From: sirus MOGHADASIAN (cyrus.mgh@gmail.com)
Date: Fri Aug 17 2007 - 22:22:38 ART
Thanks for your detail answers ,I examine with iperf and achieve 21Mbps for
my wireless connection instead of 327kbps by increasing window size from
8.0 KB to 100KB.
I should inspect other aspects to see what other factors are in!!!
By the I have a difficulty with changing window size on windows although I
change TcpWindowSize and
GlobalMaxTcpWindowSize registry but windows does not change it's actuall
window size and also I chck my interfaces there is no WindowSize vlue define
for them in registry , if anybody know what's going on plz inform me.
Sirus
Thanks
Thanks
On 8/17/07, ROBERT LEUGERS <RLeugers@familydollar.com> wrote:
>
> Because the TCP stack on your machine has throughput limitations based
> on a single thread? Your disk write speed is bad? Full buffers (tcp
> frozen windows) or packet drops (requiring retransmit requests, data
> retransmission, acknowledgements, and back-off with slow start from a
> fraction of the previous rate) on the destination might do it also?
> Because your testing methodology/program can't push the rate based on
> the way the application under test behaves (use a UDP or TCP-based test
> instead of ICMP)?
>
> These are just a few possibilities. Very hard to tell. BTW -- this
> doesn't eliminate the possibility the switch is another source of the
> problem.
>
> I can tell you from experience that some applications misbehave over
> very good networks because of the way they actually operate over the
> network. Database queries, for example, don't always behave very well
> over high-delay WANs if they are searching through a huge table
> line-by-line until they get the answer(s). The same database works
> blindingly fast if the query is executed locally on the database.
>
> Another thing is that Ping does not really do a throughput test. It
> tests request/response and rough two-way delay based on ICMP echo and
> echo-reply, but not necessarily throughput as no data is really being
> transmitted.
>
> Multiple concurrent ftp's from different windows might push the rate up.
>
> A traffic generator might do it also. There are free ones about.
>
> Use one of these methods and the throughput might get better.
>
> Rob Leugers
>
> -----Original Message-----
> From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
> Sean.Zimmerman@clubcorp.com
> Sent: Friday, August 17, 2007 10:49 AM
> To: sirus MOGHADASIAN
> Cc: Cisco certification; nobody@groupstudy.com
> Subject: Re: simple question: why I get throughput of maximum 1Mbps in
> network that consists of 2 hosts and 1 switch!!?
>
> If I had to guess, I'd say duplex mismatch. Is it a 10/100 or
> 10/100/1000
> switch? How about the NICs?
>
>
> Sean #18225
>
>
>
>
> "sirus MOGHADASIAN" <cyrus.mgh@gmail.com>
> Sent by: nobody@groupstudy.com
> 08/16/2007 08:43 PM
> Please respond to
> "sirus MOGHADASIAN" <cyrus.mgh@gmail.com>
>
>
> To
> "Cisco certification" <ccielab@groupstudy.com>
> cc
>
> Subject
> simple question: why I get throughput of maximum 1Mbps in network that
> consists of 2 hosts and 1 switch!!?
>
>
>
>
>
>
> Hi group,
>
> simple question: why I get throughput of maximum 1Mbps in network that
> consist of 2 host and 1 switch!!?
>
>
> My OS is windows , even more interesting if I use access pint for one of
>
> two
> hosts with 802.11g I reach only 120KBps .
>
>
>
> What this issue stem from?
>
>
>
> Sirus
>
> _______________________________________________________________________
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
>
> _______________________________________________________________________
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
>
> -----------------------------------------
> ************************************************************ NOTE:
> This e-mail message contains PRIVILEGED and CONFIDENTIAL
> information and is intended only for the use of the specific
> individual or individuals to which it is addressed. If you are not
> an intended recipient of this e-mail, you are hereby notified that
> any unauthorized use, dissemination or copying of this e-mail or
> the information contained herein or attached hereto is strictly
> prohibited. If you receive this e-mail in error, notify the person
> named above by reply e-mail and please delete it. Thank you.
This archive was generated by hypermail 2.1.4 : Sat Sep 01 2007 - 11:32:11 ART