HP3000-L Archives

December 2000, Week 1

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Stan Sieler <[log in to unmask]>
Reply To:
Stan Sieler <[log in to unmask]>
Date:
Mon, 4 Dec 2000 11:17:31 -0800
Content-Type:
text/plain
Parts/Attachments:
text/plain (27 lines)
Re:
> I am willing to concede that whether or not *any* mirrored disc setup can be
> considered RAID depends on how transparent the mirroring is to the OS.  How
> can we split this hair, though?  Is HP's mirroring on MPE a part of the OS,
> or more like a driver?

Well, the driver is part of the OS, so...

However, to answer your question more usefully:

  HP's software mirroring is absolutely above the driver level.

(I.e., you wouldn't want to consider it to be driver level)

You can see this by doing "low level" (but above the various
disk drivers level) I/O to a disk drive.  If the drive is
mirrored, the I/O doesn't get mirrored.
To be more exact, you can do a sendio to read/write a disk
ldev, and *only* that ldev gets accessed, not the mirror (if any).

How do I know?  Actual experiments with <plug> DiskPerf </plug>,
where we can easily test one half of a mirrored pair by
using sendio to read from the ldev.

Stan Sieler                                           [log in to unmask]
www.allegro.com/sieler/wanted/index.html          www.allegro.com/sieler

ATOM RSS1 RSS2