Re: How I built a 2.8TB RAID storage array

From: Folkert Rienstra (see_reply-to_at_myweb.nl)
Date: 03/03/05


Date: Thu, 3 Mar 2005 01:04:43 +0100


"Jon Forrest" <forrest@ce.berkeley.edu> wrote in message news:4224E4B9.5070407@ce.berkeley.edu
> John-Paul Stewart wrote:
>
> > Yes. If you look at the CPUs on RAID cards, they're a lot less
> > powerfull than the host CPU (even on the most expensive $1000+ cards).
>
> That's because, other than performing the XOR operations
> for writes, they don't have to do very much.
>
> > I haven't seen any benchmarks comparing software RAID to hardware
> > RAID where the host CPU was heavily used. They always seem to be done
> > on otherwise unloaded systems. But then, everything I've read agrees with
> > the previous poster's assesment that hardware RAID will win when the
> > host CPU is otherwise occupied.
>
> Right. Even when a server is busy, satisfying read requests and non-RAID-
> 5 requests shouldn't add much to the load. Most of the work is done by
> the intelligence built-in to the ATA or SCSI electronics on the disk.
> The latency imposed by the movement of the arms and platters dominates
> the latency caused by a busy CPU.
>
> For a while I was a big fan of those cheap IDE pseudo-RAID 0 and 1
> controllers but I now realize that they really don't provide much benefit
> compared to just adding more IDE channels since those controllers do so
> little.

> That's one reason why you can convert one of those
> Promise IDE boards into a RAID controller by simply adding a resistor.

That was only possible with the original Ultra 66 and 100 boards.
Next they used different PCI IDs for Ultra and Fasttrak.

>
> Jon



Relevant Pages

  • Disk I/O DEAD slow after upgrade from 5.5 -> 6.1
    ... We recently upgraded a mail server from FreeBSD 5.5 to 6.1 and also moved the /home partition from a shared RAID 1 Volume with the ... Could this be related to FreeBSD 6 and the RAID card? ... CPU: IntelXeonCPU 2.80GHz ... <ACPI PCI bus> on pcib0 ...
    (freebsd-questions)
  • Re: Raid0 auf 100MB/s limitiert (WinXP, ICH8R)
    ... getakete moderne CPU. ... Allerdings erhöht ein RAID 1 beispielsweise die I/O-Last auf dem Bus beim Schreiben um einen Faktor 2. ... Auch eine Last auf dem Speicherbus von mehr als 30% der Zeit hat deutliche Auswirkungen, da nun jeder Speicherzugriff im Mittel etwas warten muss. ... Vergleiche mal ein RAID1 mit 2 Platten mit einem mit 32 Platten! ...
    (de.comp.hardware.laufwerke.festplatten)
  • aacraid and smp kernel: SCSI hang
    ... Our main server is an Opteron system with an Adaptec SATA RAID 2410SA ... controller and 4x 250GB Seagate Barracuda disks attached to it, ... We recently upgraded the server with a second Opteron CPU. ... Please note that aacraid 1.1.5-2400 has been used in the single CPU ...
    (Debian-User)
  • Re: filesystem: 12h to delete 32GB of data
    ... RAID5 is good for sequential reads -- eg. ... blob of data up to the CPU. ... -- which works a great deal better on RAID10. ... For most H/W raid that equates to ...
    (freebsd-performance)
  • Re: filesystem: 12h to delete 32GB of data
    ... RAID5 is good for sequential reads -- eg. ... blob of data up to the CPU. ... -- which works a great deal better on RAID10. ... For most H/W raid that equates to ...
    (freebsd-questions)