Handbook of Local Area Networks, 1998 Edition:LAN Interconnectivity Basics
Click Here!
Search the site:
ITLibrary
ITKnowledge
EXPERT SEARCH
Programming Languages
Databases
Security
Web Services
Network Services
Middleware
Components
Operating Systems
User Interfaces
Groupware & Collaboration
Content Management
Productivity Applications
Hardware
Fun & Games
EarthWeb sites
Crossnodes
Datamation
Developer.com
DICE
EarthWeb.com
EarthWeb Direct
ERP Hub
Gamelan
GoCertify.com
HTMLGoodies
Intranet Journal
IT Knowledge
IT Library
JavaGoodies
JARS
JavaScripts.com
open source IT
RoadCoders
Y2K Info
Previous
Table of Contents
Next
REPEATERS
LAN repeaters were initially developed to overcome the distance restrictions on single LAN segments. Ethernet has a 500-m length restriction for baseband and a 100-m restriction for a 10Base-T station. Token Ring has a distance limitation of 100 m per ring segment. If internetworking devices did not exist these distance limitations would preclude the use of Ethernet or Token Ring in larger multistory office buildings or factories.
Repeaters are the oldest technology in use today for connecting multiple LAN segments. A repeater is primarily a hardware device, equipped with a microprocessor and some code in ROM. Repeaters were originally packaged as standalone units, but now they are typically integrated into LAN concentrators. The basic function of a LAN repeater is to reproduce the exact bit stream detected on one segment on another LAN segment. One way to describe a repeaters operation is, whatever it gets it throws over the wall.
Repeaters can be local, remote, and multisegment. The local repeater has two LAN ports that connect directly to two LAN segments, which allows processors on either side of the repeater to communicate.
Local Repeaters
Local repeaters (shown in Exhibit 3-3-1) used alone are fine for connecting LAN segments on different floors, but not for connecting LAN segments in different buildings. Repeaters can be used in pairs to support remote repeating. Each repeater is connected to its local LAN segment through a transceiver and they are connected to each other through local twisted pair or fiber-optic cable, dial-access phone lines and modems, or leased lines and modems or DSU/CSUs. The local twisted pair or fiber connections are typically used in campus environments to allow LAN-to-LAN communication between different buildings and make use of customer owned cable or fiber.
Exhibit 3-3-1. Local Repeater Diagram
Wide area public carrier analog or digital services are used to connect LANs that are remote from each other. Analog point-to-point links are established through dial or leased line modems and operate from 2,400 to 56K bps, the speed depending on line type, modulation techniques, and compression algorithms. Digital point-to-point lines use DSU/CSUs and offer data rates of 144K, 1.5M (T1) and 45M (T3)bps.
Fractional T1
Fractional T1 options on T1 circuits introduce another option and a source of confusion, because the carrier implies that a digital circuit can be ordered in bandwidth blocks of N × 56K bps; N = 5 would yield a 280K-bps circuit.
To understand how fractional T1 circuits are configured, the anatomy of a digital circuit must e further explained. A carrier circuit is actually comprised of three components,the near end local exchange carrier (LEC) portion, the interexchange carrier (IXC) portion, and the far end LEC portion (as shown in Exhibit 3-3-2).
Exhibit 3-3-2. The Carrier and Customer Components of a Communication Circuit
IXCs (e.g., AT&T, MCI, Sprint) offer tariffed fractional T1 services that provide N × 56K-bps channels between the LECs. However, because the LEC portion of the T1 circuit must be installed on a point-to-point basis between the LEC central office and the customer premise, the T1 circuit cannot be shared or fractionalized. Therefore, a fractional T1 connection is actually allocated as a full T1 on both ends of the LEC portion of the circuit; the fractional T1 bandwidth is restricted to the IXC portion of the circuit.
Remote Repeaters
Remote repeaters (shown in Exhibit 3-3-3) can use any of the WAN transmission carrier services for LAN-to-LAN connection. The limitation of the remote repeater technique is the WAN links speeds: on the order of 9.6K to 1.5M bps, as opposed to 10M bps for a typical LAN. LAN transmission media differ from WAN transmission media, but the repeater still performs the same bit regeneration and signal conditioning.
Exhibit 3-3-3. Remote Repeater Diagram
Multiport Repeaters
The last type of repeater to be discussed is the multiport repeater (shown in Exhibit 3-3-4). The multiport repeater is actually a special adaptation of the local repeater, allowing many LAN segments to be connected together through a single device. Multiport repeaters were common in thinnet and thicknet coaxial cable LANs, but have declined in popularity with the advent of 10Base-T concentrators.
Exhibit 3-3-4. Multiport Repeater Diagram
All repeaters function at the physical layer. Their primary function is to faithfully reproduce the original bit pattern generated on the original LAN segment on all other attached LAN or serial connections. This regeneration function provides electrical isolation between segments and is completely independent of the higher layer protocols. Because repeaters do not have to process higher layer protocol information, data forwarding between segments occurs in real time.
The disadvantages of an internetworking device that processes physical layer bit patterns are that connection between dissimilar LANs (e.g., Token Ring and Ethernet) is not supported, and delay sensitive higher level protocols may experience time-outs. And because repeaters do not process LAN link layer protocols, repeated segments function as one large segment, imposing a limitation on the number of attached devices. But the primary problem is that the round-trip propagation delay of the repeated signal can exceed the link layer IEEE 802.3 and 802.5 specification. The reader should check with the respective concentrator vendor for this specific limitation.
FRAME RELAY SWITCHES
Frame relay is a protocol that was developed to handle LAN-to-LAN traffic more efficiently than point-to-point WAN links. Frame relay is more than a physical layer protocol because it has a partial link layer implementation. When comparing internetworking devices, a frame relay switch has features in common with repeaters and bridges.
The development of frame relay reflects the evolution of customer applications. Traditional mainframe data networks were based on a hierarchical model and constructed on a star topology. The topology was typically defined by point-to-point digital or analog connections set up in a star or tree structure. Fixed bandwidth lines were used because the polling activities of the central processor infrastructure created deterministic traffic. That model, while still useful, is being augmented, and in some instances replaced with a model that includes the distributed processing requirements of LAN-based intelligent workstations.
Previous
Table of Contents
Next
Use of this site is subject certain Terms & Conditions.
Copyright (c) 1996-1999 EarthWeb, Inc.. All rights reserved. Reproduction in whole or in part in any form or medium without express written permission of EarthWeb is prohibited.
Please read our privacy policy for details.
Wyszukiwarka
Podobne podstrony:
README (285)285,2,artykuldemo cgi 289220 289285 (2)285?1303 monter elektroniki samochodowej285 aSHSpec 289 6307C24 ARC Breaks and the Comm Cycle289 291287 289więcej podobnych podstron