| 1 |
On Saturday 07 October 2006 02:38, Richard Broersma Jr wrote: |
| 2 |
> I thought it might be nice to recap the characteristics of the possible |
| 3 |
> RAID configurations. Please feel free to correct any points that are |
| 4 |
> incorrect or inaccurate. |
| 5 |
|
| 6 |
> (4) Disk RAID5: |
| 7 |
> |
| 8 |
> Total Array Capacity: 750G |
| 9 |
> Reading Throughput: <= 2 x a single drive read cap. (similar to RAID0) |
| 10 |
> 1 file can be read at-most 1 time |
| 11 |
> or 4 separate file can be read simultaneously |
| 12 |
> Write Throughput: <= 1 x a single drive write cap. |
| 13 |
> Fault tolerance: can survive 1 disk failures |
| 14 |
|
| 15 |
I'm not sure of the raid 5 figures. For very big writes, i.e., lot's of full |
| 16 |
stripes, you'll get up to 3x write performance ( if your controller can write |
| 17 |
to all disk in paralell and is smart enough to calculate parity in memory ), |
| 18 |
but for small writes ( i.e, directory entries or DB updates ), you'll get |
| 19 |
about 0.5x ( exact figure is difficult, but for a single sector update a raid |
| 20 |
5 has to make 2 reads ( old sector, old parity ), then recalculate new parity |
| 21 |
as and then two writes AFTER them ( new sector, new parity ). Also read |
| 22 |
performance should be <= 3x if you give <= 4x for raid 0. |
| 23 |
|
| 24 |
F.O.S. |
| 25 |
-- |
| 26 |
gentoo-server@g.o mailing list |
Archives