Dear James,
Darrin Bulik has answered all but 2c clearly. I shall attempt to answer this for you.
Firstly, what is the Performance Log?
It is a log that describes the performance and the geometry of the drive
and makes the link between these two drive characteristics.
Why a Performance Log?
Such information is required for any form of advanced scheduling.
A Consumer Electronics device using a HDD will want to GUARANTEE
it's real-time performance, but simultaneously offer the best possible
performance for normal PC like data traffic. Only with some knowledge
about the drive performance can a system donate any 'spare' performance
to the PC like data whilst maintaining some guarantees.
The knowledge could be obtained during a drive qualification
process direct from the HDD vendor or, as with drive capacity, it could be
communicated by the drive itself. This latter is preferred as drives can be
interchangeable in an intelligent system. The Performance Log as such will not
make or break a streaming feature set, but it could ease qualification when
working with multiple HDD vendors and it clearly helps for servicing, for example.
As an aside, similar scheduling techniques have been used for many years in SCSI drives
which have operated as video servers. In the SCSI world the geometry is normal reported
and performance is measured. The fact remains that advanced scheduling requires both
geometry and performance.
In my opinion the Performance Log is clearly explained in T13 document:
http://www.t13.org/technical/d99135r5.pdf
This document describes the A/V model and definitions used in the Streaming Feature Set.
Generally, it has been neglected for some time and may no longer be up to date , but it
is nevertheless a useful reference document.
I hope I have indicated here why a Performance Log could be useful.
Finally, as Darrin also indicated the original proposal was based primarily on Seagate's SeaStream
work and was presented at T13 by Jonathan Haines, but I believe he is not with Seagate anymore.
Kind regards,
Stephen Cumpson.
--
Dr. Stephen Cumpson, Senior Scientist,
Group Storage Technologies, Building WY 1 45, (WY12),
Philips Research Labs., Prof. Holstlaan 4, 5656 AA Eindhoven, The Netherlands
Tel.: +31 40 274 2762 Fax: +31 40 274 4648
Email: mailto:[EMAIL PROTECTED]
Web: http://www.research.philips.com/
| [EMAIL PROTECTED]
Sent by: [EMAIL PROTECTED] 01-11-2002 19:14
|
To: [EMAIL PROTECTED] cc: [EMAIL PROTECTED] (bcc: Stephen Cumpson/EHV/RESEARCH/PHILIPS) Subject: [t13] Some Streaming cmd set questions Classification:
|
This message is from the T13 list server.
I have been asked to get answers to a few questions about the
currently specified Streaming cmd feature set.
Please (someone) answer these if you can remember from the AV
discussions of the last several years of T13 AV evolution.
1) CONFIGURE STREAM
Are the 'stream id' and 'AU' parameters expected to be preserved
across
* power cycle
* hardware reset
* SRST
2) READ LOG EXT (Performance Log)
a) Are there supposed to be any limits on the values of 'K', 'L' and
'M' ?
b) Is it preserved across power cycles and hardware reset ?
(like the read and write stream error logs)
c) Can anyone provide any rationale or description of how these
data are intended to be used ? With the non-existent
description currently in the document, these appear
to be worthless.
* sector time array
* position array
* access time array
Thanx !!!
-----------------------------------------------------------------
Jim Hatfield
ATA Interface Firmware & T13 (ATA/ATAPI) Standards Representative
Seagate Technology - PSG
e-mail: [EMAIL PROTECTED]
s-mail: 389 Disc Drive; Longmont, CO 80501 USA
voice: 720-684-2120
fax : 720-684-2711
====================================================
