Re: Disk sizes and performance

[PaulMmn <PaulMmn@xxxxxxxxxxxxx>]

Our CE mentioned new, bigger disk drives in passing.  I replied
irreverently that we'd end up with poorer performance due to fewer arms.

He said he didn't think so;  the newer disks spin faster, so compensate for
the fewer arms.

--Paul E Musselman
PaulMmn@ix.netcom.com




>I am looking for some ball park estimates on when to use a greater number of
>small disk drives versus a smaller number of larger disk drives.
>
>Conventional wisdom is that the more disk arms you have the better the
>performance. Obviously this is directly related to the amount of I/O being
>done on the system. I feel however that there is some point (total disk
>required) at which adding additional disk arms does not have any significant
>improvement on performance and at that point it is more beneficial to use
>larger disks which results in financial savings in the form or lesss $/GB,
>less controllers, less racks etc.
>
>Any opinions as to what the magic number might be. ie if less than n GB on a
>system, use 4 GB drives. for n-m GB on a system use 8.5 GB drives and for
>greater than m GB on a system use 17 GB drives.
>
>Also what is the general opinion as to the DASD % utilisation at which I/O
>performance begins to degrade.


+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---