Re: software raid X controler raid

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Fri, Feb 01, 2002 at 10:27:15PM -0200, Dinamica Sonora wrote:
> If i understood well, only DPT can do hardware RAID, all other are "more
> or less" software RAID. The ataraid support is just a layer between the
> requests and the physical disks.

yep
> 
> I mean "more" when running soft RAID, without RAID controler, since all
> disk requests/management would be done by kernel using main CPU and "less"
> when running w/ a RAID controler since some of the disk requests would be
> handled by the controler itself (releasing the CPU). OK. but what about an
> onboard RAID controler?  Wouldn't the performance be similar for both
> controler RAID and soft RAID?

In fact, no. Softwareraid will be faster. The kernel has more higher level
knowledge and uses that to work more efficiently....

> So, why are there so many MOBO w/ RAID onboard?

Looks nice on the box

> How would one (performance) benefit from using a RAID 0 controler onboard
> in a Linux system?

For Raid 0 or Raid 1, software raid in practice always beats the pants of
hardware raid, and doesn't use lots of cpu (think "100 cycles per 4Kb
block").

Raid5 is a different story where the XOR operation takes cpu time; hower a
pIV cpu runs at 2.2 Ghz while a raid controller cpu at, typically,
230Mhz....

Greetings,
   Arjan van de Ven







[Index of Archives]     [Linux RAID]     [Linux Device Mapper]     [Linux IDE]     [Linux SCSI]     [Kernel]     [Linux Books]     [Linux Admin]     [GFS]     [RPM]     [Yosemite Campgrounds]     [AMD 64]

  Powered by Linux