1-43
PVC Operations and Management Concepts1-44
For More Information1-46
Where to Go Next1-47
117374-C Rev 00
1-1
Configuring ATM and MPLS Services
ATM General Information
Asynchronous transfer mode (ATM) is a connection-oriented, cell-based
technology that relays traffic across a Broadband Integrated Services Digital
Network (B-ISDN). ATM provides a cost-effective way of transmitting voice,
video, and data across a network.
ATM Cells
An ATM cell is a fixed-length packet of 53 bytes. It consists of a 5-byte header
containing address information and a fixed, 48-byte information field. Figure 1-1
shows a diagram of an ATM cell.
5-byte
header
Figure 1-1.ATM Cell
48-byte
information field
ATM0001A
This fixed-length cell size allows you to predict network delays, making ATM
suitable for carrying real-time information (for example, voice and video) as well
as data.
ATM allows the network to operate at a much higher rate than typical
packet-switching protocols (for example, X.25), because it provides no error
protection or flow control. Instead, ATM relies on the source and destination
devices to perform er ror-recovery functions such as retra nsmissi on of lost pack et s.
1-2
117374-C Rev 00
Understanding ATM, MPOA, ATM Router Redundancy, and OAM
Cell Header
After dividing the data into 48-byte segments for transmission, the end device --
that is, the ATM da ta ser vice un it/c hann el ser vice un it (DSU/CSU) or native ATM
device -- attaches the required header information (Figure 1-2
The fields in each ATM cell header provide all the information necessary for
networking. These fields include the following:
•Generic flow control (GFC): The first 4 bits of the cell header contain the
GFC. The GFC controls traffic flow onto the ATM network by contr olli ng the
user-to-network interface (U NI).
Cell loss
priority
1
2
3
4
5
Bytes
ATM0002A
117374-C Rev 00
•V irt ual path id entifier (VPI): The next 8 bits of the cel l heade r (tha t is, the las t
half of byte 1 and the first half of byte 2) contain the VPI. The VPI specifies a
virtual path on the physical ATM link. See the next section, “Data
Transmission,” for additional information about virtual paths.
1-3
Configuring ATM and MPLS Services
•Virtual channel identifier (VCI): The next 16 bits of the cell header (that is,
the last half of b yte 2, b yte 3, a nd the first half of byte 4) contain the VCI. Th e
VCI specifies a virtual channel within the virtual path on the physical ATM
link. See the next section, “Data Transmission,” for additional information
about virtual channels.
•Payload type (PT): The next 3 bits (that is, b it s 5 through 7 of byte 4) indicat e
the type of information the cell is carrying (for example, user data or
management information).
•Cell loss priority (CLP): The last bit of byte 4 indicates the priority of the cell
and whether the network can discard the cell under heavy traffic conditions.
Setting the bit to 1 indicates the network may discard the cell if necessary.
•Header error control (HEC): The last byte of the header field contains the
HEC. Its primary function is to guard against misdelivery of cells due to
header or single-bit errors. However, the HEC does not gauge the quality of
the data in the information field.
Cell Information Field
Following the 5-byte cell header is a 48-byte information field containing user
data. The ATM adapta ti on l ayer (AAL) organizes the data in thi s field. See “ATM
Layers” on page 1-6 for additional information about the AAL.
Data Transmission
Data transmission (also called cell switching) through the ATM network relies on
the establishment of logical connections between ATM devices. ATM is a
connection-oriented service. This means that an ATM device cannot transmit
information until it establishes a connection with a receiving device. These
connections consist of virtual channels, virtual paths, and transmission paths.
A virtual channel is a logical connection between two communicating ATM
devices. Each virtual channel can carry a different protocol or traffic type. The
virtual channel tra nsport s cel ls tha t ha v e a c ommon ide ntifier, the VCI, that is part
of the cell header. You can establish virtual channels permanently or set them up
dynamically, allowing the network to adjust itself to the traffic demand.
1-4
117374-C Rev 00
Understanding ATM, MPOA, ATM Router Redundancy, and OAM
A virtual path is a set of virtual channels between a common source and
destination. The virtual channels in a virt ual path are logically associated with a
common identifi er, the virtual path identifier (VPI), that is part of the cell header.
You can base cell switching on either the VPI alone, or on a combination of the
VPI and VCI.
V irt ual paths en able you t o separa te networ k trans por t funct ions in to thos e relat ed
to an individual logical connection (virtual channel) and those related to a group
of logical connections (virtual path).
A transmission path is a physical connection that comprises several virtual paths,
each virtual path containing several virtual channels. The transmission path can
support multiple virtual paths across a single connection to the network.
Figure 1-3
shows the relationships betw een the vir tual channel, the virtual path,
and the transmission path.
Transmission Path
VC
VC
VC
Figure 1-3.ATM Transmission Components
VP
VP
VP
VP = Virtual path
VC = Virtual channel
VP
VP
VP
VC
VC
VC
ATM0006A
117374-C Rev 00
1-5
Configuring ATM and MPLS Services
Permanent and Switched Virtual Connections
Virtual channels and virtual paths allow you to establish virtual channel links
(VCLs). You can create VCLs as ei ther permanent virtual circuits (PVCs) or
switched virtual circuits (SVCs). After you establish a PVC, you can transfer
information over it at any time. SVCs activate, through signaling and network
switching, only when there is information ready for transmission.
ATM Layers
The B-ISDN protocol reference model, on which ATM is based, consists of four
layers (Figure 1-4
and the layer directly below it.
). Each layer communicates only wit h the l ayer di rect ly abo v e i t
Higher protocol layer
Convergence sublayer (CS)
Segmentation and reassembly sublayer (SAR)
ATM layer
Transmission convergence sublayer
Physical medium dependent sublayer (PMD)
Figure 1-4.B-ISDN ATM Protocol Reference Model
ATM
adaptation
layer
(AAL)
Physical
layer
(PHY)
ATM0003A
1-6
117374-C Rev 00
Understanding ATM, MPOA, ATM Router Redundancy, and OAM
The following layers relate directly to how Bay Networks routers support ATM:
•Physical layer
•ATM layer
•ATM ada ptation layer (AAL)
Physical Layer
®
In a Bay Networks Backbone Node (BN
) router, Intelligent Link Interface (ILI)
pairs provide access and processing. An ILI pair consists of a link module and a
processor module that work together to process and transmit information over a
network. Bay Networks provides an ATM Routing Engine (ARE) link module in
conjunction with an ARE processor.
In addition, Bay Networks provides a hub version of the ATM ARE ILI pair, the
Model 5782 ATM router. This single hub module, incorporating the ILI functions
of both the link module and processor module, resides in a System 5000BH
chassis. See Using the Model 5782 ATM Virtual Network Router for more
information.
The ILI pair and the Model 5782 ATM router are functionally identical, and you
configure them in much the same way. Unless otherwise noted, the remainder of
this guide treats the router ILI pair and the Model 5782 as if they were the same.
Depending on the ATM router type, you can connect directly to an ATM network
over any of the following phys ical interfaces:
•Optical carrier level 3 (OC-3)
•Digital s ervice level 3 (DS-3)
•E-3 (the European equivalent of the North American DS-3)
117374-C Rev 00
1-7
Configuring ATM and MPLS Services
Although Bay Networks uses raw bandwidth to describe line rates, inherent
overhea d wit hin the me di a li mit s maxi mum band width for the line. Use Table 1-1
to determine the maximum bandwidth for the media you are using.
The ATM layer d efines how two nodes transmit in formation between them. It is
concerned with the format and size of the cells and the contents of the headers.
The addresses of the cells are meaningful only to the two adjacent local nodes
(that is, usually not to the end nodes).
ATM Adaptation Layer
The ATM adaptation layer (AAL) converts upper-layer protocol data into formats
that are compatible with the requirements of the ATM layer, enabling ATM to
handle different t ypes of infor mation within the same format.
The AAL is divided into two sublayers: the convergence sublayer (CS) and the segmentation and reassembly (SAR) sublayer. These two sublayers convert
variable-length messages into 48-byte segments, while ensuring the integrity of
the data.
1-8
The CCITT (now ITU-T) has defined different types of AALs to handle different
kinds of traffic. Bay Networks ATM routers support a CS function compatible
with AAL 5, as defi ned in Section 1 of the CCITT (now ITU-T) B-ISDN Protocol Reference Model (PRM).
In AAL 5, only a trailer attaches to the data from the upper-layer protocols to
create a CS PDU. AAL 5 divides the protocol data uni t (PDU) into a 48-octet SAR
PDU; howe v er , it does not add an SAR header and trailer. This 48-octet SAR PDU
becomes the payload of the ATM cell (Figure 1-5
).
117374-C Rev 00
Understanding ATM, MPOA, ATM Router Redundancy, and OAM
Service
class
AAL 5
convergence
sublayer
AAL
layer
AAL 5
SAR
layer
ATM
layer
Note: AAL 5 does not attach an SAR header or trailer to the PDU.
Figure 1-5.ATM Adaptation Layer 5
Service Records and Virtual Circuits
SAR PDU
ATM cell
Service data unit (SDU)
SDU
SAR PDU
ATM cell
Trailer
SAR PDU
ATM cell
ATM0005A
117374-C Rev 00
ATM devices communicate using virtual circuits (VCs). These VCs transmit and
receive ATM cells containing upper-layer protocols. Bay Networks ATM routers
use service records to provide a way of creating logical interfaces (within the
physical ATM interface) for these upper-layer protocols. In essence, these service
records allow you to:
•Organize multiple VCs into logical groups.
•Create direct point-to-point connections.
•Assign, delete, and modify upper-layer protocols for just one PVC or for a
group of VCs at any given time.
1-9
Configuring ATM and MPLS Services
Supported Protocols
Depending on the data encapsulation type and virtual connection type (PVC or
SVC) you choose for the service record, the router supports various protocols.
Table 1-2
lists all supported protocols for standa rd PVCs and SVCs using
LLC/SNAP, NLPID, NULL, LANE 802.3, or LANE 802.5 data encapsulation.
Table 1-2.Protocols Supported for Standard PVCs and SVCs
PVC Using LLC/SNAP,
NLPID, or NULL
BridgeIPBridgeBridge
Spanning TreeRIPSpanning TreeSpanning Tree
Native Mode LANBGPNative Mode LAN
IP
RIP
EGPBGPBootP
BGPOSPF
OSPFBootPRIP/SAP
BootPRouter Discovery
IGMPIGMPSR Spanning Tree
SVC Using LLC/SNAP
or NULL (RFC 1577)SVC Using LANE 802.3SVC Using LANE 802.5
IP
OSPF
IPv6
IP
RIPOSPF
RIP (XNS)
AppleTalk
LLC2
RIP
IPX
Source Routing
LLC2
1-10
117374-C Rev 00
Understanding ATM, MPOA, ATM Router Redundancy, and OAM
Caution:
Ethernet and token ring emulated LANs can support different
protocols. When adding a protocol to a LANE service record with an
unspecified emulated LAN type, ensure that the protocols you add are
supported by the emulated LAN (Ethernet or token ring) that you want to join.
Things to Remember
When enabling protocols on a service record, keep the following in mind:
•A PVC service record requires that you add at least one PVC for the service
record to operate.
•Each ATM service record globally controls:
--All protocols for any standard PVCs and SVCs that it contains
--All nonbridging protocols for any hybrid PVCs that it contains
•Selecting LANE to run on an SVC service record defines that service record
as belonging to an emulated LAN. Any protocols on that service record
operate as if they were running over a traditional Ethernet or token ring LAN.
•When running IP over a NULL encapsulated PVC service record, you must
change the A ddress Resolution parameter to None. You must then add an IP
adjacent host with the MAC address equal to the VPI/VCI of the PVC. See
Configuring IP Services for information about the Address Resolution
parameter.
117374-C Rev 00
1-11
Configuring ATM and MPLS Services
Rules for Editing Protocols
Depending on the type of virtual circuit you are using, Site Manager requires you
to add additional protocols, or delete and edit existing protocols, from specific
protocol menus.
Use Table 1-3
Table 1-3.Locating and Using Site Manager Protocol Menus
Site Manager Menu LocationPVCs and SVCsHybrid PVCs Only
ATM Service Records List window
ATM Virtual Channel Link window
* For nonbridging protocols
†
For bridging protocols
to locate the appropriate protocol menu for each access mode.
Remember, hybrid PVCs use their service record configurations for nonbridging
protocols and their individual configurations for bridging protocols.
Data Encapsulation Methods
Bay Networks ATM routers support multiprotocol encapsulation (as defined in
RFC 1483), enabling the ro uter to mul tiple x (combine) and demulti ple x (separa te)
bridged or routed protocol data units (PDUs).
For transmission, the encapsulation process adds a header from 2 to 8 octets in
length to the PDU to a llow decoding. The decoding process det ermine s the pro per
service access point (SAP).
99
†
*
9
1-12
When receiving information, the encapsulation method evaluates the header to
determine whether the PDU is a valid routed or bridged cell. If it is valid, the
encapsulation met hod then s trips t he hea der fro m the ce ll and p asses th e cell t o the
appropriate SAP for routing or bridging.
117374-C Rev 00
Understanding ATM, MPOA, ATM Router Redundancy, and OAM
You can choose from four data encapsulation types:
•LANE
•LLC/SNAP
•NULL
•NLPID
How you assign a data encapsulation type and which type takes precedence
depends on the virtual circuit type and, for PVCs, the order in which you assign
the encapsulation type.
Each ATM device must encapsulate PDUs before sending them to the SAR
sublayer.
LANE Encapsulation
LANE provides Ethernet (IEEE 802.3) or token ring (IEEE 802.5) encapsulation
of ATM PDUs for transmission over an emulated LAN. You can assign LANE
encapsulation to SVCs only.
117374-C Rev 00
LLC/SNAP Encapsulation
Logical Link Control/Subnetwork Access Protocol (LLC/SNAP; RFC 1483)
allows multip lexi ng of multi ple protocol s ov er a si ngle ATM vir tual circui t. In thi s
approach, an IEEE 802.2 Logical Link Control (LLC) header prefixes each PDU.
You can assign LLC/SNAP encapsulation to:
•PVC service records
•SVC service records
•Individual PVCs
Assigning LLC/SNAP to an SVC service record automatically uses the
Note:
technology defined in RFC 1577, Classical IP and ARP over ATM. For
information about RFC 1577, see “Classical IP over ATM Concepts
page 1-28
.
” on
1-13
Configuring ATM and MPLS Services
NULL Encapsulation
RFC 1483 refers to NULL encapsulation as “VC-based multiplexing.” This
method performs upper-layer protocol multiplexing implicitly using ATM virtual
circuits. You can assign NULL encapsulation to:
•PVC service records
•SVC service records
•Individual PVCs that are also members of an LLC/SNAP service record
Assigning NULL to an SVC service record automatically uses the
Note:
technology defined in RFC 1577, Classical IP and ARP over ATM. For
information about RFC 1577, see “Classical IP over ATM Concepts
page 1-28
.
NLPID Encapsulation
You can use Network Layer Protocol ID (NLPID; RFC 1490) in an ATM
environment for frame relay/ATM internetworking. You can assign NLPID
encapsulation to PVC service records.
” on
Selecting a Data Encapsulation Method
Generally speaking, th e desi gners of these data e ncapsul atio n methods envisioned
that NULL encapsulation would dominate in environments where the dynamic
creation of large numbers of ATM VCs is fast and economical. These conditions
usually exist in private ATM networks.
LLC/SNAP encapsulation is an alternative for environments in which it is not
practical to ha v e a se parat e VC for e ach car ried pro tocol ( for e xa mple , if the ATM
network supports only PVCs, or if billing depends heavily on the number of
simultaneous virtual circ uits).
The choice of multiplexing methods that two ATM stations use to exchange
connectionless network traffic depends on the type of virtual circuit involved:
•For PVCs, you select the multiplexing method when you manually configure
the connection.
1-14
117374-C Rev 00
Understanding ATM, MPOA, ATM Router Redundancy, and OAM
•For SVCs, the stations themselves negotiate the multiplexing method by
sending B-ISDN signaling messages. These messages include “low-layer
compatibility” information that allows negotiation of AAL5 and the carrie d
(encapsulated) pr otocol.
Routed and bridged PDUs are always encapsulated within the payload
Note:
field of the AAL5 CPCS PDU, regardless of the selected multiplexing method.
Selecting LLC/SNAP Encapsulation
When the same virtual circuit carries several protocols, select LLC/SNAP
encapsulation. LLC/SNAP encapsulation attaches an LLC/SNAP header before
the PDU. This header includes information that the receiving ATM station needs
to properly process the incoming PDU. For bridged PDUs, this header also
includes the type of the bridged media.
In NULL enca psulation, the carried netwo rk protocol is identified implicitly by
the virtual circuit connecting the two ATM stations. Because each protocol must
travel over a separate virtual circuit, there is no need to include explicit
multiplexing information in the payload of the PDU. For this reason, the
bandwidth requirements and processing overhead remain minimal.
117374-C Rev 00
You can either manually configure the carried protocol or let the signaling
procedures negotiate it dynamically during call establishment.
NULL encapsulated cells do not receive a header in a routed environment. In a
bridged environment, the content of the PDU itself includes the necessary
information for bridging the multiplexed protocols.
1-15
Configuring ATM and MPLS Services
Encapsulation Rules for PVCs
How you assi gn data encaps ulati on to individual PVCs depends to a degre e on th e
data encapsulation type you assigned to the service record that contains those
PVCs. Table 1-4
and hybrid PVCs that reside on these service records.
Table 1-4.Assigning Data Encapsulation to Individual PVCs
provides suggestions for assigning data encapsulation to PVCs
Service Record Data
Encapsulation Type
LLC/SNAPNULL or LLC/SNAPLLC/SNAP
NULLNULL or LLC/SNAPLLC/SNAP
NLPIDNLPIDNLPID
Individual PVC Data
Encapsulation Type
Hybrid PVC Data
Encapsulation Type
When assigning a data encapsulation type to a PVC or group of PVCs, keep the
following in mind:
•When you add a PVC, it reads and uses the data encapsulation type specified
in its ATM service record.
•You can globally assign a data encapsulation type to all nonhybrid PVCs in a
particular service record, or you can assign a data encapsulation type to
individual group PVCs.
•If you change the data encapsulation value for the service record, all new
PVCs that you add to that service record use the new value.
•You must assign a data encaps ulati on type to h ybrid- mode PVCs individually.
You cannot assign data enca psulatio n to a hybri d-mode PVC using the service
record.
•When you use the copy function, the new PVC uses the data encapsulation
type of the existing PVC.
1-16
•When running IP over a NULL encapsulated PVC service record, you must
change the A ddress Resolution parameter to None. You must then add an IP
adjacent host with the MAC address equal to the VPI/VCI of the PVC. See
Configuring IP Services for information about the Address Resolution
parameter.
117374-C Rev 00
PVC Access Methods
You can set up PVCs to access an ATM network in the following ways:
•Multiple PVCs per service record
•One PVC pe r service reco rd
•Hybrid access PVCs
Multiple PVCs
Upper-layer pro tocols treat each service record on an ATM network interface as a
single access point . These protocols use a single networ k address to send al l traf f ic
destined for the network to the ATM network interface. Figure 1-6
conceptual drawing of multiple PVCs accessing an ATM network through each
service record.
ATM
physical
interface
Service
record
Upper-layer
protocol
Service
record
Understanding ATM, MPOA, ATM Router Redundancy, and OAM
shows a
Site A
PVC
PVC
PVC
PVC
PVC
ATM
network
Site B
Site C
Site D
Site E
ATM0018B
Figure 1-6.Multiple PVCs per Service Record
Configuring multiple PVCs per service record uses network addressing most
efficiently. Although you need to configure each PVC manually, you need only
define and associate protocols with the ATM network service record. All the
PVCs that you configure for a given ATM service record carry the protocols that
you select and configure to run on that service record.
117374-C Rev 00
1-17
Configuring ATM and MPLS Services
When you configure multipl e PVCs per service recor d, all PVCs use the
Note:
data encapsulation type that you set for the ATM service record. See “Data
Encapsulation Methods” on page 1-12 for more information.
A configuration usi ng mult iple PVCs per service record works best in either fully
meshed environments or in nonmeshed environments where systems not directly
connected to each other ha ve no need to communicate. You can conf igure mult iple
PVCs per service record as long as you do not need to separate protocols by PVC
(that is, all PVCs accept the same protocols).
There are, however, ways to configure upper-layer protocols, such as IP or
Internetwork Packet Exchange (IPX), to al low systems in nonmeshed ne tw orks to
fully communicate. See the documentation for these protocols for more
information.
One PVC
A configuration using one PVC per service record works the same way as one
using multiple PVCs per service record. When you define only one PVC per
service record, upper-layer protocols treat the ATM network as a series of direct
point-to-point connections, viewing each PVC as an individual network interface.
1-18
You can configur e each PVC wi th different protocols and parameter set tings . This
allows you to connect to dif fer ent netw ork si tes usin g, for e xample , dif fe rent type s
of data encapsulation (Figure 1-7
).
117374-C Rev 00
Understanding ATM, MPOA, ATM Router Redundancy, and OAM
ATM
physical
interface
Upper layer
protocol
Service
record
Service
record
PVC
PVC
Figure 1-7.One PVC per Service Record
Assigning one PVC per service record allows you to dedicate a PVC to a
particular protocol, but at the expense of some configuration overhead, memory,
and address space.
This type of configuration is best suited to small, nonmeshed configurations or to
configurations in which protocols must reside on separate PVCs.
Note:
The maximum number of PVCs you can configure in this way varies,
depending on the configuration of the router, the number of protocols running
on the circuits, and the number of routing entries.
ATM
network
Site A
Site B
ATM0020A
117374-C Rev 00
Hybrid Access PVCs
PVCs do not typically allow bridging in nonmeshed environments. If your
network combines bridging and routing over the same interface, you need to use
the service record portion of each PVC for routing, while at the same time
allowing bridging to operate. To do this, you must define the PVC as a hybrid/
bridged VC.
Defining the PVC as a hybrid/bridged VC allows the bridge to view each PVC as
a separate bridge interface while allowing the routing protocols to view all PVCs
as part of the same interface (Figure 1-8
).
1-19
Configuring ATM and MPLS Services
Bridge protocol sees two interfaces
to the network
Routing
protocol
I = Interface
Bridge
I
protocol
Routing protocol sees
one interface to the network
I
I
Figure 1-8.Hybrid Access PVCs
Use hybrid PVCs when creating nonmeshed network configurations that use both
bridging and routing over a single ATM interface. These PVCs work best for
spanning tree bridging.
Note:
When you define a PVC as a hybrid/bridged VC, Site Manage r provides
additional Bridge, Spanning Tree, Source Routing (SR), SR Spanning Tree,
Translational/Learning bridge (Translate/LB), and Native Mode LAN (NML)
protocol options. These protocols run on the PVC along with the protocols
defined in the ATM service record.
ATM
network interface
Hybrid PVC
Hybrid PVC
Direction of data
Site A
ATM
network
Site B
ATM0012B
1-20
117374-C Rev 00
Understanding ATM, MPOA, ATM Router Redundancy, and OAM
Using Hybrid PVCs for Transparent Bridging
In Figure 1-9
, traff ic is bridged between site A and site B. The bridge (r out er 1) i s
running on the ATM network interface, and its PVCs are not defined as
hybrid/bridged VCs.
Site ASite B
Bridge port sees one
path to Sites A and B
C
Router 1
ATM
network
D
Figure 1-9.Example of a Bridged Network
Router 2
Router 3
A
E
B
F
ATM0013A
117374-C Rev 00
In this exampl e, when the bridge receives da ta from site A and does not recognize
the destination address, it tries to direct traffic through another bridge port.
However, because the PVCs are not defined as hybrid/bridged VCs, the ATM
bridge port views the paths to site A and site B as the same.
A bridge does not send the same data over the bridge port fr om which it just
received the data, so the bridge cannot direct the data to site B. To resolve this
problem, you need to designate the PVCs on router 1 as hybrid/bridged VCs.
If you define the PVCs as hybrid VCs, each PVC acts as a separate bridge port.
This enables the bridge running on the ATM in terface to view the traffic from
site A as arriving on a different port from that of site B. When the bridge sends
data, it now has access to all its ports, including the port that accesses site B.
Therefore, data from site A can reach site B.
1-21
Configuring ATM and MPLS Services
SVC Access Methods
SVCs use signaling messages to dynamically establish, maintain, and clear a
switched virtual connec tion at the UNI. These message s (as defi ned by the Q.2931
standard for signaling protocols) allow the router to assess the availability of an
ATM end point (device), establish a connection with that device, mainta in that
connection for the duration of data transfer, and then clear the connection when
the transfer is complete.
Assigning ATM Addresses
An ATM address is composed of a network prefix and a user part. Bay Networks
routers use the autogeneration feat ure to create the user part of the ATM address
by combining the MAC address of the ATM interface with a unique selector byte
to create unique addresses (Figure 1-10
manually.
390000000000000000000000000000A20CA98F00
). You can also enter ATM addr esses
ATM address
1-22
Network prefix
39000000000000000000000000
0000A20CA98F00
MAC address
0000A20CA98F00
Figure 1-10.ATM Address Components
User part
Selector
byte
ATM0037A
117374-C Rev 00
Understanding ATM, MPOA, ATM Router Redundancy, and OAM
Entering an ATM Address Network Prefix
The ATM address network prefix specifies the ATM domain of which the service
record is a part. This 13-byte portion of the ATM address can range from
XX000000000000000000000000 to XXFFFFFFFFFFFFFFFFFFFFFFFF.
The XX byte must contain a value of 39, 45, or 47. These values define the
authority and format identifier (AFI). The AFI byte identifies the group
responsible for allocating the prefix and the format the p refix uses. For more
information about the AFI byte, refer to the ATM Forum UNI specification.
Entering an ATM address ne twork pr ef ix is opt ional. If you do not e nter a netw ork
prefix in the spe cif i ed r ange, t he ser vice r ec ord acc epts t he first prefix v alue t hat it
receives from the switch.
Entering an ATM Address User Part
The ATM address user part (su ff ix) consi sts of a 6-byt e end-statio n identif ier and a
1-byte selector field. This 7-byte portion of the ATM address can range from
00000000000000 to FEFFFFFFFFFFFF.
You can either allow the router to generate this value automatically, or you can
enter the value manually.
ATM Traffic Parameters
The ATM User-Network Interface Specification defines the following traffic
parameters:
•Peak cell rate (PCR) -- The upper traffic rate limit for a n individual VC
•Sustainable cell rate (SCR) -- The upper bound on the conforming average
rate of an individual PVC or control VC
•Maximum burst s ize (MBS) -- Th e maximum length of a cell stream allowed
on a particular VC
These parameters help to prioritize and cont rol the traffic on each VC. How you
configure your ATM traffic parameters depends on the characteristics of the
individual c onnection s that you w ant to set up (for e xample, t he desired maximum
cell rate, average cell rate, and burst size).
117374-C Rev 00
1-23
Configuring ATM and MPLS Services
You can change ATM traffic parameters several times before deciding on a
particular set. The following sections describe the traffic parameters and provide
some basic guidelin es for customizing them on an ATM PVC or ATM control VC
(that is, the signaling VC or ILMI VC).
You do not need to manually configure traffic parameters for SVCs (as
Note:
you must for PVCs and control VCs), because SVCs dynamically negotiate
these parameters before sending data.
Using the PCR
The PCR specifies the upper traffic limit, in cells/second, that the ATM
connection can support.
How you set the PCR de pends on:
•The optical transmission rate of your ATM device
•The amount of traffic you expect on a particular VC
•The rate you want for each VC
When setting the PCR, keep the following considerations in mind:
1-24
•Each VC can have its own PCR.
•The PCR cannot exceed the maximum rate for the physical media. For
example, you cannot exceed 149.76 Mb/s for an OC-3c line.
•The PCR specifies the desired rate for the attached physical media (that is,
OC-3c, DS-3, or E-3). It does not specify the rate for the ATM network as a
whole. For example, you can specify a full 149.76 Mb/s for each PVC or
control VC on an OC-3c connection. However, if the VC ultimately connects
to a lower-speed link (for example, T1 or E1), your PCR is limited to the
maximum rate for that media.
•ATM VCs may fail to operate with PCR values lower than 128 cells/s.
•The E-3 framing mode setting affects the maximum PCR setting.
117374-C Rev 00
Understanding ATM, MPOA, ATM Router Redundancy, and OAM
Using the SCR
The SCR is the upper bound on the conf orming average rate of an indivi dual PVC
or control VC. The average rate is the number of cells transmitted over the link
divided by the duration of the connection. The duration of the connection is the
total amount of time it takes from connection setup to connection release.
The SCR allows you to define future cell flow on a PVC or control VC in greater
detail than by using only the PCR. The SCR controls the rate over time -- not at a
specific instant of time -- and can help you use your ne twork resources more
efficiently. In othe r words, the SCR allows sufficient bandwidth for operation, but
does not allow a bandwidth as high as the PCR.
The SCR value maps directly to an MCR (minimum cell rate) value. In other
words, when you conf igu re the SCR on a device, you actually configure th e uppe r
bound of an a vera ge rate. Like t he SCR, the MCR def ines the minimum am ount of
guaranteed bandwidth allowed for PVCs and control VCs on the ATM line. The
MCR (that is, SCR) not only controls the rate over time, it guarantees this rate.
When setting the SCR, keep the following considerations in mind:
117374-C Rev 00
•The SCR maps directly to the MCR.
•The MCR provides guaranteed bandwidth for PVCs and control VCs while
allowing sufficient ban dwidth for S VCs to opera te.
•To be useful, the SCR must not exceed the PCR.
•If you know the user average rate, set the SCR approximately 10 percent
higher than this value.
•ATM VCs may fail to operate with SCR values lower than 128 cells/s.
•Entering 0 for the SCR turns off this function and specifies that the ATM
router uses “best effort” for SCR.
•The E-3 framing mode setting affects the maximum SCR setting.
1-25
Configuring ATM and MPLS Services
Using the MBS
The MBS specifies the maximum number of sequential cells allowed on a VC
before that VC must relinquish bandwidth to other VCs waiting to transmit. This
burst occurs at or close to the peak cell rate.
When setting the MBS, we suggest that you select a value larger than the largest
packet your PVC or control VC can transmit (that is, the size of the maximum
AAL CPCS transmit SDU). For example, if your VC accepts packets that are less
than 4608 bytes long (PVC default), set your MBS value between 45 and 50 cells.
As a guideline, use this formula to determine your MBS value:
Maximum packet size (in bytes)
48 bytes/cell
For example:
4608 bytes (default)
48 bytes/cell
= MBS value (in cells)
= 96 cells
ATM0016A
1-26
117374-C Rev 00
Understanding ATM, MPOA, ATM Router Redundancy, and OAM
ARP and Inverse ARP Support
ATM supports the Address Resolution Protocol (ARP), enabling the router to
dynamically resolve IP network layer protocol-to-VPI/VCI address mappings.
ATM learns the address of the virt ual circuit by d etecting the virtual cir cuit that
delivered the ARP response.
ATM also supports Inverse ARP. However, you can use Inverse ARP only if both
the local and remote routers support it.
Bay Networks uses both pro prietar y and sta ndard ARP and Inverse ARP for PVCs
that run IP. The method that the PVC uses depends on how you configure address
resolution for the IP interface. See Configuring IP Services for additional
information about configuring address resolution.
Bay Networks use s s ta ndar d ARP and Inverse ARP for SVCs r unni ng classical IP
(RFC 1577) and SVCs running LAN emulation.
ATM Error Checking
ATM verifies that the VPI/VCI is valid with respect to the PVCs configured for
the ATM circuit. It also verifies the header format. ATM verifies valid SVC
connection through signaling messages.
Simulated Multicast Packet Support
Simulated multicasting is generally used in certain address resolution techniques
and for applications that require the d elivery of identical information to multiple
recipients. Bay Netwo rks ATM router s simulate mult icasti ng by se nding a cop y of
the multicast or broadcast packet to every available virtual circuit on a particular
logical interface.
Converting Mb/s to Cells/s
Several ATM traffic parameters require you to enter values in cells per second
(cells/s). To convert to cells/s, divide the number of bits/s by 424 (the number of
bits per ATM cell).
117374-C Rev 00
1-27
Configuring ATM and MPLS Services
Number of bits/second
Number of bits/ATM cell
For example:
100,000,000 bits/s
424 bits/cell
Classical IP over ATM Concepts
RFC 1577, Classical IP and ARP over ATM, describes an administrative entity
within an ATM network called a logical IP subnet (LIS). Each ATM LIS consists
of multiple network de vice s -- hosts and router s -- connected to the ATM network
and configured with interfaces to the same IP subnet.
Each LIS operates and communicates independently in an ATM network. A host
connected to an ATM network communicates directly with other hosts in its own
LIS. To communicate with hosts in another LIS, the host must use an IP router.
This router c an connect to multiple LISs.
= Number of cells/second
= 235,849 cells/s
ATM0021A
1-28
An ATM LIS must meet the fo llowing requirements:
•All memb ers of the LIS (hosts and routers) must have the same IP network/
subnet address and mask.
•All members must be directly connected to the ATM network using SVCs.
•All members must access hosts outside the LIS through a router.
•All members must be able to communic ate by means of ATM with e very ot her
member of the LIS (that is, the virtual connection topology must be fully
meshed).
117374-C Rev 00
Understanding ATM, MPOA, ATM Router Redundancy, and OAM
An ATM LIS can replace an IP LAN. In Figure 1-11, for example, three IP host
systems and an IP router have interfaces to an Ethernet LAN. To communicate
with each other on the LAN, the devices use the MAC addresses that they obtain
using ARP or static routes. For communication beyond the LAN, the devices use
IP addresses.
140.250.200.1
00 00A2 00 00 01
Host
C
140.250.200.4
00 00A2 00 10 40
ATM0035A
140.250.200.0
Host
A
140.250.200.2
00 00A2 00 10 20
Host
B
140.250.200.3
00 00A2 00 10 30
Router
Figure 1-11.IP Local Area Network
In Figure 1-12, an ATM network replaces the LAN interfaces, creating a LIS. For
communication within the LIS, the devices use ATM addresses obtained using
ATMARP; for communication beyond the LIS, the devices use IP addresses.
For example, to send a message to host B, host A uses host B’s ATM address. To
send a message to a h ost beyond the LIS, host A uses an IP address to identify the
remote host and sen ds the me ssage to t he local router ( route r A), using the rout er’s
ATM address. The router then forwards the message.
117374-C Rev 00
1-29
Configuring ATM and MPLS Services
ATMARP client
Router A
ATM network
ATMARP server
Host
A
Host
B
Key
SVC supporting ATM LIS
Host
Figure 1-12.IP Logical IP Subnet
Router B
C
ATM0037A
1-30
117374-C Rev 00
Understanding ATM, MPOA, ATM Router Redundancy, and OAM
ATM Address Resolution
An address resolution protocol defines a mechanism that enables an IP router to
use the IP address of a ne tw ork device to learn the physical addres s of th at device.
An Ethernet LAN uses ARP as its address resolution scheme. A LIS uses an
address resolution scheme called ATMARP as defined by RFC 1577.
On a LAN, defined as a broadcast medium, a router obtains the physical address
of a network device by broadcasting an ARP request. In a LIS, which uses a
nonbroadcast ATM medium, a router sends an ATMARP request to an ATMARP
server.
Each IP interface on the LIS opens a V C to the ATMARP server and registers its
IP address and ATM address (Figure 1-12
the server builds and maintains a table that maps LIS IP addresses to ATM
addresses.
A router that needs the ATM address of a host on the LIS sends an ATMARP
request to the server. When the server returns a response containing the address,
the router extracts the ATM address of the host from the response and opens an
SVC directly to the host using ATM UNI signaling.
on page 1-30). Using this information,
117374-C Rev 00
If the server does not have an entry for the re quested IP address, it returns a
negative acknowledgment, signifying that the destination is unreachable.
1-31
Configuring ATM and MPLS Services
Configuring an ATM Service Record for ATMARP
When configuring a service record to act as an ATMARP client or server:
•Define a classical IP service record.
--Specify SVC as the v irtua l circui t type for t he serv ice re cord. Al l netw or k
devices on a LIS must connect ove r SVCs.
--Specify LLC/SNAP or NULL as the encapsulation type for the service
record. RFC 1577 defines LLC/SNAP as the encapsulation type for
ATMARP.
For more information about how to create an SVC service record to run
classical IP, see Chapter 3, “Starting ATM, ATM MPOA Server, ATM Router
Redundancy, and MPLS.”
•Add IP and IP routing protocols to the circuit.
•Configure ATMARP Mode as either a client or a server.
For a description of the ATM-specific IP parameters necessary for classical IP
ATMARP operation over ATM, see “ATMARP Parameters” on page A-102.
For full compatibility with RFC 1577, you may have to specify a maximum
transmission unit (MTU) size of 9188 bytes. See “Defining the Interface MTU”
on page 4-3 (for the BCC) or the Site Manager Interface MTU parameter
description on page A-6 for additional information about setting the MTU size.
1-32
When configuring th e router as a client , you must def ine the serv er switch add ress.
The client sends ATMARP requests to the server switch address.
Note:
If you remove and replace a link module that is configured as an
ATMARP client, the client loses connectivity until the ATMARP server
registration refresh interval for that client expires (900-second default). This
loss of connectivity occurs only when the client is configured to autogenerate
the ATM address user part (see “Disabling and Reenabling User Part
Autogeneration” on page 7-4 for additional information about autogenerating
ATM addresses).
117374-C Rev 00
Understanding ATM, MPOA, ATM Router Redundancy, and OAM
Configuring an ATM Address for an Adjacent Host
An adjacent host is a network device on the local LIS. You must configure an
ATM address for all hosts on the LIS that do not use ATMARP.
For a description of the IP parameters required for creating adjacent hosts in a
classical IP over ATM environment, see “Adjacent Host Parameters” on
page A-104.
For more information about adding, editing, and deleting adjacent hosts, see
Configuring IP Services.
ATM LAN Emulation Concepts
LAN emulation allows virtual communication of traditional LAN devices and
applications over an ATM network. An ATM network can run one or more
emulated LANs. However, each emulated LAN is independent of the others and
devices cannot communicate directly across emulated LAN boundaries.
Note:
Communication between emulated LANs is possible through routers
and bridges only (possi bly implemented on the same end station).
This section provides general information about LAN emulation as described by
the ATM Forum. For more information about LAN emulation, refer to the ATM
Forum document LAN Emulation Over ATM (Version 1.0).
For instructions on how to customize LAN emulation on your ATM router, see
Chapter 8, “Customizing LAN Emulation Service Records and Clients.”
LAN Emulation Connectivity
An emulated LAN can provide Ethernet (IEEE 802.3) or tok e n rin g (IEEE 802.5)
connectivity. With an emulated Ethernet or token ring network over ATM,
software applications can interact as if they were connected to a traditional LAN.
117374-C Rev 00
1-33
Configuring ATM and MPLS Services
LAN Emulation Components
Each ATM domain contains a LAN emulation configuration server (LECS). Each
emulated LAN comprises a group of LAN emulation clients (LECs), a LAN
emulation server (LES), and a broadcast and unknown server (BUS). These
servers provide specific LAN emulation services.
LAN Emulation Configuration Server
The LAN emulation configuration server (LECS) assigns individual LE clients to
different emulated LANs. The LECS does this by giving the client the ATM
address of the LAN emulation server (LES). This method allows you to assign a
client to an emulated LAN based on the client’s physical location (ATM address)
or the identity of a LAN destination it re presents.
LAN Emulation Clients
The LAN emulation client (LE client or LEC) is the interface, or virtual portion of
an interface, through which an end station forwards data, resolves addresses, and
provides other control functions. The LE client provides the MAC-level emulated
Ethernet or tok en rin g servic e int erfac e to the upper -lay er prot ocol. It also con trols
the LAN emulation UNI (LUNI) interface when communicating with other
devices on the emulated LAN.
1-34
The LE client uses the configuration protocol to obtain information from the
LECS. This protocol allows the LE client to locate the LES and set up a
bidirectional, control direct virtual channel connection. The LE client
automatically obtains all of the necessary configurat ion data (including the LES
address) from the LECS.
LAN Emulation Server
The LES controls and coordinates LE client access to the emulated LAN. When
an LE client joins an emulat ed LAN, it registers its ATM address with the LES.
When the LES obtains the ATM address of the LE client, it also obtains the
client’s physical location (from the MAC address or route descriptor).
LE clients query the LES to obtain the ATM address associated with a specific
MAC address or route descriptor. After an LE client receives the ATM address of
the LE client it wants to reach, the individual clients communicate directly.
117374-C Rev 00
Broadcast and Unknown Server
T o emulate a tradi tional LAN, the emul ated LAN must provi de the connect ionless
data deli very of a shared network to its LE cl ie nts and be able to handle broadcast
and multicast data. The broadcast and unknown server (BUS) fulfills this
requirement by distributing all broadcast, multicast, and unknown traffic to and
from all LE clients on an emulated LAN.
For example, when an LE client sets up its initial configuration, it obtains the
MAC address of the BUS from the LES. Using this MAC address, the LE client
sets up a multic ast se nd vir tual channe l conne ction (VCC) t o the B US. In tu rn, the
BUS registers t he LE client as part of its emulated L AN.
To broadcast data, an LE client uses the multicast send VCC to transmit
information to the BUS. The BUS then retransmits the data, through multiple
point-to-point connections or one point-to-multipoint connection, to each LE
client on the emulated LAN.
Redundant LES/BUS
Understanding ATM, MPOA, ATM Router Redundancy, and OAM
117374-C Rev 00
Bay Networks ATM routers support LAN emulation server (LES) and broadcast
and unknown server (BUS) redundancy. This redundancy reduces the risk of
network failure by overcoming a single point of failure in accessing the LES.
1-35
Configuring ATM and MPLS Services
LAN Emulation States
As defined in the ATM Forum LAN Emulation Over A T M specif ica tion, L E clients
enter va ri ous st at es of communication while attempting to jo in an emulated LAN.
These states (referred to as “phases” by the ATM Forum) indicate the progress of
an LE client as it connects with an emulated LAN (Figure 1-13
).
Initial state (1)
LECS connect state (2)
After experiencing
any failure, or
terminating its
connection to the
emulated LAN,
the LE client
returns to the
initial state.
Configure state (3)
Join state (4)
Initial registration state (5)
BUS connect state (6)
Operational state (7)
Figure 1-13.LAN Emulation States
Note:
The numbers that follow each state appear in the ATM LEC status
record (wfAtmLecStatusEntry).
If the LE client
loses the BUS
connection, it can
attempt to reconnect.
ATM0034A
1-36
117374-C Rev 00
Understanding ATM, MPOA, ATM Router Redundancy, and OAM
The following sections briefly describe each LAN emulation state. Refer to the
ATM Forum LAN Emulation Over ATM specification for more information about
LAN emulation states.
Initial State
An LE client always starts in the initial state before attempting to connect to the
LECS.
LECS Connect State
An LE client enters the LECS connect state when it attempts to connect to the
LECS.
Configure State
An LE client enters the configure state when it attempts to retrieve the necessary
information (that is, the ATM address of the LES, LAN type, LAN name,
Maximum MTU, and various timeout values) required to join an emulated LAN.
117374-C Rev 00
Join State
An LE client enters the join state when it attempts to join an emulated LAN.
Joining an emulated LAN requires that the LE client:
•Set up a control VCC to communicate with the LES.
•Send a join request (c ontaining the client MAC address) to the LES.
•Accept a control distributed VCC to receive control data from the LES.
•Receive a valid join response from the LES containing a LAN emulation
client ID (LECID).
Only one ATM LE client per Bay Networks ATM router can join an
Note:
emulated LAN at any point in time. However, you can always move a LAN
emulation client t o t he app rop ri at e e mu la ted LAN us in g ne twork management
software.
1-37
Configuring ATM and MPLS Services
Initial Registration Sta te
An LE client enters the initial registration state when it attempts to register
multiple MAC addresses with the LES.
The router LE client provides the MAC address only for its own ATM
Note:
interface. Because it does not register multiple MAC addresses, the router
never enters this state. The router LE client acts as a proxy for bridge MAC
addresses not lear ned on this circuit.
BUS Connect State
An LE client enters the B US connec t stat e when i t att empts t o set u p a VCC to th e
BUS.
Operational State
An LE client enters the operational state after successfully completing the
requirements to join an emulated L AN.
1-38
117374-C Rev 00
Understanding ATM, MPOA, ATM Router Redundancy, and OAM
Multi-Protocol over ATM Concepts
As defined by the ATM Forum, Multi-Protocol over ATM (MPOA) maps routed
and bridged traffic flows to ATM SVCs, thus mitigating the performance
limitations imposed by hop-by-hop routing of individual packets. This technique
of mapping identifiable traffic flows to virtual channels creates network
“shortcuts” between source and destination, and is generally referred to as
cut-through or zero-hop routing.
Cut-through routing i s based o n the f act t hat, in most cases , data t ransfe r occur s at
a steady rate of flow. For example, data or file transfer from one legacy Ethernet
LAN to a remote counterpart usually involves multiple frames. A file transfer of
approximately 45 KB requires about 30 Ether net frames, all address ed to the same
destination.
In an MPOA environment, it is possible to identify the recipient from addressing
data contained within the first frame and to establish an SVC to the recipient.
Then all 30 or so frames can be broken down into approximately 900 ATM cells
and transmitted to the recipient via the virtua l channel provided by the SV C.
Network performa nce i mp roves as the cells follow a predetermined direct path, in
contrast to the hop-by-hop routing of the Ethernet frames. Network performance
improves markedly in the case of steady-stream deterministic data flows, such as
video.
MPOA Logical Components
MPOA operati ons are based on l ogical compone nts, which can be im plemented in
various configurations of hardware and software. MPOA logical components
include the following:
•MPOA client
The MPOA client (MPC) resides in an ATM edge device adjacent to the
router. BayRS does not provide MPC functionality. The primary function of
the MPC is to act, in ATM terminology, as an ingress or egress point for traffic
using network cut-throughs.
117374-C Rev 00
1-39
Configuring ATM and MPLS Services
The MPC moni tors traffic flows between a local source an d remote
destinations. Wh en traffic volume between a s ource a nd a d estin ation e x cee ds
a preconfigured threshold level (for example x packets to the same network
layer address in y seconds), the MPC attempts to set up an SVC between the
source and destination workstations.
In attempting to set up an SVC, the MPC first looks in a local cache of
network layer-to-ATM address mappings. If the MPC finds the destination
address in its local cac he, it immediately be gins to est ablish the SVC. If it f ails
to locate the destination address in the local cache, it generates an MPOA
address resolution request to an adjacent MPOA server.
•MPOA router
The MPOA router is a collection of logical functions that map network layer
addresses to ATM addresses. The MPOA router maintains tables of adjacent
network layer (IP), MAC layer, and ATM addresses, in addition to standard
routing tables derived from a routing protocol (generally OSPF or RIP).
MPOA routers communicate over NHRP to map network layer addresses to
ATM addresses. BayRS provides MPOA router functionality to map IP
addresses to their ATM counterparts.
•MPOA server
The MPOA server (MPS) is a logical function that mediates between local
MPCs and the MPOA router. It receives MPOA address resolution requests
from MPCs and passes them to the MPOA routing function. The MPOA
router, using NHRP, resolves the address and passes the requested ATM
address back to the MPS. The MPS, in turn, forwards the resolved address to
the requesting MPC. BayR S provides MPOA server func tionality.
MPOA Basic Elements
MPOA services are dependent on three basic elements:
•LANE
LANE (LAN emulation) provides transparent support for legacy LANs in an
ATM topology. LANE enables intrasubnet communication, and MPOA
provides intersub net communication.
1-40
117374-C Rev 00
Understanding ATM, MPOA, ATM Router Redundancy, and OAM
•Next Hop Resolution Protocol (NHRP)
NHRP is a sophisticated address resolution protocol that maps network layer
addresses (for example, IP addresses) to NBMA addresses (in the case of
BayRS, ATM addresses). For instructions on conf iguring an NHRP serv er , see
Appendix C, “Configuring NHRP for ATM Services.”
•Virtual router
A virtual router is a set of network devices and associated software that
collectively provides the functionality of multiprotocol routed networks.
Establishing a Network Cut-Through
MPOA componen ts and elements fu nction together to establish an SVC bet ween a
source host and a destination host (Figure 1-14
1.
A local MPC monitors traf f ic f lo w and mainta ins a count of pack ets addre ssed
over a specific interval to remote hosts. When the count exceeds a threshold
value, the MPC attempts to establish an SVC to the host.
To establish the virtual connection, the MPC needs the ATM address of the
host.
).
117374-C Rev 00
2.
The MPS first checks a local address re solution cache to map th e known
network layer address with an ATM equivalent.
3.
If the cache search fails, the MPC issues an MPOA resolution request to the
local MPS function resident on the adjacent router.
4.
The local MPS hands the resolution request to the MPOA router component.
5.
The MPOA rou ter gener ates an NHRP address resolut ion reque st for t he ATM
address of the destination host. Standard routing protocols move the NHRP
request through the ne twork toward the destination host. Eventually, the
NHRP request reaches the egress router, that is, th e router that serves the host.
6.
The egress router forwards the request to the remote MPS.
7.
The remote MPS provides the ATM address of the destination host to the
remote MPOA router.
If the destination host is connected to a legacy LAN, the MPS provides the
ATM address of the router that connects to the legacy LAN. If the destinat ion
host is ATM-attached, the MPS provides the ATM address of the destination
host.
1-41
Configuring ATM and MPLS Services
8.
The remote MPOA router generates an NHRP address resolution reply
containing the ATM address provided b y th e MPS. Standar d routing pr otocols
move the NHRP reply through the network to the local MPOA router.
9.
After caching the addr ess resolution information, the MPOA router sends the
resolved addr es s t o th e M PS, whi ch, i n t urn, sends the resolved address to the
MPC that initiated the resolution process.
10.
The local MPC caches the ad dress reso lution inf ormati on and no w esta blishes
an SVC, either directly t o t he de st ination host (if that host is an ATM device),
or to the egress router that serves the dest ination host.
ATM MPC
10BASE-T
ATM network
ATM MPS
ELAN 1
ELAN 2
10BASE-T
ATM MPC
Figure 1-14.MPOA with Cut-Through VC
Note:
Creating a separate NHRP control VC is optional. If you configure
NHRP on each LEC, the control information passes over the emulated LAN.
1-42
ELAN 3
ELAN 4
ATM MPS
Key
Cut-through VC
Logical connection
NHRP control VC
ATM0055A
117374-C Rev 00
Understanding ATM, MPOA, ATM Router Redundancy, and OAM
ATM Router Redundancy Concepts
Bay Networ ks ATM routers support warm sta ndby router redundancy. This
redundancy protects a network from the irrecoverable failure of an entire ATM
router. You configure routers to be members of a router redundancy group.
Token ring
100BASE-T
10BASE-T
Token ring
Figure 1-15
Centillion 100 switch
Centillion 100 switch
Centillion 100 switch
Centillion 100 switch
illustrates rout er redundancy in a Bay Networks ATM environment.
ATM network
ATM router 1
ELAN 1
ELAN 2
ELAN 3
ELAN 4
Key
ATM router 2
Physical connection
Logical connection
ATM00002
Figure 1-15.Router Redundancy
117374-C Rev 00
1-43
Configuring ATM and MPLS Services
Router redundancy requires at least two routers to be members of a router
redundancy group. One router acts as the primary router and provides normal
routing/bridging services. The other router acts as the secondary, backup router
and takes over if the primary router fails.
All members of an ATM router redundancy group must have the following
characteristics:
•Be the same ATM router type, for example, BCN
•Have the same hardware configuration, including CPU and interface module
types, and slot and port locations of the backed-up interfaces.
•Have the same sof tw a re configuration. That is, you must be running the same
router software ve rs ion and have the same loadable modul es configured.
•Contain both a primary configuration file and a secondary configuration file.
•Have at least one legacy LAN connection between the routers in the
redundancy group.
Note:
Bay Networks recommends at least two legacy LAN connections
between routers in the redundancy group.
®
, BLN®, or System 5000™.
PVC Operations and Management Concepts
In most ATM networks, if a PVC fails, the remote device does not receive
notification of the failure at the ATM layer. Instead, the device receives this
information from a non-ATM source such as a routing protocol that operates
above the ATM layer. The PVC Operations and Management (OAM) feature
provides a mechanism by which ATM devices can receive prompt failure
information.
PVC OAM has two methods of detecting PVC failure: loopback and alarms. You
can use either of these methods separately or both methods together. When using
OAM loopback cells, the time can decrease to only a few seconds; when using
OAM alarms, the detection is almost instantaneous.
1-44
117374-C Rev 00
OAM Loopback
The OAM loopb ack meth od u ses lo opback c ells to det ect a lost connec tion. When
OAM loopba ck is enabled, the PVC sends loopback cells at a designated interval.
If a remote device does not return loopback responses, and the PVC loses a
specified number of cells, the service record alerts the upper-layer protocol that
the link is down . The PVC continues to se nd O AM loopbac k cells ov er the se rvice
record. When it receives a specified number of OAM loopback responses, it
declares the link operational and begins sending traffic again.
Note:
PVCs on an interface. Link status is based on the OAM loopback status of all
PVCs on the in terface.
OAM Alarms
If the ATM router connects to an ATM switch that uses OAM alarms, you can
enable alarm detection on the ATM router. With OAM alarms enabled, when the
router recei ve s an alar m indicat ion signa l (AIS) a larm from the switch, i t alert s the
upper-layer protocol that the ATM link is down; it does not wait for an OAM
loopback response fr om the remote device.
Understanding ATM, MPOA, ATM Router Redundancy, and OAM
For OAM loopback to function properly, you must configure it on all
117374-C Rev 00
After receiving an AIS alarm, the router sends a remote defect indication (RDI)
response to the switch to indicate that it received the alarm. The switch continues
to send AIS alarms u ntil the link is operat ional ag ain. If the rout er does not rece i ve
an AIS alarm for 3 seconds, it declares the link operational and begins sending
traffic again.
1-45
Configuring ATM and MPLS Services
For More Information
For more information about ATM, refer to the following documents:
ATM Forum. ATM User-Network Interface Specification. Version 3.0. September
1993.
ATM Forum. LAN Emulation Over ATM. Version 1.0. January 1995.
Bellcore Document SR-NWT-001763, Issue 1. Prelimi nar y Report on Broadband
ISDN Transfer Protocols. December 1990.
———, FA-NWT-001109. Broadband ISDN Transport Network Elements
Framework Generic Criteria. December 1990.
———, FA-NWT-001110. Broadband ISDN Switching System Framework
Generic Criteria. December 1990.
De Prycker , M. Asynchr onou s T r ansfer Mode: Soluti on for Br oad band ISDN. Ellis
Horwood Limited, 1991.
Grossman, D., E. Hoffman, F. Liaw, A. Malis, A. Mankin, and M. Perez. ATM
Signaling Support for IP over ATM. RFC 1755. Network Working Group.
February 1995.
1-46
Handel, R., and M. Huber. Integrated Broadband Networks: An Introduction to
ATM-Based Networks. Reading, Mass.: Addison-Wesley, 1991.
Heinanen, J. Multiprotocol Encapsulation over ATM Adaptation Layer 5.
RFC 1483. Network Working Group. July 1993.
ITU-T. B- ISDN -- ATM Adaptation Layer -- Service Speci f ic Co nnec tion Orie nted Protocol (SSCOP). Final Draft. March 10, 1994.
Laubach, M. Classical IP and ARP over ATM. RFC 1577. Network Working
Group. January 1994.
117374-C Rev 00
Where to Go Next
Use the following table to determine where to go next.
If you want toGo to
Learn about MPLS concepts.Chapter 2
Start ATM or MPLS.Chapter 3
Change default settings for ATM interface
parameters.
Change default settings for ATM signaling
parameters.
Change default settings for ATM PVC service record
and PVC parameters.
Change default settings for classical IP service
record parameters.
Change default settings for LAN Emulation client
service record parameters.
Change default settings for Multi-Protocol over ATM
server parameters.
Change default settings for the ATM router
redundancy parameter.
Change default settings for MPLS parameters.Chapter 11
Obtain information about Site Manager parameters.Appendix A
Monitor ATM using the BCC show commands.Appendix B
Configure NHRP for MPOA services.Appendix C
Understanding ATM, MPOA, ATM Router Redundancy, and OAM
Chapter 4
Chapter 5
Chapter 6
Chapter 7
Chapter 8
Chapter 9
Chapter 10
117374-C Rev 00
1-47
Chapter 2
Understanding MPLS
This chapter desc ribes th e conce pts un derlyi ng MPLS and, where a ppropr iate, th e
specific ways Bay Networks implements these concepts on its routers. It contains
the following information:
TopicPage
MPLS General Information2-2
The MPLS Network2-5
Supported Proto cols2-7
For More Information2-7
Where to Go Next2-8
117374-C Rev 00
2-1
Configuring ATM and MPLS Services
MPLS General Information
Multiprotocol Label Switching (MPLS) is an emerging Internet Engineering T ask
Force (IETF) sta ndard that i s currentl y in draft fo rm. Its pr imary goal is to provi de
a standardized solution that superce des existing proprietary solutions for
integrating label-swapping and forwarding with ne twork layer routing. MPLS
works in an environment where traditional network layer routing protocols (for
example, OSPF and BGP) are used to maintain the routing topology and
forwarding in formation base (FIB) for each router.
In connectionless networks (those using connectionless network layer protocols),
as a packet travels from one hop to another, each router must determine where to
forward the packet based on the individual packet header. This decision process
can be broken down into two major tasks: classifying a set of packets as part of a
forwarding equivalence class (FEC) and mapping each FEC to a next hop.
By classifying a set of packets as part of an FEC, the router uses the same
forwarding criter ia for each packet. All packets that belong to a particular FEC
and that trav el from a particular node foll ow the same path. This group of packets
is called a “stream.” A packet stream is a group of packets that follows the same
path to a destination. I n a con v ention al IP netw ork, each ro uter hop e xamines ea ch
packet to determine its destination.
2-2
Using MPLS, the examination of the packet is done only once. The first router
assigns a label that defines the specific packet stream. Each intervening router
then forward s packets ba sed on the fixed-length labels. Labels reside in the label
information base (LIB), which contains both inbound and outbound labels
associated with inbound and outbound interfaces.
Looking up a label is faster than interpreting the destination of an individual
packet and routing data based on that destination. By assigning labels to packets
or packet streams, the transmission speed of your network increases.
117374-C Rev 00
MPLS System Overview
The Bay Networks MPLS implementation consists of three major components :
•Label distribution entity
•MPLS label management (MLM) entity
•Forwarding entity
Understanding MPLS
Other
interfaces
Figure 2-1
OSPF/RIP
IP
Forwarding
illustrates the basic MPLS system architec tur e.
Routing
table
LDP
MLM
Driver
Key
MIB
External
component
MPLS
component
Data path
Control path
Figure 2-1.The MPLS System
117374-C Rev 00
ATM0058A
2-3
Configuring ATM and MPLS Services
Label Distribution Entity
The label dis tribution entity is essentially the implementation of the label
distribution protocol (LDP). LDP is the set of procedures and messages by which
label-switching routers (LSRs) establish label-switched paths (LSPs) through a
network. LDP establishes these paths by mapping network layer routing
information directly to data link layer switched paths.
LDP associates a packet stream with a specific LSP and assign the LSP a specific
label. The label infor mation is distributed betwe en the LSRs and LERs to
maintain stream mapping information.
MPLS Label Management
The MPLS label management (MLM) entity communicates with LDP. It is
responsible for:
•Establishing the default VC (0/32)
•Responding to requests from LDP (for example, requests for a label and
establishing VC communications)
Forwarding
2-4
•Communicating with the ATM driver to set up and tear down VCs
The forwarding entity encapsulates and decapsulates the data that it sends and
receives over the MPLS interface.
Outbound data is delivered to the encapsulation process by the higher layers and
delivered to the lower-level driver for transmission to the MPLS network.
Inbound data is received from the MPLS network by the lower-level driver and
delivered to the decapsulation process, where it is stripped of layer 2 protocol
headers. The decapsulation process then passes the inbound data to higher layers
for further processi ng.
117374-C Rev 00
The MPLS Network
The MPLS network (Figure 2-2) consists of two major components:
•Label switching router (LSR)
•Label edge router (LER)
Understanding MPLS
100BASE-T
10BASE-T
LER
MPLS network
LER
100BASE-T
LSR
LER
LSR
LSR
100BASE-T
LER
Key
Label-switching router (LSR)
Label edge router (LER)
Virtual connection
10BASE-T
100BASE-T
Figure 2-2.Sample MPLS Network
117374-C Rev 00
ATM0057A
2-5
Configuring ATM and MPLS Services
Label Switching Router
A label-switching router (LSR) is a router that contains all label- switching
interfaces. The LSR controls MPLS forwarding in the MPLS network. An LSR
performs table lookup on received packets and, based on the packet label,
forwards the packet or packet stream to the specified outgoing inter f a ce. The LSR
swaps the labels of the packet headers before transmitting the packets to the
MPLS network.
Note:
An ATM LSR currently consists of an ATM switching device th at works
in conjunction with a UNIX Ultrasparc workstation running Solaris and LDP.
For information about how to configure an ATM LSR, refer to the
documentation provided with your switching device.
Label Edge Router
A label edge router (LER) is an LSR that resides between the IP and MPLS
networks. This router performs two generalized functions:
•It receives non-MPLS traffic, labels that traffic, and forwards it to another
label-switching interface.
2-6
•It receives labeled MPLS traffic, strips the label from the packets, and
forwards the traffic over a non-MPLS interface.
Note:
This guide describes how to configure the LER. For information about
how to start MPLS on the router, see “Starting MPLS” on page 3-30. For
information about how to customize the ATM router interface for MPLS, see
Chapter 11, “Customizing MPLS Confi gur ation.”
117374-C Rev 00
Supported Protocols
MPLS supports the following protocols:
•IP
•RIP
•BGP
•OSPF
For More Information
For more information about MPLS, refer to the following documents:
Black, D. Switching Solutions Expanding Networks. Draft document. Expected
publication date, 1999.
“LDP Specification,” Andersson, Doolan, Feldman, Fredette, Thomas, work in
progress, Internet Draft <draft-ietf-mpls-ldp-01.txt>. August, 1998.
Understanding MPLS
117374-C Rev 00
“Multiprotocol Label Switching Arc hitecture,” Callon, Rosen, Viswanathan, w ork
in progress, Internet Draft <draft-ietf-mpls-arch-01.txt>. March, 1998.
2-7
Configuring ATM and MPLS Services
Where to Go Next
Use the following table to determine where to go next.
If you want toGo to
Learn about ATM concepts.Chapter 1
Start ATM or MPLS.Chapter 3
Change default settings for ATM interface
parameters.
Change default settings for ATM signaling
parameters.
Change default settings for ATM PVC service record
and PVC parameters.
Change default settings for classical IP service
record parameters.
Change default settings for LAN Emulation client
service record parameters.
Change default settings for Multi-Protocol over ATM
server parameters.
Change default settings for the ATM router
redundancy parameter.
Change default settings for MPLS parameters.Chapter 11
Obtain information about Site Manager parameters.Appendix A
Monitor ATM using the BCC show commands.Appendix B
Configure NHRP for MPOA services.Appendix C
Chapter 4
Chapter 5
Chapter 6
Chapter 7
Chapter 8
Chapter 9
Chapter 10
2-8
117374-C Rev 00
Chapter 3
Starting ATM, ATM MPOA Ser ver,
ATM Router Redundancy, and MPLS
This chapter describes how to create a basic ATM or MPLS configuration by
specifying v alues for requir ed parameter s only and accep ting def ault v alues for all
other parameters. This chapter contains the following information:
TopicPage
Starting Configuration Tools3-2
Starting ATM Services3-2
Starting the MPOA Server3-15
Starting ATM Router Redundancy3-23
Deleting ATM from the Router3-29
Starting MPLS3-30
Deleting MPLS from the Interface3-39
Where to Go Next3-40
117374-C Rev 00
For overview information about ATM, see Chapter 1, “Understanding ATM,
MPOA, ATM Router Redundancy, and OAM.” For overview information about
MPLS, see Chapter 2, “Understanding M PLS.”
3-1
Configuring ATM and MPLS Services
Starting Configuration Tools
Before configur ing ATM, refer to the following us er guides for instructions on
how to start and use the Bay Networks configuration tool of your choice.
Configuration ToolUser Guide
Bay Command Console (BCC)
Site Manager
Starting ATM Services
You can use the BCC or Site Manager to start ATM on the router using default
values for all parameters.
Using the BCC
To start ATM on a router using the BCC:
1.
Add ATM to the configuration.
2.
Enable ATM signaling (if you plan to define either a LANE or classical IP
service record).
3.
Define an ATM service record.
4.
If you defined a PVC service record, add at least one virtual circuit to that
service record.
5.
Enable protocols on the ATM service record.
Using the Bay Command Console
Configuring and Managing Routers with
Site Manager
3-2
117374-C Rev 00
Starting ATM, ATM MPOA Server, ATM Router Redundancy, and MPLS
Adding ATM to the Configuration
To add ATM to the configuration, navigate to the top-level prompt and enter:
atm slot
<connector_number>
<slot_number>
[
module
<module_number>
]
connector
slot_number
module_number
is the number of the chassis slot containing the link module.
is a convention used for other routers within the System 5000
chassis. You need only enter a module number when configuring an ATM router
in the System 5000 chassis (that is, the Model 5782 Centillion Multiprotocol
Engine). The module number for the ATM router is always 1.
connector_number
The top-level prompt for BCC configuration on the System 5000
Note:
is the number of a connector on the link module.
platform is “stack.” However, the remainder of this guide uses the “box”
prompt associated with the BN platform in its examples.
For example, the following command adds ATM to the BN configurat ion on
slot 5, connector 1:
box#
atm slot 5 connector 1
atm/5/1#
Enabling Signaling (LANE and Classical IP Service Records Onl y)
To enable signaling on an ATM interface, navigate to the ATM prompt and enter:
signaling
117374-C Rev 00
For example, the following command enables signaling on slot 5, connector 1:
atm/5/1#
signaling/5/1#
signaling
3-3
Configuring ATM and MPLS Services
Defining an ATM Service Record
Using the BCC, you can define PVC, classical IP, and LANE service records.
Defining PVC Service
To define a PVC servi ce record, navigate to the ATM interface prompt and enter:
pvc-service
service_name
encapsulation_type
<service_name>
is a unique text string that you assign to the service record.
encapsulation
is the data encapsulation type that you wa nt the PVC service
<encapsulation_type>
record to use.
For exampl e, the foll o w ing comman d def i nes a PVC servi ce recor d with the na me
“boston” on ATM slot 5, connector 1:
atm/5/1#
pvc-service/boston#
Note:
pvc-service boston encapsulation llc-snap
You must add at least one PVC to a PVC ser vi ce r ecord. Go to “Addi ng
PVCs” on page 3-5 for instructions.
Defining Classical IP Service
To define a classical IP service record, navigate to the ATM interface prompt and
enter:
classical-ip-service
<service_name>
encapsulation
<encapsulation_type>
3-4
service_name
encapsulation_type
is a unique text string that you assign to the service record.
is the data encapsulation type that you wa nt the classical IP
service record to use.
For examp le, the follo wing command def ines a cl assical IP serv ice record with the
You must add at least one PVC to a PVC service record for that service record to
operate. A PVC is defined by its VPI/VCI pair.
A virtual path is a set of virtual channels between a common source and
destination. The virtual channels within a virtual path logically associate with a
common iden tifier. This identifier is called the virtual path identifier (VPI) and is
part of the cell header.
A virtual channel is a logical connection between two communicating ATM
entities. Each virtual channel can carry a different protocol or traffic type. The
virtual channel transports cells that have a common identifier. The identifier is
called the virtual channel identifier (VCI) and is part of the cell header.
117374-C Rev 00
Note:
ATM does not allow duplicate VPI/VCI pairs on the same physical
interface (tha t is, on the same li nk module ). However, dupli cate VPI /VCI pair s
can exist on different physical interfaces (that is, on different link modules).
3-5
Configuring ATM and MPLS Services
To add a PVC to a PVC service record, navigate to the ATM PVC service prompt
and enter:
pvc vpi
vpi_number
<vpi_number>
identifies the virtual path of the PVC. The header can contain a
<vci_number>
vci
maximum of 8 VPI bits for a UNI connection. This bit range allows for path
identifiers from 0 to 255.
vci_number
identifies the virtual channel of the PVC. The header can contain a
maximum of 16 VCI bits. This bit range allows for channel identifiers from 32 to
65535.
Note:
Following the recommendation of the ATM Forum, virtual channel
identifiers from 0 to 31 are reserved for signaling and added functionality.
For example, the following command adds PVC 0/130 to the configuration on
PVC service record boston:
pvc-service/boston# pvc vpi 0 vci 130
pvc/0/130#
Adding Protocols to an ATM Service Record
The BCC currently supports only IP and IPX configuration over ATM. Table 3-1
indicates which service records support the IP and IPX protocols.
3-6
Table 3-1.Service Record Protocol Support
Service Record TypeIPIPX
pvc-service
classical-ip-service
lec-service
99
9
99
117374-C Rev 00
Starting ATM, ATM MPOA Server, ATM Router Redundancy, and MPLS
Adding IP
To add IP to a service record, navigate to the ATM service record prom pt (for
classical IP or LEC service records), or the ATM PVC to which you want to add
IP, and enter:
ip address
address
and
<address>
mask
are a v alid IP address an d its a ssociate d subnet mask, e xpressed
mask
<mask>
in either dotted-decimal notation or in bit notation.
For example, the following command configures an IP address of 2.2.2.2 and a
subnet mask of 255.255.255.0 on ATM PVC 0/130:
pvc/0/130# ip address 2.2.2.2 mask 255.255.255.0
ip/2.2.2.2/255.255.255.0#
Adding IPX
To add IPX to a service record, navigate to the ATM service record prompt (for
LEC service records), or the ATM PVC to which you want to add IPX, and enter:
ipx address
address
<address>
is a valid IPX address expressed in either dotted-decimal notation or in
bit notation.
For example, the following command configures an IPX address of 0000001a on
ATM PVC 0/130:
pvc/0/130# ipx address 0000001a
ipx/0000001a#
117374-C Rev 00
3-7
Configuring ATM and MPLS Services
Using Site Manager
To start ATM on a router using Site Manager:
1.
Create an ATM circuit.
2.
Define an ATM service record.
3.
Enable protocols on the ATM service record.
4.
If you defined a PVC service record, add at least one virtual circuit to that
service record.
Creating an ATM Circuit
To create an ATM circuit, complete the following tasks:
You do thisSystem responds
Site Manager Procedure
1. In the Configuration Manager window,
click on an ATM link module interface
(
).
ATM1
2. Click on OK to accept the default circuit
name.
3. Click on
4. Click on OK to accept the default settings. The Edit ATM Connector window opens.
5. Go to the next section to define a service
record on the circuit.
Note:
Accepting the defaults in the Initial ATM Signaling Config window
.The Initial ATM Signaling Config window
ATM
The Add Circuit window opens.
The Select Connection Type window
opens.
opens.
enables signaling on the interface. If you are running only PVCs on the
interface, you do not need signaling enabled.
3-8
117374-C Rev 00
Starting ATM, ATM MPOA Server, ATM Router Redundancy, and MPLS
Defining an ATM Service Record
The Configuration Manager allows you to define service records for a specific
data encapsulation type. Depending on the data encapsulation type you choose,
the Configuration Manager also allows you to select a PVC or SVC connection
type for that service record.
Caution:
You cannot edit the Data Encapsulation Type or Virtual Connection
Type parameters after you assign them to a service record. However, you can
edit the Data Encapsulation Type for individual PVCs.
Table 3-2
identifies which data encapsulation types you can apply to permanent
and switched virtual circuits. For an explanation of the different data
encapsulation types, and rules for assigning data encapsulati on, see Chapter 1,
“Understanding ATM, MPOA, ATM Router Redundancy, and OAM.”
Table 3-2.Valid Data Encapsulation Types for PVCs and SVCs
Data Encapsulation TypePermanent Virtual CircuitSwitched Virtual Circuit
LANE
LLC/SNAP
NLPID
NULL
99
9
99
9
117374-C Rev 00
3-9
Configuring ATM and MPLS Services
Adding a Service Record for PVCs
Note: The values for some parameters are contingent on the va lues of others.
If you change one parameter, you must press the Enter or Tab key to advance
from one parameter cell to another. Pressing either key acknowledges any
changes to a parameter. If you neglect this step, the Configuration Manager
may not provide the appropriate option for other parameters.
To add a service record for PVCs, complete the following tasks:
You do thisSystem responds
Site Manager Procedure
1. In the Configuration Manager window,
click on an ATM link module interface
(
).
ATM1
2. Click on
3. Click on
4. Click on
5. Set the
parameter to
Click on
description on page A-16.
6. Press the
the Virtual Con nection Type parameter.
7. Change the
parameter to
the parameter description on page A-16.
8. Click on OK.The Select Protocols window opens.
9. Go to “Enabling Protocols on an ATM
Service Record” on page 3-12.
.The Edit ATM Connector window opens.
ATM
Service Attributes
.The ATM Service Record Parameters
Add
Data Encapsulation Type
LLC/SNAP, NLPID
or see the parameter
Help
or
Enter
Tab
Virtual Connection Type
. Click on
PVC
.The ATM Se rvice Records List w indow
, or
NULL
key to advance to
or see
Help
The Select Connection Type window
opens.
opens.
window opens.
.
3-10
117374-C Rev 00
Starting ATM, ATM MPOA Server, ATM Router Redundancy, and MPLS
Adding a Service Record for Classical IP
To add a service record for classical IP, complete the following tasks:
Site Manager Procedure
You do thisSystem responds
1. In the Configuration Manager window,
click on an ATM link module interface
(
).
ATM1
2. Click on
3. Click on
4. Click on
5. Set the
parameter to
on
Help
on page A-16.
6. Click on OK.The Select Protocols window opens.
7. Go to “Enabling Protocols on an ATM
Service Record” on page 3-12.
.The Edit ATM Connector window opens.
ATM
Service Attributes
.The ATM Service Record Parameters
Add
Data Encapsulation Type
LLC/SNAP
or see the parameter description
.The ATM Se rvice Records List w indow
or
NULL
. Click
The Select Connection Type window
opens.
opens.
window opens.
Adding a Service Record for LANE
To ad d a service record for LANE, complete the following tasks:
Site Manager Procedure
117374-C Rev 00
You do thisSystem responds
1. In the Configuration Manager window,
click on an ATM link module interface
(
).
ATM1
2. Click on
3. Click on
4. Click on
.The Edit ATM Connector window opens.
ATM
Service Attributes
.The ATM Service Record Parameters
Add
.The ATM Se rvice Records List w indow
The Select Connection Type window
opens.
opens.
window opens.
(continued)
3-11
Configuring ATM and MPLS Services
Site Manager Procedure
You do thisSystem responds
5. Click on OK.The Select Protocols window opens.
6. Go to the next section, “Enab ling Protocols
on an ATM Service Record.”
(continued)
Enabling Protocols on an ATM Ser vice Record
Depending on the data encapsulation type and virtual connection type (PVC or
SVC) that you choose for the service reco rd, the ro ute r supports various protocols.
You can select and configure protocols immediately after you create a service
record, or you can exit the Select Protocols window and add protocols at a later
time.
Adding Protocols Immediately After Creating a Service Record
To add protocols to a service record immediately after creating it, complete the
following tasks:
Site Manager Procedure
3-12
You do thisSystem responds
1. In the Select Protocols window, click on
the protocols you want to add.
2. Click on OK.For each protocol you select, the
A check mark appears in the bo x f or each
protocol that you select.
Configuration Manager displays a
protocol-specific window prompting you
for required information.
Click on
the appropriate protocol-specific guide.
for any parameter, or see
Help
117374-C Rev 00
Starting ATM, ATM MPOA Server, ATM Router Redundancy, and MPLS
Adding Protocols to an Exis ting Service Record
To add protocols to an existing service record, complete the following tasks:
Site Manager Procedure
You do thisSystem responds
1. In the Configuration Manager window,
click on an ATM link module interface
(
).
ATM1
2. Click on
3. Click on
4. Click on the servi c e record to which you
want to add protocols.
5. Click on
6. Choose
7. Click on the protocols that you want to
add.
8. Click on OK.For each protocol you select, the
.The Edit ATM Connector window opens.
ATM
Service Attributes
Protocols
Add/Delete
.The Protocols menu opens.
.The Select Protocols window opens.
.The ATM Se rvice Records List w indow
The Select Connection Type window
opens.
opens.
The Protocols menu selection becomes
active.
A check mark appears in the box for each
protocol that you select.
Configuration Manager displays a
protocol-specific window prompting you
for required information.
Click on
the appropriate protocol-specific guide.
for any parameter, or see
Help
117374-C Rev 00
3-13
Configuring ATM and MPLS Services
Adding PVCs
You must add at least one virtual circuit to a PVC service record for that service
record to operate. When you finish configuring the protocols for a PVC service
record, the ATM Virtual Channel Li nk window opens.
To add a PVC to a PVC service record, complete the following tasks:
You do thisSystem responds
Site Manager Procedure
1. In the Configuration Manager window,
click on an ATM link module interface
(
).
ATM1
2. Click on
3. Click on
4. Click on the PVC service record to which
you want to add a virtual circuit.
5. Click on
6. Click on
7. Set the
or see the parameter description on
Help
page A-21.
8. Set the
or see the parameter description on
Help
page A-22.
9. Click on OK.You return to the ATM Virtual Channel
10. Click on
11. Click on
12. Click on
13. Click on
.The Edit ATM Connector window opens.
ATM
Service Attributes
.The ATM Virtual Channel Link window
PVC
.The ATM Virtual Channel Link
Add
VPI Number
VCI Number
.You return to the ATM Service Records
Done
.You return to the Edit ATM Connector
Done
.You return to the Select Connection Type
Done
.You return to the Configuration Manager
Done
.The ATM Se rvice Records List w indow
parameter. Click on
parameter. Click on
The Select Connection Type window
opens.
opens.
opens.
Parameters window opens.
Link window.
List window.
window.
window.
window.
3-14
117374-C Rev 00
Loading...
+ hidden pages
You need points to download manuals.
1 point = 1 manual.
You can buy points or you can get point for every manual you upload.