All of lore.kernel.org
 help / color / mirror / Atom feed
* TCP session management
@ 2011-05-24 22:29 Nauman Tahir
  2011-05-25  3:26 ` Eric Dumazet
  0 siblings, 1 reply; 2+ messages in thread
From: Nauman Tahir @ 2011-05-24 22:29 UTC (permalink / raw)
  To: Kernel Newbies, Linux Kernel Mailing List

Hi

I am interested in learning about highly practiced/recommended
algorithms used for TCP session management in the stack. I am writing
code for an embedded processor which needs to process TCP/IP flows and
it does not have the stack. I am interested in implementing most
efficient algorithm to maintain/lookup about 5 million full duplex TCP
sessions.

Any ideas, pointers would be greatly appreciated.

Thanks
Nauman

^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: TCP session management
  2011-05-24 22:29 TCP session management Nauman Tahir
@ 2011-05-25  3:26 ` Eric Dumazet
  0 siblings, 0 replies; 2+ messages in thread
From: Eric Dumazet @ 2011-05-25  3:26 UTC (permalink / raw)
  To: Nauman Tahir; +Cc: Kernel Newbies, Linux Kernel Mailing List

Le mardi 24 mai 2011 à 15:29 -0700, Nauman Tahir a écrit :
> Hi
> 
> I am interested in learning about highly practiced/recommended
> algorithms used for TCP session management in the stack. I am writing
> code for an embedded processor which needs to process TCP/IP flows and
> it does not have the stack. I am interested in implementing most
> efficient algorithm to maintain/lookup about 5 million full duplex TCP
> sessions.

Hi

Not sure what are your needs...

Linux has no particular problems handling 5 million councurrent tcp
sessions, granted you have enough memory.

What do you mean by "it does not have the stack" ?




^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2011-05-25  3:26 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2011-05-24 22:29 TCP session management Nauman Tahir
2011-05-25  3:26 ` Eric Dumazet

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.