[Bioclusters] Xserve G5 memory

[James Cuff]

Sorry I know bad form to follow up on own post.  But I did a bit of STFW:

http://www.macosxhints.com/article.php?story=20030712130414339

Basically:

/usr/sbin/sysctl -w kern.maxvnodes=<some number bigger than it is now>

Looks like it will do the job to boost the UBC.  Let us know how you get 
on, if it even works at all...  should do though.

Best,

j.

James Cuff wrote:
> 
> Hrrm - so I wonder if that old HFS+ and the Mach kernel would seem to 
> the the issue here....  XNU does have a UBC (unified buffer cache), 
> although I've never really been a fan of Mach...
> 
> However, back to reality, I'd check that there really is enough memory 
> on the xserve for the database to fit, and look at the output of vmstat 
> for the IBM/Xeon and OSX as you run the tests.
> 
> But sorry, to answer your real question, I'm pretty sure the UBC on OSX 
> is 'tunable', I've just no idea how.  To be honest, the database should 
> either fit, and the kernel back magic will do the job for you, or it 
> will not, in which case it will flush it out, and you will see that in 
> {vm,io}stat
> 
> Best,
> 
> J.
> 
> -- 
> James Cuff, D. Phil.
> Group Leader, Applied Production Systems
> Broad Institute of MIT and Harvard. 320 Charles Street,
> Cambridge, MA. 02141.  Tel: 617-252-1925  Fax: 617-258-0903
> 
> 
> Victor M.Ruotti wrote:
> 
>> Hi Juan,
>> How exactly do you hold your databases in memory. Do you it through 
>> programming? It may help to describe how exactly this is done. I am 
>> also curious to know how you do it.
>>
>> Victor
>> On Oct 5, 2004, at 12:09 PM, Juan Carlos Perin wrote:
>>
>>> I have been running benchmarks with blastall on several different 
>>> machines.
>>> We've come to realize that one of the biggest differences affecting 
>>> search
>>> times is how the machines actually maintain the search databases in 
>>> memory.
>>>
>>> Eg..  On our IBM 8-way machine, the databases are held in the memory, 
>>> which
>>> seems to be an effect of the architecture of the machine, and search 
>>> times
>>> become incredibly fast after an initial run, which stores the 
>>> database in
>>> memory.  The same effect seems to take place on our Dual Xeon Dell (PE
>>> 1650), which also outpaces the Xserves significantly after an initial 
>>> run to
>>> populate the db in  memory.
>>>
>>> It would appear the the Xserves dump the db from memory after each 
>>> search,
>>> even when submitting batch jobs with multiple sequences in a file.  Is
>>> anyone aware of how this functions, and how this effect might be 
>>> changed to
>>> allow the db to stay in memory longer?  Thanks
>>>
>>> Juan Perin
>>> Child. Hospital of Philadelphia
>>>
>>> _______________________________________________
>>> Bioclusters maillist  -  Bioclusters@bioinformatics.org
>>> https://bioinformatics.org/mailman/listinfo/bioclusters
>>
>>
>>
>> _______________________________________________
>> Bioclusters maillist  -  Bioclusters@bioinformatics.org
>> https://bioinformatics.org/mailman/listinfo/bioclusters
> 
> 
> _______________________________________________
> Bioclusters maillist  -  Bioclusters@bioinformatics.org
> https://bioinformatics.org/mailman/listinfo/bioclusters