× 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: Setting up ASP
  • From: dhandy@xxxxxxxxxxx (Douglas Handy)
  • Date: Fri, 15 Oct 1999 00:37:32 -0400

James,

>Using your situation in relation to ours, let's say we have 30 drives
>(6713) and are pushing 70% capacity.  So we go out and buy 6 more drives
>(6607).

 (snip)

>We add the 6 new drives (without RAID) in their own ASP AND move the
>historical file to it.  Now the remaining 30 drives are pushing 40%

The math seems creative.  If we ignore capacity lost to device parity
for the sake of simplicity, we have:

  30 drives @ 8.58MB = 257 MB; 70% full = 180 MB

We add 6 drives @ 4.19 MB = 25 MB.  Even if full, it would only reduce
the 180 MB down to 155 MB, which is still just over 60% used, not 40%.
Even if 6 of the original drives were lost to device parity, we'd be
at 70% of 206 MB (i.e. 144 MB) and moving 25 MB to another ASP would
only lower the % used to 58%.  And in all likelihood the 25MB ASP
would not be full so the actual reduction should be under 10%.

If the file is being used by < 5% of the people at once and the data
is < 10% of the dasd capacity, I think the other 95% of the users
count for an awful lot of the disk arm activity.

According to my Aug '99 copy of the "AS/400 Disk Arm Requirements
Based on Processor Model Performance" worksheets, a 640-2237 should
have an estimated minimum of 29 arms when using RAID-5 on the 6713
drives.

In my estimation, when you're so close to the theoretical minimum
number of arms, having the extra 6 arms available to help with the 95%
of the users not seeking historical data will stand a good chance of
doing more for overall throughput than a separate 25MB ASP.  And it
would certainly be easier to administer, unless the historical data
was well below the 25MB cap on the ASP size using the six 6607 drives.

I'm not arguing the theory, but I think you'd need more arms left in
the system ASP in relation to the recommended minimum and/or for the
historical data to be a bigger percentage of the total capacity to
come out ahead.  Those scenarios can well exist, and evidently did in
your case, but I think they are the exception rather than the norm.

How many arms did you move to a history data ASP, and how many arms
were left in the live data ASP?  What is your system's recommended
minimum disk arms?

YMMV,
Doug
+---
| 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 ...

Follow-Ups:
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.