[OSM-dev] 0.6 move and downtime (re-scheduled)

[Florian Lohoff]

On Fri, Mar 13, 2009 at 03:41:16AM +0100, Stefan de Konink wrote:
> Mirroring will not increase performance because your RAID card will not 
> a priori know what files you are interested in, only the blocks you are 
> interested in and in the worst case will grab the same data from the 
> same disks and compare it ;)

In case of raid1 (mirroring) it should not compare data on simple read
but only do so in case of a reported ECC/disk block error.

Mirroring will increase performance with multiple concurrent readers
which is obviously the case with the OSM API Server. In the case of a
single reader you are right - it would not help latency but probably
streaming data (which would need intelligent read ahead)

So it is sensible to make it mirroring and it might even be a benefit to
do an 1 -> N mirroring.

Rule of thumb:

	More concurrent readers -> More spindles

This is why 72GB disks are still common in storage systems.

Flo
-- 
Florian Lohoff                  flo at rfc822.org             +49-171-2280134
	Those who would give up a little freedom to get a little 
          security shall soon have neither - Benjamin Franklin
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.openstreetmap.org/pipermail/dev/attachments/20090313/770be5fb/attachment.pgp>