Forums before death by AOL, social media and spammers... "We can't have nice things"
|    comp.protocols.tcp-ip    |    TCP and IP network protocols.    |    14,669 messages    |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
|    Message 12,935 of 14,669    |
|    Martijn Lievaart to arya    |
|    Re: Differences in PUSH flag implementat    |
|    22 Jun 09 08:38:48    |
      e8c685ee       From: m@rtij.nl.invlalid              On Sun, 21 Jun 2009 18:01:24 -0700, arya wrote:              > I've run an application on three different OS with different results.       > The application is supposed to a file to an embedded device, waiting for       > an 'OK' response from the embedded devices between each line of the       > embedded code that it transfers. This is done over a telnet connection.       >       > When I measured the time between successive telnet packets sent by the       > PC, I measured 500ms (Windows 2003 Server), 300 ms (Vista) and 1ms       > (Linux 2.6). When I captured the TCP traffic, the delay seems to be at       > the PC end. The embedded device responds with the 'OK' within 1ms every       > time. The PC application take 300 - 500ms to turn it around and send the       > next command.              Sounds like Nagle being sometimes enabled and sometimes not. Google       Nagles algorithm.              HTH,       M4              --- SoupGate-Win32 v1.05        * Origin: you cannot sedate... all the things you hate (1:229/2)    |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
(c) 1994, bbs@darkrealms.ca