hat does emerge from studies of both local and wide- area data traffic over the years is that computer traffic is bursty. It does not flow in constant streams; rather, "the level of traffic varies widely over almost any measurement time scale" (Fowler and Leland, 1991). Dynamic bandwidth allocations are therefore preferred for data traffic, since static allocations waste unused resources and limit the flexibility to absorb bursts of traffic. This requirement addresses traffic patterns, but it says nothing about the absolute level of load. How can we evaluate a system when we never know how much capacity is enough? In the personal computing industry, this problem is solved by defining "enough" to be "however much I can afford today," and relying on continuous price-performance improvements in digital technology to increase that level in the near future. Since both of the infrastructure upgrade options rely heavily on digital technology, another criteria for evaluation is the extent to which rapidly advancing technology can be immediately reflected in improved service offerings. Cable networks satisfy these evaluation criteria more effectively than telephone networks because: Coaxial cable is a higher quality transmission medium than twisted copper wire pairs of the same length. Therefore, fewer wires, and consequently fewer pieces of associated equipment, need to be installed and maintained to provide the same level of aggregate bandwidth to a neighborhood. The result should be cost savings and easier upgrades. Cable's shared bandwidth approach is more flexible at allocating any particular level of bandwidth among a group of subscribers. Since it does not need to rely as much on forecasts of which subscribers will sign up for the service, the cable architecture can adapt more readily to the actual demand that materializes. Telephony's dedication of bandwidth to individual customers limits the peak (i.e. burst) data rate that can be pr...
