Re: How to turn off the "File System Real-time Protection" in Symantec Antivirus Corporate Edition?

From: Phil Weldon (notdisclosed_at_example.com)
Date: 04/28/04 

Date: Wed, 28 Apr 2004 13:54:16 GMT

Sorry for the typo. The I/O bandwidth should be 1 Mbit per second, not 1
MByte per second. The Amdahl/Case Rule is a 'rule of thumb' like 'Moore's
Law, and only approximate. A 'Northwood' 1.8 GHz Pentium might have 2,400
MIPS, indicating 2.4 GBytes memory and 2.4 GBits I/O bandwidth by the
Amdahl/Case Rule. The I/O bandwidth for a 32-bit, 33 MHz PCI bus would be
32 X 33 MHz ~= 1 GBit/second. The I/O bandwidth for a 32-bit 66 MHz X 8 AGP
bus would be ~= 16 GBit/second, but much of the AGP bus bandwidth is unused
for most applications. A 66 MHz 64-bit PCI bus is available on server
motherboards and a faster replacement for the PCI bus is soon to appear in
general motherboards; server applications approach the 1 MByte per MIPS
ratio, and for general workstation use, the 1 MByte per MIPS ratio is
approached if the swap file is included.

I suggest you use the performance data available in 'Task Manager' (select
'Processes', 'View', 'Select Columns') to find out where your bottleneck is. 

-- 
Phil Weldon, pweldonatmindjumpdotcom
For communication,
replace "at" with the 'at sign'
replace "mindjump" with "mindspring."
replace "dot" with "."
"Phil Weldon" <notdisclosed@example.com> wrote in message
news:YRxjc.15038$e4.58@newsread2.news.pas.earthlink.net...
> The request might have something to do with CPU resources.  What might not
> be noticed with a 3 GHz Pentium 4 in a workstation might be very noticable
> if the CPU can't keep up with disk bandwidth.  Without knowing that keep
> piece of information...  Amdahl/Case Rule:  A balanced computer system
needs
> about 1 MByte of memory and 1 MByte per second of I/O bandwidth per MIPS
of
> CPU performance.
>
> -- 
> Phil Weldon, pweldonatmindjumpdotcom
> For communication,
> replace "at" with the 'at sign'
> replace "mindjump" with "mindspring."
> replace "dot" with "."
>
>
> "Bill Sanderson" <Bill_Sanderson@msn.com.plugh.org> wrote in message
> news:Oh$p5iHLEHA.2660@TK2MSFTNGP09.phx.gbl...
> > Agreed.
> >
> > I don't think this is a universal issue with Symantec Corporate--my
> clients
> > haven't run into it--so I suspect this is fixable.  Turning off the scan
> > should be an absolute last resort, although reasonable for testing.
>
>

Relevant Pages