[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference netcad::hub_mgnt

Title:DEChub/HUBwatch/PROBEwatch CONFERENCE
Notice:Firmware -2, Doc -3, Power -4, HW kits -5, firm load -6&7
Moderator:NETCAD::COLELLADT
Created:Wed Nov 13 1991
Last Modified:Fri Jun 06 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:4455
Total number of notes:16761

2611.0. "Utilization % on Probe clarification" by DWCU05::MCGRATH (Tom McGrath-NYFD) Fri Aug 04 1995 05:18

In LTM, the utilization statistics collected reflected MAC layer monitoring 
only and did not include the physical layer statistics( ie. preamble (8 
bytes, runt packets and the inter packet interval(equivalent to 12 bytes) 
shown by the statistics in some other monitors. This is documented in the 
LTM manual. 

While the documentation for the probe states that utilization percentages 
include the preamble, what about the IPG time? This is a more correct way 
of measuring the actual theoritical wire utilization. Any insight into what 
the actual utilization statistics on the probe measure would be appreciated.

tm
T.RTitleUserPersonal
Name		DateLines
2611.1does not consider interpacket gapNAC::FORRESTFri Aug 04 1995 15:3619
    As stated in the probe manual, the equation used to calculate
    utilization is 
    
    	octets + (8 * pkts)
    	-------------------
    	seconds * 12500
    
    The (8 * pkts) part of the equation takes into consideration the
    preamble. If I divide 10Mbs by 8 to get the potential number of octets,
    I get 1250000. If I were to put 100 in the numerator to turn the
    fraction into a percentage, then that is the same as putting 12500 in 
    the denominator, which appears to be what Frontier has done.
    
    So the probe will never show 100% utilization due to interpacket gap 
    requirements, (also collisions). I won't take a stand as to whether 
    or not this is philosophically the right approach.
    
    jack