Top strangeness

EM1897 at aol.com EM1897 at aol.com
Thu Feb 10 14:40:35 PST 2005


In a message dated 2/10/2005 3:05:46 PM Eastern Standard Time, Matthew Dillon 
<dillon at xxxxxxxxxxxxxxxxxxxx> writes:

>
>:Ok, well I decided to do a little more testing before I dismantled my 
testbed.
>:What I found is that I get the following:
>:
>:Network load = 15K pps:     System 5% Interrupt 9-12%
>:
>:Network load = 34Kpps:      System 10% Interrupt 10-14%
>:
>:(the interrupt number bounces around a bit)
>:
>:Something doesn't seem right. Generally the load increases are linear for
>:this sort of test. The interrupt load seems about the same for what should
>:be twice the load. Even if no additional interrupts were generated I can't 
see
>:that the interrupt work could be that similar for double the load.
>
>    This sounds about right, actually.  The EM device is aggregating
>    packets and thus the interrupt code becomes more efficient at higher
>    loads.  It's also tight enough that it probably fits entirely in the 
cpu's
>    L1 cache.

I find that difficult to fathom. If that were the case you should also see 
such efficiencies in FreeBSD and its typically pretty linear. Is there a
way to reset the counters in vmstat so that I can monitor the marginal
interrupt rate?





More information about the Users mailing list