January 12, 2012 at 12:16 am
@SQLRNNR exactly, although if you were to lose half the drives performance would suck in a big way. Your data is still intact though.
@steve-2 Jones that is, unfortunately a disadvantage of the RAID5. Array rebuild times for R5 and R10 are also a consideration, with lower rebuild times R10 is often the favourite.
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
January 12, 2012 at 8:25 am
Perry Whittle (1/12/2012)
@steve-2 Jones that is, unfortunately a disadvantage of the RAID5. Array rebuild times for R5 and R10 are also a consideration, with lower rebuild times R10 is often the favourite.
That's why you need a spare in the rack.
And if you do that, why not go with R10 then and slightly bigger drives? I've gambled on R5 in the past, but with today's capacities of drives, not sure I would anymore.
Side note: we started with R5 here in 2001, but when we purchased larger servers in 2005, we just went with multiple R1 arrays for protection since we were putting a spare drive in a 6 slot server, we just did 3 R1 arrays.
December 22, 2016 at 5:10 am
RAID5 is good for DWH application for reading operation.
Go through below link
http://database.ca/blog.aspx?blogid=1024
😎
December 22, 2016 at 5:14 am
Please note: 4 year old post
Gail Shaw
Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability
December 22, 2016 at 6:47 am
GilaMonster (1/11/2012)
... RAID 5 has about the worst write performance of the common RAID levels because of the need to compute and then write parity...
I had asked for tempdb on RAID 10 once, and the hardware guys ended up putting it on RAID 6, when I reviewed write latency metrics I didn't know whether to laugh or cry.
Viewing 5 posts - 16 through 19 (of 19 total)
You must be logged in to reply to this topic. Login to reply