Re: Slowdown on high-load machines with 3000 sockets

[Christian Schmid <webmaster@rapidforum.com>]

This issue has been tracked down more. This bug does NOT appear if I disable preemtive kernel.
Maybe this helps.

Nick Piggin wrote:
> Christian Schmid wrote:
> 
>> I already tried with 300 KB and even used a perl-hash as a 
>> horrible-slow buffer for a readahead-replacement. It still slowed down 
>> on the syswrite to the socket. Thats the strange thing.
>>
> 
> Do you have to use manual readahead though? What is the performance
> like if you just let the kernel do its own thing? The kernel's
> readahead provides things like automatic scaling and thrashing
> control, so if possible you should just stick to that.
> 
> Although you may want to experiment with the maximum readahead on your
> working disks:
> /sys/block/???/queue/read_ahead_kb
> 
> Also, can we get a testcase (ie. minimal compilable code) to reproduce
> this problem?
> 
> Thanks,
> Nick
> 
> 
>