Monday, April 27, 2009

Delay Line Memory

Steven Davis saw the Friday Link on racetrack memory and sent in a link to the Wikipedia entry on delay line memory, which is conceptually almost identical. Delay line memory was invented in the 1940's, though. Here's an excerpt from the Wikipedia entry:
Delay line memory was a form of computer memory used on some of the earliest digital computers. Like many modern forms of electronic computer memory, delay line memory was a refreshable memory, but as opposed to modern random access memory, delay line memory was serial access. In the earliest forms of delay line memory, information introduced to the memory in the form of electric pulses was transduced into mechanical waves that propagated relatively slowly through a medium, such as a cylinder filled with a liquid like mercury, or a magnetorestrictive coil, or a piezoelectric crystal. The propagation medium could support the propagation of hundreds or thousands of pulses at any one time. Upon reaching the other end of the propagation medium, the waves were re-transduced into electric pulses, amplified, shaped, and reintroduced to the propagation medium at the beginning, thus refreshing the memory. Accessing a desired part of the propagation medium's memory contents required waiting for the pulses of interest to reach the end of the medium, a wait typically on the order of microseconds. Use of a delay line for a computer memory was invented by J. Presper Eckert in the mid-1940s for use in computers such as the EDVAC and the UNIVAC I.

The entire Wikipedia is fascinating.

Mike Harris also sent me a link to a video showing an actual delay line memory unit (it's freaking huge, of course), and it's amazing as well: 1951 UNIVAC I - Mercury delay line Memory for computers.

Site Meter