GroupStudy.com GroupStudy.com - A virtual community of network engineers
 Home  BookStore  StudyNotes  Links  Archives  StudyRooms  HelpWanted  Discounts  Login
Re: Sniffer results posted 03/06/1999
[Chronological Index] [Thread Index] [Top] [Date Prev][Date Next] [Thread Prev][Thread Next]


Let's do a little math:

10 Mbps * 3600 s = 36 Gb/hour at 100% utilization

36Gb/hour * 0.05 = 1.8 Gb/hour at 5% utilization

1.8 Gb/hour / 8 bits/Byte = 225 MB/hour at 5% utilization

Worst case scenario: all packets are of the maximum size of roughly 1500
bytes.  Thus,

225 MB/hour / 1500 Bytes/packet = 150,000 packets/hour

Of course these were the successfully transmitted packets.

6500 / (150,000+6500) = 4.15% collisions (worst case)

If you had smaller packets, which alot undoubtedly were, the collision
percentage is lower because you had more packets. Not to mention I probably
goofed somewhere either in my analysis or my arithmetic.


Matt

-----Original Message-----
From: Robert Klemme <rklemme@xxxxxxxxxxxxxxxxxx>
To: 'Cisco Listserv' <cisco@xxxxxxxxxxxxxx>
Date: Friday, March 05, 1999 3:57 PM
Subject: Sniffer results


>I am doing some monitoring of a network with a Sniffer, here are the
>results.
>
>In a 1 hour capture there were 6,500 collisions and the max utilization was
>89%,
>the average utilization was less than 5%.  Is this normal amount of
>collisions, this
>is anEthernet segment so I know it is contention based network but it seemd
>a b
>bit high to me.  I welcome any comments.
>
>thanks
>
>Robert
>---------------------
>To remove your name from the mailing list send a message to
Majordomo@xxxxxxxxxxxxxx with the body containing "UNSUBSCRIBE CISCO"

---------------------
To remove your name from the mailing list send a message to Majordomo@xxxxxxxxxxxxxx with the body containing "UNSUBSCRIBE CISCO"