Re: Yet another disk transfer speed problem

[Holger Lubitz]

Bharath Krishnan proclaimed:
> 
> Hi,
> 
> I would expect the disk which acts slower(maxtor) to be atleast as fast
> as the other one (ibm).

Could you provide fdisk -l for both? For some odd reason unknown to me
some filesystems give slower results with hdparm than others, even with
the buffer-cache reads (which are intended to measure memory speed, not
drive speed, and thus should be the same for all drives on a given
mainboard). Also, hdparm directly on the drive device is often a bit
slower than hdparm for the first (outermost) partition. These problems
have been far worse in older kernels, though. With 2.2 I once
benchmarked a vfat-partition at half the speed the same partition gave
as ext2.

Holger