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 13,790 of 14,669    |
|    Christoph Kukulies to All    |
|    time between two packets    |
|    01 Aug 11 17:09:53    |
      From: kuku@physik.rwth-aachen.de              A question about packet processing speed:              I have a server and a client which communicate via a socket (ipv4, AF_INET).       The client sends a connect. The server does an accept and waits       for further messages from the client on that socket (using select()).              The design of the higher level protocol is the following way:                     client server                     send(REQUEST) ---------------------------> recv(REQUEST)       recv(DATA) <-------------------------- send(DATA or ERROR CODE)       recv(END) <-------------------------- send(END)              Now it turns out that the doubly sending of packets makes the whole       communication very slow and I have no idea why at the moment.       Each transaction in the above way takes 0.2 seconds and calculatiing alone       from the wire speed (100Mbit/s) for 256 bytes I would expect smaller       transmission times. Lowering the packet size down to 70 doesn't have       any impact on the time I'm achieving.       But when I remove the sending of the END packet, I'm getting a drastic       performance increase, that is the transaction goes down to 600 in 1 second.       A 600fold performance increase.              --       Chris Christoph P. U. Kukulies kukulies (at) rwth-aachen.de              --- 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