Re: tcp/ip performance problem

[Vitaly Ivanov]

Hi.
We have solved this problem. We had tuned TCP/IP stack in Linux Kernel
but it hadn't helped us because there was very law value backlog in
listen function
int listen(int sockfd, int backlog);
This function is used in nginx and lighttp demons. In nginx this value
had been set to 511 and in lighttp to 1024. We set both value to 65535
or -1 and solve our problem. Now we have about 700Mbit per second
instead 70.
But now we got another trouble - our server under heavily load because
ksoftirqd use 100% CPU. We suppose that this process is "bottle-neck".
Can we tune this process?
Thanks.

2008/9/26 Pekka Savola <pekkas@xxxxxxxxxx>:
> On Fri, 26 Sep 2008, Vitaly Ivanov wrote:
>>>
>>> Output how?  A single-stream output, or output to multiple (hundreds or
>>> thousands?) clients?  Where is the test client, in the same lan or
>>> further
>>> away in the network?
>>
>> Clients are the thousands HTTP browsers in the Internet. Both servers
>> contain the same static content size from 2Kbytes to 1Mbyte.
>
> How many simultaneous users is the typical figure?
>
> Is polling enabled? How have you tuned FreeBSD performance?
>
> Have you been able to figure out what's the bottleneck on linux vs on
> freebsd?  Is it CPU (which process?), interrupts, disk-IO, ...? (top,
> vmstat, etc. maybe help in that)
>
> One thing that is different in Linux/FreeBSD is rfc1323/window-scaling
> support.  I suspect you've already checked it, but I'd verify that both
> systems are sending out and receiving similar-looking packets (for example,
> similar MTUs, no fragmentation, no ICMP packet too bigs, ...)
>
> --
> Pekka Savola                 "You each name yourselves king, yet the
> Netcore Oy                    kingdom bleeds."
> Systems. Networks. Security. -- George R.R. Martin: A Clash of Kings
>
--
To unsubscribe from this list: send the line "unsubscribe linux-net" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html