Xref: utzoo comp.arch:5842 comp.os.vms:7641
Path: utzoo!attcan!uunet!peregrine!elroy!ames!amdcad!weitek!dms!shepperd
From: shepperd@dms.UUCP (Dave Shepperd)
Newsgroups: comp.arch,comp.os.vms
Subject: Re: Pagefault time
Message-ID: <520@dms.UUCP>
Date: 2 Aug 88 00:29:34 GMT
References: <8103@rama.UUCP>
Distribution: na
Organization: Atari Games Inc., Milpitas, CA
Lines: 25

From article <8103@rama.UUCP>, by jdc@rama.UUCP (James D. Cronin):
> 
> 
> Do any of you netlanders know what the paging performance is for
> Vaxen (or any other machine, for that matter)?  For example, how
> long does it take to service a "soft" page fault (i.e. the desired
> page is in the free/modified page list) vs. a "hard" page fault
> (i.e. going out to disk).  

Measured softfault times:
	Vax 11/730 	1700
	Vax 11/750	 800
	Vax 11/780	 500
	uVax II		 500
	Vax 3xxx	 180

All times in microseconds and are reasonably accurate. The one I remember 
for sure is the the 11/780 at 498us.

The 730 and 750 were measured with VMS 3.x. The 780 was measured under
VMS 3.x and VMS 4.x (no difference). The uVax II and Vax 3xxx were
measured under VMS 4.7.

Hard fault times are a function of your paging disk's specs coupled
with what other I/O demands your system is placing on it.