Polling slows down bandwidth
Коньков Евгений
kes-kes at yandex.ru
Fri Oct 29 17:56:29 UTC 2010
Zdravstvuyte, Chuck. (How do you do, Chuck ;-)
Вы писали 29 октября 2010 г., 20:23:19:
CS> On Oct 28, 2010, at 11:39 PM, Коньков Евгений wrote:
>> Здравствуйте, Chuck.
CS> Um, greetings?
yes, it is
>> Вы писали 28 октября 2010 г., 23:41:58:
>>
>> CS> On Oct 28, 2010, at 1:21 PM, Коньков Евгений wrote:
>>>> [ ... ]
>>
>> CS> What is "sysctl kern.clockrate", and have you increased kern.hz
>> CS> in /boot/loader.conf to at least 1000, if not 2000 or 4000?
>>
>> # vmstat -i
>> interrupt total rate
>> irq14: ata0 193948 6
>> irq16: rl0 42829515 1464
>> irq23: nfe0 41224044 1409
>> cpu0: timer 58494158 1999
>> irq256: igb0 106911 3
>> irq257: igb0 254606 8
>> irq258: igb0 2 0
>> Total 143103184 4892
>>
>> # sysctl kern.clockrate
>> kern.clockrate: { hz = 1000, tick = 1000, profhz = 2000, stathz = 133 }
>>
>> # sysctl kern.hz
>> kern.hz: 1000
>> but I have configured and installed kern with 2000HZ
>> "systat -v" shows that: 2002 cpu0: time
CS> Actually, the interrupt rate is tracking profile hz, which is
CS> roughly double the actual kern.hz-- per sysctl, you should try to at least double kern.hz.
ok, I will.
>>
>> CS> Polling mode operation generally performs better when using older
>> CS> 100Mbs ethernet NICs which do not support interrupt mitigation and
>> CS> various capabilities like TSO4; gigabit ethernet NICs are smarter
>> CS> hardware and can generally outperform polling mode.
>>
>> so using polling on gigabit NICs is a bottle neck? and is cause of low performance, is not?
CS> Simple answer is yes. It should be possible that you could tune
CS> polling to get similar performance, or at least better performance
CS> than you see now, but the additional hardware capabilities of
CS> gigabit NICs are likely to outperform polling mode, just as
CS> polling mode can generally outperform old 100MBs ethernet NICs.
CS> Regards,
--
С уважением,
Коньков mailto:kes-kes at yandex.ru
More information about the freebsd-net
mailing list