flâneur

The Early Years of Academic Computing

cs.cornell.edu · 1,694 words · saved by 1 readers

Mainframe computers were expensive. The central computer that Dartmouth bought in 1975 cost more than four million dollars, a great deal of money for a small university. Moreover, these computers had huge economies of scale. Several years later, when Cornell studied the options for replacing its mainframe computer, an IBM 370/65 cost twice as much as a smaller IBM 370/55, yet was four times as powerful. With such economies of scale, every organization followed the same strategy: buy the largest computer you could afford and find ways to maximize the amount of work done on it. Computer hardware improved rapidly during the 1960s, both in performance and reliability, but processor cycles were a scarce resource not to be wasted. Computers of the previous generation had run one program at a time and the central processor might spend much of its time idling. For example, a program to process a tray of punched cards would spend most of its time waiting for the next card to be read. By the mid

The Early Years of Academic Computing Early Timesharing Multiprograming, multitasking, and timesharing Mainframe computers were expensive. The central computer that Dartmouth bought in 1975 cost more than four million dollars, a great deal of money for a small university. Moreover, these computers had huge economies of scale. Several years later, when Cornell studied the options for replacing its mainframe computer, an IBM 370/65 cost twice as much as a smaller IBM 370/55, yet was four times as powerful. With such economies of scale, every organization followed the same strategy: buy the large

saved by

related reading