At 12:14 7-12-2003 -0500, Gaspar Bakos wrote:
I am about to build a system for data reduction, but before I do so, I
thought of posting this in case anyone has useful hints (don't do that!,
or be careful with...)
It would be a dual xeon mobo + 3ware Escalade card + 4x250Gb (WD) disk,
Good choice. I assume you mean the 4 port card.
running most probably RH9.0 and kernel 2.4.22-xfs. I haven't decided yet
about the arrangement of the 4x250Gb disks, but definitely there will be
XFS on them. My possibilities are: (I need total space more than 500Gb)
1. JBOD, each disk one partition
(drawback: I have to take care of not filling either of them)
2. RAID-0, one single 1Tb XFS partition
I would suggest using Raid 10 ( size = n/2) if you have have a enviroment
with heavy writes. If you won't write to the fs too much and it's mostly
reads you could use raid 5 (size = n-1).
If it's a production server I tend to waste money and opt for the raid 10
option instead since it's so much faster for database workloads and write
heavy environments. The raid 10 performance is also a lot more consistent
relatively big files (8Mb and 16Mb) accompanied with very small files
(<1kB). Recovery issues: I saw xfs_check run out of memory on a single
120Gb partition after an unexpected power failure. 3Ware configuration
issues that might be related to XFS, speed, efficiency.
I you read/write/modify lot's of small files create the fs with a larger log.
mkfs.xfs -l size=32768b /dev/foo
Being an astronomer, I am not that experienced with sw/hw issues...
I was always wondering when people write "we have been testing XFS with
60Tb filesystems" (and other magic numbers) - how they do that?
If it's 60TB I don't think it's a single filesytem under linux. The current
limit is 2TB per device. AFAIK this is fixed in the upcoming 2.6
I don't make sense, I don't pretend to either. Questions?