|
On Fri, 28 Jun 2002, R. Bruce Hoffman, Jr. wrote: > Reads are about the same... the real performance boost on mirroring comes > from writes. I believe this is incorrect. On a mirrored set the write command must be completed twice, whereas a single read operation on either side of the mirrored pair will fulfill a read request. Thus reads should be faster and writes much slower. > > Also, isn't mirroring considered a form of RAID? I seem to recall RAID > > having a number associated to indicate the form of RAID being implemented. > > RAID 5 is the data-striping method, RAID 1 is mirroring..... (?) > > Yup. mirroring is described in the original raid docs, but I consider them > as two different beasts. Raid usually requires special hardware (raid > controllers, iops) whereas mirroring is done by the system and can be done > at device, iop or bus level. Raid is a set on a controller. While this may be true on the iSeries, it isn't true in general. Other OSes do not require any special hardware for all levels of RAID. There is even some debate on whether hardware RAID outperforms software RAID. At the very minimum, software RAID (meaning the RAID functions are carried out by the kernel, not any hardware) is far more flexible in the way you use disks and the types of disks you use. It is also interesting to note that the iSeries by default implements what everyone else calls RAID level 0, i.e. spreading one filesystem over multiple disks. So every iSeries uses RAID, just most people don't call it that. James Rich
As an Amazon Associate we earn from qualifying purchases.
This mailing list archive is Copyright 1997-2025 by midrange.com and David Gibbs as a compilation work. Use of the archive is restricted to research of a business or technical nature. Any other uses are prohibited. Full details are available on our policy page. If you have questions about this, please contact [javascript protected email address].
Operating expenses for this site are earned using the Amazon Associate program and Google Adsense.