This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Chapter 1 The Story Of Bridging Challenge-and-response is a formula describing the free play of forces that provokes new departures in individual and social life. An effective challenge stimulates men to creative action . . . Arnold Toynbee This set of notes is organized around a description of the history of bridges. This chapter also describes some of the stimuli that lead to innovation, and introduces some of the people that produced such inventions. Arnold Toynbee [TC72] describes history using a challenge-response theory in which civi- lizations either grow or fail in response to a series of challenges. Similarly, the history of bridges can be described as a series of three challenges that are described in the three sections of this chapter: Ethernets under fire (Section 1.1), wire speed Ethernet forwarding (Section 1.2), and scaling to higher speeds (Section 1.3. The responses to these challenges led to what is now known as 802.1 Spanning Tree bridges [IEE97]. 1.1 Challenge 1: Ethernet Under Fire The first challenge arose in the late 1980s. Ethernet, invented in the 1970s as a low-cost, high bandwidth interconnect for Personal Computers, was attacked as behaving poorly at large loads and being incapable of spanning large distances. Recall that if two or more nodes on an Ethernet send data at the same time, a collision occurs on the shared wire. All senders then compute a random retranmission time and retry, where the randomization is chosen to minimize the probability of further collisions. 1 Theoretical analyses (e.g., [Bux80]) claimed that as the utilization of an Ethernet grew, the effective throughput of the Ethernet dropped to zero as the entire bandwidth was wasted on retransmissions. A second charge against Ethernet was its small distance limit of 1.5 Km, much smaller than the limits imposed by say the IBM Token Ring. While the limited bandwidth charge turned to be false in practice ([BMK88]), it remained a potent marketing bullet for a long time. The second limited distance charge was, and remains, a true limitation of a single Ethernet. In this embattled position around 1980, network marketing people at Digital Equipment Corporation (DEC) pleaded with their technical folks for a technical riposte to these attacks. Could not their bright engineers find a clever way to extend a single Ethernet such that it could become a longer Ethernet with a larger effective bandwidth? First, it was necessary to discard some unworkable alterntives. Physical layer bit repeaters were unworkable because they did not avoid the distance and bandwidth limits of ordinary Ethernets. Extending an Ethernet using a router did, in theory, solve both problems but introduced two other problems. First, in those stone ages, routers were extemely slow and could hardly keep up with the speed of the Ethernet....
View Full Document
This note was uploaded on 10/10/2010 for the course CSE 123 taught by Professor Varghese,g during the Spring '08 term at UCSD.
- Spring '08
- Computer Networks