Re: [patch 1/4] - Potential performance bottleneck for Linxu TCP


* David Miller <davem@xxxxxxxxxxxxx> wrote:

I want to point out something which is slightly misleading about this
kind of analysis.

Your disk I/O speed doesn't go down by a factor of 10 just because 9
other non disk I/O tasks are running, yet for TCP that's seemingly OK
:-)

disk I/O is typically not CPU bound, and i believe these TCP tests /are/
CPU-bound. Otherwise there would be no expiry of the timeslice to begin
with and the TCP receiver task would always be boosted to 'interactive'
status by the scheduler and would happily chug along at 500 mbits ...

(and i grant you, if a disk IO test is 20% CPU bound in process context
and system load is 10, then the scheduler will throttle that task quite
effectively.)

Ingo
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Relevant Pages

  • Re: Vista Myths
    ... However, disk I/O on ... It shouldn't be with SATA drives, ... or that the scheduler handles ... > There is no way anyone can claim that OSX doesn't feel snappy. ...
    (comp.sys.mac.advocacy)
  • Fully and Partially pre-emptible kernel
    ... some computational stuff while Process B is waiting for disk I/O ... interrupt saying disk I/O is complete. ... Now can the scheduler be pre-empted and disk interrupt's ISR be called ...
    (comp.unix.solaris)