Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watmath!clyde!rutgers!ames!sdcsvax!ucbvax!LBL.ARPA!nagy%warner.hepnet
From: nagy%warner.hepnet@LBL.ARPA.UUCP
Newsgroups: comp.os.vms
Subject: RE: VAX Performance and Coverage Analyzer
Message-ID: <870606055152.01g@lbl.arpa>
Date: Sat, 6-Jun-87 08:51:51 EDT
Article-I.D.: lbl.870606055152.01g
Posted: Sat Jun  6 08:51:51 1987
Date-Received: Sun, 7-Jun-87 05:03:13 EDT
Sender: daemon@ucbvax.BERKELEY.EDU
Distribution: world
Organization: The ARPA Internet
Lines: 18

We have PCA but I have not personally used it.  A couple of others
in my group have used it a bit and been pretty happy with it (their
only complaint was that PCA, particularly the Analyzer, was pretty
complex, i.e. lots of commands,d required a bit of playing with
to get the results desired).

I first learned of PCA at a DECUS session given by Digital describing
the use of PCA to improve the performance of the VAX FORTRAN compiler
and of PCA itself.  As I remember, particularly in the case of PCA,
the improvement factors were enormous and quite surprising (mostly
as a result of eliminating excessive page faulting by changing
memory allocation schemes and data structures)

= Frank J. Nagy
= Fermilab Research Division EED/Controls
= HEPNET: WARNER::NAGY (43198::NAGY) or FNAL::NAGY (43009::NAGY)
= BitNet: NAGY@FNAL
= USnail: Fermilab POB 500 MS/220 Batavia, IL 60510