Xref: utzoo comp.unix.ultrix:7060 comp.sys.dec:5682
Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sun-barr!decwrl!pa.dec.com!mogul
From: mogul@wrl.dec.com (Jeffrey Mogul)
Newsgroups: comp.unix.ultrix,comp.sys.dec
Subject: Re: Performance Tuning Ultrix 4.1
Keywords: paging swapping fast large load BSD
Message-ID: <1991May2.004642.24019@pa.dec.com>
Date: 2 May 91 00:46:42 GMT
References: <1991Apr30.160331.16215@milton.u.washington.edu> <1991Apr30.235901.3132@brolga.cc.uq.oz.au>
Sender: news@pa.dec.com (News)
Organization: DEC Western Research
Lines: 25

In article <1991Apr30.235901.3132@brolga.cc.uq.oz.au> ggm@brolga.cc.uq.oz.au (George Michaelson) writes:
>>	    Performance Tuning a DEC Ultrix 4.1 Workstation
>>			       Round 2
>>		Corey Satten, corey@cac.washington.edu
>
>This is absolutely fascinating reading. For the faint of heart (like me)
>Is there even a REMOTE chance that DECs Ultrix development people will
>fold in similar functionality sometime soon? I would really rather run
>something passed into me via the CD, since if it breaks I get to complain.
>
>....probably I'm asking for a DECtype to reply...

I'm a "DECtype", but I'm in a research group, so you can't believe
what I say.  I took the liberty of forwarding the "Performance Tuning"
paper to the people in the Ultrix group who are working on VM.  They
will probably find it fascinating reading, too.

Whether these changes make it into Ultrix is an interesting question.
One of the many potential benefits of a switch to an OSF/1 kernel
is that it comes with the Mach virtual memory system.  I suspect that
Corey Satten's discoveries won't apply, since the Mach VM system has
little or no connection with the BSD VM system that he has been fixing.
Of course, it might have interesting new bugs.

-Jeff