nv-l

RE: BandwidthUtilHdx statistics in Netview

2000-11-13 11:53:36
Subject: RE: BandwidthUtilHdx statistics in Netview
From: Gareth_Holl AT tivoli DOT com
To: nv-l AT lists.tivoli DOT com
Date: Mon, 13 Nov 2000 11:53:36 -0500
if you assume that if*Octets are snapshots of the number of bytes since the
agent was initialised (or something), then yes you would find that the
units would cancel to give seconds:
bits/(bits/sec)=sec
BUT
as pointed out by Alfred, snmpCollect is smart enough to provide delta
values for If*Octets (I think), thus seconds is also cancelled out.



Gareth Holl
Software Engineer
gholl AT tivoli DOT com

Tivoli Systems / IBM Corporation
Research Triangle Park,  NC  27703.    1-800-TIVOLI-8


Tsubasa_Kimura AT tivoli DOT com on 11/13/2000 10:34:03 AM

Please respond to IBM NetView Discussion <nv-l AT tkg DOT com>

To:   IBM NetView Discussion <nv-l AT tkg DOT com>
cc:    (bcc: Gareth Holl/Tivoli Systems)
Subject:  RE: [NV-L] BandwidthUtilHdx statistics in Netview






The follow-up question:

Do you know why are we still left with the unit of second after the units
in the formula cancel out each other?

Thank you,
Tsubas

Please respond to IBM NetView Discussion <nv-l AT tkg DOT com>

To:        "'IBM NetView Discussion'" <nv-l AT tkg DOT com>
cc:         (bcc: Tsubasa Kimura/Tivoli Systems)
Subject:        RE: [NV-L] BandwidthUtilHdx statistics in Netview



[IMAGE]
hi !

:)

since the if*Octets are counters, snmpcollect already gives you the right
delta values


mfG. Alfred Reibenschuh
INFORMATIONSTECHNOLOGIE AUSTRIA GMBH
TELEKOMMUNIKATION
Networkmanagement
Intranet Info-koordinator
A-1020 Wien, Lassallestrasse 5
T: ++43-1-21717-58947
F: ++43-1-21717-58979
E: alfred.reibenschuh AT it-austria DOT com

-----Original Message-----
From: Tsubasa_Kimura AT TIVOLI DOT COM [mailto:Tsubasa_Kimura AT TIVOLI DOT COM]
Sent: Wednesday, November 08, 2000 5:01 PM
To: nv-l AT tkg DOT com
Subject: [NV-L] BandwidthUtilHdx statistics in Netview



List,

      The formula in mibexpr.conf is BandwidthUtilHdx =
(8*100*(IFin+IFout))/IFSpeed. Now, the formula should be a percentage but
IFin and IFout are both in bytes which get converted to bits by multiplying
by 8 but IFSpeed in bits per second. The result of the division is still in
seconds with a multiplication of 100 for it to be a percentage. So the
question is whether we still need to divide the formula with the collection
duration to get a real percentage figure.


Thank you,
Tsubasa


[IMAGE]


(Embedded image moved to file: pic21132.pcx)
(See attached file: C.gif)
(See attached file: att1.htm)

Attachment: pic21132.pcx
Description: Paintbrush

Attachment: C.gif
Description: Compuserve GIF

hi !
 
:)
 
since the if*Octets are counters, snmpcollect already gives you the right delta values
 

mfG. Alfred Reibenschuh
INFORMATIONSTECHNOLOGIE AUSTRIA GMBH
TELEKOMMUNIKATION
Networkmanagement
Intranet Info-koordinator
A-1020 Wien, Lassallestrasse 5
T: ++43-1-21717-58947
F: ++43-1-21717-58979
E: alfred.reibenschuh AT it-austria DOT com

-----Original Message-----
From: Tsubasa_Kimura AT TIVOLI DOT COM [mailto:Tsubasa_Kimura AT TIVOLI DOT COM]
Sent: Wednesday, November 08, 2000 5:01 PM
To: nv-l AT tkg DOT com
Subject: [NV-L] BandwidthUtilHdx statistics in Netview


List,

      The formula in mibexpr.conf is BandwidthUtilHdx = (8*100*(IFin+IFout))/IFSpeed. Now, the formula should be a percentage but IFin and IFout are both in bytes which get converted to bits by multiplying by 8 but IFSpeed in bits per second. The result of the division is still in
seconds with a multiplication of 100 for it to be a percentage. So the question is whether we still need to divide the formula with the collection duration to get a real percentage figure.


Thank you,
Tsubasa




<Prev in Thread] Current Thread [Next in Thread>