Re: Hardware RAID vs. Software RAID

"Anna C. Dent" <anacdent_at_hotmail.com> wrote in message news:Ye%kb.79328$Ms2.67542_at_fed1read03...

> >
> > What's wrong with RAID 5?
> With Raid-5 the redundancy information is stripped across all disks.
> For EVERY data block which is written to a RAID-5 volume the parity
> block must be recalcuated and also written to disk. For all intents
> and purposes RAID-5 doubles the number of write operations to disk.
> As the percentage of WRITE to READ operations increases available
> useful/application disk throughput decreases.

The interesting bit here is of course that RAID5 stripes the parity writes. And it also spreads the two write operations (data+parity) over at the very least two drives. With a good controller channel combination, it should be possible to overlap the writes so they proceed as one or slightly over that.

This is not bad at all considering what you gain. The problem of course are some of the SAN implementations of "RAID5" (remaining nameless in the interest of poetry - WTH!), which are in fact RAID3 and have all the disadvantages of this last one vis-a-vis databases.

If all SANs actually implemented RAID5 as it is defined, in all probability it wouldn't have gotten the bad name it has.

Throw in oodles of battery-backed cache and most problems even out. Some boxes are of course smarter than others at using that cache.

Would I use RAID5? Darn right I would, if it was properly implemented and suitable for the task. As is, in most cases it's not and it's better to go for SAME. If possible.

-- 
Cheers
Nuno Souto
wizofoz2k_at_yahoo.com.au.nospam