Experiment to measure Internet communication performance

[Tom Fitzgerald]

golding at saturn.ucsc.edu (Richard A. Golding) writes:
> We are asking that people at several sites throughout the Internet run
> a simple daemon for two weeks, from Thursday, 2 May 1991 through
> Thursday, 16 May 1991.  We will be sending this daemon a UDP packet a
> few times each day.  The daemon simply echoes this packet back to the
> sender, so we can time how long it takes.

Why aren't you simply using ICMP ping packets?

All your statistics are going to be distorted by the time it takes to
page your daemon in from disk, and schedule it, every time a packet
arrives.

---
Tom Fitzgerald   Wang Labs        fitz at wang.com
1-508-967-5278   Lowell MA, USA   ...!uunet!wang!fitz