2d187bf7   
   From: rick.jones2@hp.com   
      
   Le Chaud Lapin  wrote:   
   > I am looking for sample data on unidirectional inter-node delay   
   > using UDP.   
      
   Over what distance?   
      
   > Ideally, someone would have set up highly-synchronized clocks   
   > between source and target to measure the actual uni-directional   
   > delay. If I may continue to be greedy, s/he would have done this   
   > independently in both directions: source-to-target and   
   > target-to-source, with a sufficient diversity of intermediate nodes   
   > between source and target for a large swath of Internet. [Data for   
   > nodes confined within same LAN would not be useful.]   
      
   At one point I toyed with the idea of hacking netperf for one-way   
   latency measurements for two systems on a LAN, and even went as far as   
   exchanging posts with the NTP gurus in the ntp newsgroup   
   (comp.protocols.time.ntp? not sure).   
      
   In the course of that discussion, some papers about WAN one-way   
   latency measurements were brought-up, using I think, "PPS" (Pulse Per   
   Second) GPS devices at the endpoints - or perhaps they just synced to   
   > stratum 1 time servers and it was the LAN idea that brought PPS GPS   
   > into play.  Anyhow, whatever they used, it got the time synchronized   
   "well enough" for the WAN.  I don't recall that it would have been   
   synchronized "well enough" for the LAN, but I abandoned the idea after   
   taking a Garmin GPS 12 (perhaps not the best choice) into the machine   
   room where my systems would reside and finding no GSP signal   
   whatsoever.   
      
   rick jones   
   --   
   oxymoron n, commuter in a gas-guzzling luxury SUV with an American flag   
   these opinions are mine, all mine; HP might not want them anyway... :)   
   feel free to post, OR email to rick.jones2 in hp.com but NOT BOTH...   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)