[113657] in North American Network Operators' Group
Re: downloading speed
daemon@ATHENA.MIT.EDU (Jay Hennigan)
Fri Apr 17 19:59:41 2009
Date: Fri, 17 Apr 2009 16:59:34 -0700
From: Jay Hennigan <jay@west.net>
In-Reply-To: <a7ade0880904171521t24f1099yd9921ef3c75e4d1d@mail.gmail.com>
Cc: nanog@merit.edu
Errors-To: nanog-bounces@nanog.org
chandrashakher pawar wrote:
> No errors on the interface.
> none of our customer on this router has complait us this issue
> i have changed this to "negotiation auto" as suggested by one of our member.
> tommorow customer will test again and reply.
> round-trip-time is good, no bacbone chocked.
> Unit will not make bit differnce as: The customer tried troubleshooting the
> issue after connecting laptop directily to the 100 mbps link.
> In that case also the result was same.
Note that your screenshot displays bytes, not bits. So it will display
one-eighth the download speed measured in bits.
Check the TCP tuning on the downloading PC. The fact that multiple
windows achieve a higher aggregate speed points to this. Use the Google
link I supplied earlier, also search "Bandwidth-delay product". Are
both ends of the link a substantial geographic distance (several miles)
apart?
Note that the adjustments for TCP tuning are to the TCP stack on the
machine doing the download, not the network gear.
--
Jay Hennigan - CCIE #7880 - Network Engineering - jay@impulse.net
Impulse Internet Service - http://www.impulse.net/
Your local telephone and internet company - 805 884-6323 - WB6RDV
