Skip to main content

RFC 32 - Some Thoughts on SRI's Proposed Real Time Clock

  • Status: Unknown
  • Published: February 5, 1970
  • Stream: Legacy
  • Errata: No Errata

Document Information​

  • RFC Number: 32
  • Title: Some Thoughts on SRI's Proposed Real Time Clock
  • Authors: Jerry Cole (UCLA)
  • Date: February 5, 1970
  • Category: Unknown

Abstract​

This memo comments on the real-time clock proposed by SRI in RFC 28 and discussed in RFC 29, and on how such a clock could be used for network measurements. It argues that adding a clock to one or more network HOSTs would allow user-oriented message delay measurements that include the internal HOST delays the present facilities ignore, and that extending the "Trace" capability to cover those delays would be an appropriate use for such a clock, although measuring an absolute HOST-to-HOST delay would require the two clocks to be synchronized. It suggests a resolution of about 1 msec., an accuracy of about 1 part in 10E7 and a range of about 24 hours, notes that a 1 msec. resolution would require the clock to be entirely hardware and may call for rather complicated interface circuitry, and reports that UCLA already has two clocks on the Sigma 7, one of which has a resolution of about 2 msec. but lacks the long term accuracy needed for absolute measurements.

Table of Contents​