Powerhouse Quiz takes 60%+ CPU in EXEC mode

[Robert J.M. Edis]

G'day Chris

Are/is your PH programs sorting the RMS data?  I don't know about more
recent versions but I was able to improve performance considerably (x2 or
x3) by extracting the unsorted data set to a temp file, using VMS sort and
then continuing the PH processing - all controlled in a DCL script.  It
reduced a PH process trying to do everything itself and taking hours down to
a hybrid of PH and DCL taking only minutes.

PowerHouse is also creating work and sort files when you run the process.
Given that you only have 10% to 30% free disk space (and other users doing
edits, etc.) I think Allison is right; disk thrashing is a major problem for
you.  I remember DEC technicians saying that anything less than 25% free
disk space can seriously impede system performance.

Another thought.  Is/are your QTP processes competing with QUICK or other
programs for access to the data files?  Record and/or file locking could be
an issue.

Blue
PowerHouse consultant
Creative Computing, Inc.
Rhode Island, USA

Disclaimer:
The opinions and ideas expressed in this message are my own and have no
relationship to my current employer, Initial Technical Staffing, its client
CCI, or any of CCI's clients.



	-----Original Message-----
	From:	Chris Sharman [SMTP:Chris.Sharman@ccagroup.co.uk]
	Sent:	Wednesday, February 24, 1999 9:48 AM
	To:	Allison.Hamilton@Cognos.COM
	Cc:	Chris.Sharman@ccagroup.co.uk; powerh-l@lists.swau.edu
	Subject:	RE: Powerhouse Quiz takes 60%+ CPU in EXEC mode

	>If your files are poorly optimized,

	We do optimise data files every weekend.
	$ analyze/rms/fdl <file>
	$ edit/fdl/nointeractive/granularity=double/analysis=<fdl> <fdl>
	$ convert/fdl=<fdl>/nosort <file> <temp>
	$ rename <temp> <file>
	Do it to the dictionaries once in a while (they're indexed files
too).
	So the files should be optimized.

	>or quite fragmented,

	We use a defragger (the Digital one) every night.
	The disks aren't fragmented. The files in question aren't
fragmented.

	>or your disk close to full

	The disks are full-ish (70-90%), but I don't see that that's a
problem in
	itself,  it just tends to lead to fragmentation - not a problem for
large
	pre-allocated files.

	So I'm still in the dark, unless the (Digital) supplied optimization
still
	results in some sort of RMS thrashing, or fullish disks have some
impact I'm
	unaware of.
	A colleague thinks he's seen this from a 3GL program too. If that's
the case
	Powerhouse is probably exonerated (I've already cross-posted to
comp.os.vms,
	I'll relay anything interesting here).

	Thanks very much for the suggestions.
	Chris
	
______________________________________________________________________
	Chris Sharman			Chris.Sharman@CCAgroup.co.uk
	CCA Stationery Ltd, Eastway, Fulwood, Preston, Lancashire, PR2 9WS.
	= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
= = = =
	Subscribe: "subscribe powerh-l" in message body to
majordomo@lists.swau.edu
	Unsubscribe: "unsubscribe powerh-l" in message to
majordomo@lists.swau.edu
	powerh-l@lists.swau.edu is gatewayed one-way to
bit.listserv.powerh-l
	This list is closed, thus to post to the list, you must be a
subscriber.
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
Subscribe: "subscribe powerh-l" in message body to majordomo@lists.swau.edu
Unsubscribe: "unsubscribe powerh-l" in message to majordomo@lists.swau.edu
powerh-l@lists.swau.edu is gatewayed one-way to bit.listserv.powerh-l
This list is closed, thus to post to the list, you must be a subscriber.