× The internal search function is temporarily non-functional. The current search engine is no longer viable and we are researching alternatives.
As a stop gap measure, we are using Google's custom search engine service.
If you know of an easy to use, open source, search engine ... please contact support@midrange.com.


  • Subject: Re: Disk issue: More arms versus improved hardware
  • From: "James W. Kilgore" <eMail@xxxxxxxxxxxxxxxxxxx>
  • Date: Tue, 27 Feb 2001 09:48:31 -0800
  • Organization: Progressive Data Systems, Inc.

Jeff,

I think that the service times for data access stays relatively flat
until you exceed a certain percentage of overall usage.  I don't think
that there is a direct linear relationship between storage used and
response times.

The scheme of the AS/400 is that you scatter the data across all disk
drives so that when user#2 wants something, there is a possibility that
their request may be serviced by a different drive and they do not have
to wait for the arm to become available.  Although I say possibility,
I'm not sure on the probability.

Somewhere in the formula has to be included the number of concurrent
requests (20 users press ENTER at the same time and that causes 20
requests per user for 400 total requests) and although some drives may
be faster than others, the wait time for each request at it stacks on
the queue needs to be taken into account.

IMHO, using raw numbers (seek time, transfer rates, etc.) are like
someone asking what is the speed of an AS/400 processor.  It's not all
that relevant to the overall performance measurement.  So basically,
under heavy demand, a single drive at one speed may not perform as well
as two drives of a lesser speed.

That's pretty much the way that I've thought of it, so I'm in the "more
arms the better" school of thought.

jeff_carey@baxter.com wrote:
> 
> I too am wondering about this.  If you have more than twice the data on
> the same sized disk, it would seem that the seek time for an arm to find
> the data would be about half (at least).  Wouldn't that help mitigate the
> effect of having fewer arms?  Also, each arm is a point of failure as
> well, so would the performance increase of more arms outweigh the
> decreased time between failures?  I haven't done the math yet, but it's
> probably worth considering.
+---
| 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
+---

As an Amazon Associate we earn from qualifying purchases.

This thread ...

Replies:

Follow On AppleNews
Return to Archive home page | Return to MIDRANGE.COM home page

This mailing list archive is Copyright 1997-2024 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.