Re: ULE/SCHED_SMP diff for 7.0, buildkernel & thanks.

From: Garrett Cooper <youshi10_at_u.washington.edu>
Date: Tue, 17 Jul 2007 21:56:29 -0700
youshi10_at_u.washington.edu wrote:
> On Tue, 17 Jul 2007, Attilio Rao wrote:
>
>> 2007/7/17, Abdullah Ibn Hamad Al-Marri <almarrie_at_gmail.com>:
>>> On 7/17/07, Jeff Roberson <jroberson_at_chesapeake.net> wrote:
>>> > With regards to buildkernel times;  I do not want to sacrafice 
>>> performance
>>> > on other benchmarks to improve buildkernel.  The problem is that 
>>> 4BSD is
>>> > as agressive as possible at scheduling work on idle cores.  This 
>>> behavior
>>> > that helps one buildworld hurts on other, in my opinion, more 
>>> important
>>> > benchmarks.
>>> >
>>> > For example: http://people.freebsd.org/~jeff/sysbench.png
>>> >
>>> > ULE is 33% faster than SCHED_4BSD at this mysql test.  This is a 
>>> direct
>>> > result of prefering to idle to make more efficient scheduling 
>>> decisions.
>>> > ULE is also faster at various networking benchmarks for similar 
>>> reasons.
>>> >
>>> > I also believe that while the real time may be slower on 
>>> buildworld the
>>> > system and user time will be smaller by a degree greater than the 
>>> delta in
>>> > real time.  This means that while you're building packages you have a
>>> > little more cpu time leftover to handle other tasks.  Furthermore, 
>>> as the
>>> > number of cores goes up things start to tip in favor of ULE 
>>> although this
>>> > is somewhat because it's harder for even 4BSD to keep them busy 
>>> due to
>>> > disk bandwidth.
>>> >
>>> > Thanks everyone for testing.  Can someone confirm that they have 
>>> tested
>>> > with x86 rather than amd64?  I will probably commit later today.
>>> >
>>> > Thanks,
>>> > Jeff
>>>
>>> Did you compare it to latest Linux fixes? is FreeBSD + ULE + MySQL
>>> still faster than linux?
>>
>> Just look at the link Jeff posted, it seems very well explaining :).
>>
>> Attilio
>>
>>
>> -- 
>> Peace can only be achieved by understanding - A. Einstein
>
> Unfortunately those results are still based on 2.6.20, not 2.6.22 (2 
> minor patch revision difference).
>
> I assume that that's for a vanilla Linux kernel?
>
> -Garrett

    Scratch my earlier comment about it works just fine. I seem to have 
broken my VM again, but it took 5+something iterations this round to get 
it to break.
    Now I have to restart Windows because of the random instabilities :).
-Garrett
Received on Wed Jul 18 2007 - 02:56:31 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:39:14 UTC