RE: [squid-users] Calculation of throughput...

From: Boosten, Peter <[email protected]>
Date: Thu, 15 Aug 2002 13:51:32 +0200

# For this kind of calculations it is better to use median than
# average...
# average can easily get substantially biased in both
# directions by a few
# "odd" requests.
#
# Also, the average response time depends a fair bit on your setting of
# half_closed_clients..
#
# For example, 0.1% of stalled requests with
# "half_closed_clients on" will
# add almost a second to the average response time.
#
# If you cannot do median, then atleast add a filter to your
# calculations to
# ignore requests with a very low or high average thruput as
# these are are
# mainly slow because of other criterias than your link or proxy.

Henrik,

Thanks for your answer.

The problem is, that multiple users complain about performance while
downloading large files. I'm trying to find the cause. In a measurement, a
download of a 7,45 Mb file took about 17 minutes, which is way to long for
these downloads over the mentioned links. (about 7.5Kb/sec)
I'm looking for the "offender" :-)

Our squid (2.4stable2) has Smartfilter as its parent and this one uses a
trendmicro virusscanner as its parent, so you see we have a lot of "possible
suspects" in cascade.

Peter

Disclaimer
1. This e-mail is for the intended recipient only. If you have received it
by mistake please let us know by reply and then delete it from your system;
access, disclosure, copying, distribution or reliance on any of it by anyone
else is prohibited.

2. If you as intended recipient have received this e-mail incorrectly,
please notify the sender (via e-mail) immediately. This e-mail is
confidential and may be legally privileged. DSM does not guarantee that the
information sent and/or received by or with this e-mail is correct and does
not accept any liability for damages related thereto.
Received on Thu Aug 15 2002 - 05:52:05 MDT

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 17:09:38 MST