Organization: China Interactive Publishing Network (http://www.china-pub.com/)
RFC Document Chinese Translation Program (http://www.china-pub.com/compters/emook/aboutemook.htm)
E-mail: Ouyang@china-pub.com
Translator: Anpengwang Anpengwang@263.net)
Translation time: 2001-4-26
Copyright: This Chinese translation copyright belongs to China Interactive Publishing Network. Can be used for non-commercial use free reprint, but must
Keep the translation and copyright information of this document.
Network Working Group Jerry Cole, ucla
REQUEST for Comments: 32 February 5, 1970
Some ideas about the real-time clock proposed by SRI
- Reply to NWG / RFC 28 and 29
(RFC 32 Some Thoughts on Sri's proposed Real Time Clock)
In order to increase the clock for one or more hosts on the network, it seems to be very meaningful, so it can be from the perspective of the user.
The delay time of metric messages. Our current network measurement means does not include hysteresis in the host, in these hysteresis
Among the total delay time in the messaging of the host may be considerable part of the latency. Use this clock to
The hysteresis in the host include coming in, we may find that this "record" ability is an appropriate mechanism. real
Now this use requires the source and destination of the message, but does not require two clocks, no need.
To sync. Other tests, such as the measurement from host A to host B absolutely complete message hysteresis, need two
Clock synchronization.
For SRI real-time clock, reasonable specifications may include 1 millisecond resolution, accuracy reaches 10m
One (10E7, so two clocks can maintain a reasonable relative accuracy within a few hours), using 24 hours
system. Medium-priced quartz clocks can easily meet this requirement.
What kind of mechanism is adopted by the host to read the clock. 1 millisecond resolution may need to completely solidize the clock
Chemical (rather than letting a certain place in the core) increased when receiving each clock pulse), this clock may require quite complex
Plane circuit.
At the University of California, the Los Angeles, UCLA, we have two clocks on Sigma 7, one of the distinctions
The rate is 2 milliseconds, which may be useful for the internal metrics of the host, but this accuracy is not enough for absolute measurements mentioned above.
of.
This document is entered into the form of RFC documentation by Richard Ames in January 1998]
RFC 32 Some ideas about the real-time clock proposed by SRI
RFC 32 Some Thought on Sri's proposed Real Time Clock
1
RFC Document Chinese Translation Program
