Am 2007-08-18 11:36:02, schrieb Petter Reinholdtsen:
> In <URL:http://wiki.skolelinux.no/DebianEdu/Documentation/Etch/HowTo>,
> under 'Resize Partitions'. I wrote it. As the partitions become
> large fsck time grow exponentially larger, and backup restore times
> grows linearly. Very large partitions will give you problems in
> emergencies.
I can confirm this definitivly since I have a FileServer with now over
150.000.000 files which had 2 Raid-5 of each 15 SCSI-Drives of 300 GByte,
which mean 3.6 TByte each...
Trying ro recover takes weeks!!!
After spliting the 15 HDD Raid-1 onto two Raid-5 of 7 HDD each plus one
HotSpare and then creating 9 partitions of 200 GByte on each Raid-5
solved the problem...
This mean, I had to recode my software to be able to use dozen of
partitions... or better subdirectories.
But doing this increased the performance over 400%.
Note: I am using a singel 4-Channel ICP/Vortex with now 60 SCSI
15000RpM drives of 300 GByte and Disaster-Recovery will
kill your last nerv. So, using 8 Raid-5 on such system and
partitions of maximum 200 GByte is highly recommended.
I have choosen partitions of 200 GByte since my LTO support
it without compression.
Thanks, Greetings and nice Day
Michelle Konzack
Systemadministrator
Tamay Dogan Network
Debian GNU/Linux Consultant
--
Linux-User #280138 with the Linux Counter, http://counter.li.org/
##################### Debian GNU/Linux Consultant #####################
Michelle Konzack Apt. 917 ICQ #328449886
50, rue de Soultz MSN LinuxMichi
0033/6/61925193 67100 Strasbourg/France IRC #Debian (irc.icq.com)
Attachment:
signature.pgp
Description: Digital signature