[Kolab-devel] Cyrus Murder and kolab

[Martin Konold]

Am Montag, 16. Mai 2005 14:50 schrieben Sie:

Hi Matt,

> I am interested in these tests, the write capabilities for a busy kolab
> server we have found is not to be underestimated, yes IMAP synchs/reads are
> the majority but delivering 8-12000 emails an hour to disk takes its toll.

Yes, this uses up significant bandwidth towards the disk subsystem but please 
remember that a HW RAID-Cache can only hide latency but _not_ increase the 
sustained bandwidth. 

For large installations bandwidth matters when writing to disk but latency is 
less important. 

Latency only matters when watching for fast updates for the users. But in the 
general case even the dIMAP read operations of the users are bound to the 
bandwidth.

In this respect Kolab is very different from other typical server applications 
which tend to be really bound to latency instead.

BTW I did my tests with a HP DL 380 2/4GB machine and a Compaq MSA 500 RAID 
array testing with different Cache read/write settings. All this was done 
using the onboard 5i acontroller nd the Smart Array 5402 with 256MB Cache.

Basically the cache was not much relevant (< 5%) when measuring throughput 
(e.g. mailbombing)

> it sure does, but it is fair to say that large SATA arrays with controllers
> from 3ware is a good budget solution, we currently have 3 servers in place
> that uses SATA with great success, they are not at all busy but the
> installations are growing slowly.

Yes, I agree but this is due to the fact that the 3ware controllers emulate a 
full featured SCSI array ;-) 

The physical disks of ATA and SCSI drives are often not very different but the 
access protocols are. This protocol pays in the case of Kolab.

Regards,
-- martin

-- 
"I am committed to helping Ohio deliver its electoral votes to the
President next year."  -- 2004, Wally O'Dell - CEO of Diebold, Inc. 
e r f r a k o n - Stuttgart, Germany
Erlewein, Frank, Konold & Partner - Beratende Ingenieure und Physiker