A quick and easy performance improvement

[Chris Hammond]

Maybe this could be a new feature request?  Have MailScanner copy the bayes db that are used by SA that is being called by it to memory and then sync back to the drive at low disk IO times or a maximum time, which ever comes first?  Or am I out in left field somewhere?  Oh wait, don't answer that......

Chris
 
>>> Richard Lynch <rich at mail.wvnet.edu> 07/26/06 12:25 PM >>> 
uxbod wrote:

>Why not hold the bayes on a RAM partition, and have a cronjob that periodically backs it up throughout the day so that changes are not lost if the server crashes ?
>  
>

That would definitely improve things.  Seek time in RAM is zero!

While monitoring disk I/Os (iostat 1) I was surprised at the high number 
for bayes.  I didn't expect to see it so high.  One my systems it was 
actually higher than the I/O for the mail queues.

-- Rich

>On Wed, 26 Jul 2006 11:21:57 -0400, Richard Lynch <rich at mail.wvnet.edu> wrote:
>  
>
>>Nathan Olson wrote:
>>
>>    
>>
>>>Would noatime affect bayes operation on /var/spool?
>>>
>>>Nate
>>>
>>>      
>>>
>>Noatime will probably help since it would reduce the number of I/O
>>operations to the disk -- fewer I/Os is good for performance.  If I
>>recall correctly, noatime means that the system will not update the last
>>access date for the file.  One less I/O will certainly help.  The
>>benefit I'm going after comes from reducing disk seek time by putting
>>the bayes DB closer to the mail queues.  For me, using pretty much a
>>default installation, the benefit was in decreasing the IOWait time to
>>1/10th that value it was.
>>
>>-- Rich
>>
>>--
>>
>>
>>
>>
>>    
>>
>
>
>  
>


-- 



-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.