🚀 go-pugleaf

Hi Richard,

At 07:52 PM 8/25/97 -0400, you wrote:
>At 06:41 PM 8/25/97 -0600, you wrote:
>>
>>So are you saying that 10-20 pings once per hour might be better
>>statistically than 1-2 pings every 5 or 6 minutes ?
>
>God god yes. Ocasionally you can do 100 pings and get results that
>are all over the place, and can vary by 1 - 2 orders or magnitude.

Just so I make sure we both understand what I'm trying to say:

Let's say I budget a maximum of 100 pings per hour per host.
I do this at least partly to avoid any criticism that I might
be adding to internet congestion.

Among other schemes I could do:

100 pings every hour on the hour consecutively, or
10 pings every 6 minutes on the sixth minute consecutively, or
1 ping every 36 seconds on the 36th second consecutively.

All of these add up to 100 total pings per hour per host. My preference
would be to take the last option and send a single ping every 36 seconds.
This gives higher time resolution and when averaged over longer periods
should be smoother.

It helps to diminish the anomolous effects, for example, of major
congestion which  could occur in the 5 seconds every hour during
which the 100 pings were sent.

Do you disagree with this ? I suspect that I may not have communicated my
example clearly enough to you.


I've considered multiple simultaneous or consecutive pings before but
mostly to attempt to determine if a host is really down or if a few
packets just got lost.


Regards,
Mike.

-------------------==== Posted via Deja News ====-----------------------
      http://www.dejanews.com/     Search, Read, Post to Usenet