Movatterモバイル変換

Network Working Group                                   Jerry Cole, UCLARequest for Comments: 32                                February 5, 1970SOME THOUGHTS ON SRI'S PROPOSED REAL TIME CLOCKRe: NWG/RFC's #28 and 29.The addition of a clock in one or more of the network HOST's seems to bevery desireable since it (or they) would allow user-oriented messagedelay measurements.  Our present network measurement facilities do notinclude any internal HOST delays, and these delays may be an appreciableportion of the total delay encountered by a HOST-to-HOST messagetransmission.  We may find that an extension of our "Trace" capabilitiesto include internal HOST delays would be an appropriate mechanism forutilizing such a clock.  Such usage would require a clock at both thesource and the destination of the message, although such clocks wouldnot have to be particularly accurate nor synchronized.  Other tests,such as the absolute overall message delay from HOST A to HOST B wouldrequire synchronization of the two clocks.A reasonable specification for the SRI real-time clock would seem toinclude a resolution of about 1 msec., an accuracy of about 1 part in10E7 (so that two such clocks could maintain reasonable relativeaccuracies over periods of many hours), and a range of about 24 hours.A crystal controlled clock should easily meet these requirements at amoderate cost.The choice of the mechanism by which the HOST can read the clock appearsto be of concern also.  The 1 msec. resolution may require that theclock be entirely hardware (as opposed to a core location which would beincremented at each clock pulse), and therefore the clock may requiresome rather compli- cated interface circuitry.At UCLA, we presently have two clocks on the Sigma 7, and one of thesehas a resolution of about 2 msec. which might be usable for someinternal HOST measurements.  However, it does not have the long termaccuracy for the absolute measurements mentioned above.       [ This RFC was put into machine readable form for entry ]         [ into the online RFC archives by Richard Ames 1/98 ]                                                                [Page 1]