I strongly suspect that measuring IDE vs SCSI isn't as simple as
that. I've compared similar drives for SCSI vs IDE, and they
measure identically (IBM DCAS 34330W vs IBM DCAA 34330; both the test
measurement tools and real-time things like rebuilding kernels come
out to almost exactly the same after I compensate for the speed
differences between the two processors)
OTOH, I've got an AMD K6-2/333 that's running with a pair of 13gb IDE
drives and about 400mb of core, and it runs like a slug for kernel
compiles. I'd be happy to blame this on the drives, except that the
kernel tree slurps up into core and never leaves. Could it be that
the processor is wasting a lot of time managing core when it could
instead be helping me hack VIA Apollo DMA into 2.0.28?
____
david parsons \bi/ Too many processors, all running 2.0.28
\/
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/