[Linux-PowerEdge] Linux-PowerEdge Digest, Vol 102, Issue 6
jal at mdacorporation.com
Mon Nov 5 10:57:11 CST 2012
Gregory Gulik wrote:
> I am working on setting up a large RAID setup for a variety of
> functions that aren't very performance critical. I'm trying to use LVM
> on the filesystem to allow me to carve up the space for different
> functions with the flexibility to adjust things later if necessary.
> However I'm finding the performance of the LVM based filesystem is
> many times slower than the raw partition. I guess I expect some
> overhead but not 10x.
> Specifics of my test configuration:
> Dell PowerEdge 1950 with 12G RAM
> Dell PERC 6/E RAID controller
> Dell MD1000 Controller
> 8 x Seagate 2TB SATA drives in RAID-6 config (otherwise using default
> settings in OpenManage)
> Operating System is CentOS 6.3
> Total space available is 12TB and partitioned using parted as follows:
> Disk /dev/sdb: 12.0TB
> Sector size (logical/physical): 512B/512B
> Partition Table: gpt
You have a couple of trade-offs here. There was a bunch of (very good) advice about aligning stripe units and stripe widths.
Or you could buy 4 more 2TB SATA disks, run RAID-10, keep your 12TB total, and avoid the whole striping logic analysis thing and issues with performance. RAID-10 is faster for more kinds of IO patterns than RAID-5 or RAID-6. And when a drive breaks, the volume rebuild time is only a couple of hours not a couple of days.
The trade-offs of course are money vs performance vs effort to get the system running.
One other hint: set the the IO readahead size. The command blockdev --setra is another good thing to do for read IO.
More information about the Linux-PowerEdge