On Wed, 11 Oct 2000, Jeremy Stanley wrote:
> On Wed, 11 Oct 2000, Bill Carlson wrote:
> > Depending on how the RAID5 is setup, it can actually be slower than a
> > single disk! Especially on data writes. RAID5 is a compromise between data
> > safety, i/o speed and price. A decent RAID1 or RAID 1+0 will cost a little
> > more (depending on your needs), but the performance will be better as
> > well.
>
> This is only true for cheap hardware and/or small numbers of drives.
No. The performance is much more influenced by the data flow and patterns.
When talking RAID and which level should be used, what data is written and
how is the determining factor. Good vs Bad hardware is irrelevant, as you
can use the same hardware to achieve different levels of RAID, in
simplistic terms.
> Unfortunately, writes to RAID-1 cannot be striped, so you will get
> theoretically half the write speed of an n-drive RAID-0 or n+1-drive
> RAID-5. This assumes there is no latency due to RAID-5 parity calcs
> (which an SMP server with lots of cycles to spare for software RAID
> or a modern hardware RAID controller with a large cache can usually
> accomplish) and that you're not maxing out your controller or bus
> bandwidth.
RAID-5 lags RAID-0 or RAID-1 in writes, in most cases. First, you have
extra data to write (n x parity size). Second, it takes some defineable
amount of time to create that parity vs not creating it at all. Granted,
plenty of CPU will make that a small difference, assuming the bottleneck
is at the drives and not bus bandwidth.
Most RAID-1 implementations will at least get single drive performance on
writes, even some of the IDE based ones. Which is why a lot of people
recommend RAID1+0 if you want performance and are willing to pay for it.
Assuming random balanced read/write, of course. :)
> The real downside to RAID-5 these days is degraded-mode performance. That
> is, when a drive dies and the controller or kernel has to "recreate" the
> missing data based on the parity on other drives on the fly. RAID-1 (and
> thus RAID0+1) don't have this problem as the data on each drive is copied
> exactly to it's mate and thus doesn't need to be "recreated" for reads
> when crippled.
The other plus to RAID1+0 (RAID0+1, I forget which means which :) ) is
withstanding multiple drive failures. Lose a second drive while the RAID 5
is reconstructing and it's all over. RAID1+0 will fail if both drives in a
single mirror die, assuming 2 drives per mirror. Chances of each depends
on the number of drives, controller and such of course.
> My personal favorite for performance, reliability, price and size is
> RAID-5+5, but that gets into some very *large* array sizes... (16x18GB
> SCSI drives between 4 controllers yields 9x18GB worth of usable space but
> you can lose an entire controller or up to one drive off each controller
> and still have a working array). And at those sizes price is usually no
> object so companies tend to rely on outsourced-management BCV 3-way
> RAID-0+1 anyway...
Hmmm, what are you running that lets you do RAID5 over RAID5? Software
based over Hardware RAID? Special high end controllers?
> But this is a religious argument that has been hashed out repeatedly on
> many mailing lists I've been on in past years. The RAID-1 crowd and the
> RAID-5 crowd never gain converts as neither seems to be able to produce
> controlled benchmarks convincing enough to sway the other. My suggestion
> is to test, test, test under your environment or the best approximation
> you can fabricate, and pick what works best for you.
I agree, which to use where depends on too many factors to say which is
a better choice without more information about the requirements. The rest
is all just arguing. :)
Later,
Bill Carlson
------------
Systems Programmer [EMAIL PROTECTED] | Opinions are mine,
Virtual Hospital http://www.vh.org/ | not my employer's.
University of Iowa Hospitals and Clinics |
