[netperf-talk] Re: Difficultly in understanding the CPU utilization results for UDP stream test

[Brijraj Vaghani]

results inlined

On 7/3/07, Rick Jones <rick.jones2 at hp.com> wrote:
> > I am sorry to have removed netperf-talk from the thread. I got a
> > delivery failure twice, when trying to cc netperf-talk and that is why
> > I removed it.
>
> Did it say something about the list being closed?  (Lets take that part
> offline)
> >> At the risk of showing I've already forgotten the start of this thread,
> >> does the bogus CPU util also appear with TCP_STREAM using the same
> >> "pacing" values?
> >
> > I carried out the same test for TCP_STREAM and below are the results
> > which I got
> >
> > Recv     Send    Send
> > Utilization    Service Demand
> > Socket  Socket  Message  Elapsed                    Send
> > Recv     Send    Recv
> > Size     Size      Size        Time       Throughput  local
> > remote   local   remote
> > bytes    bytes     bytes      secs.       10^6bits/s    % U
> > % S      us/KB   us/KB
> >
> > 87380  16384   1024       121.18    0.08            -1.00
> > 12.90    -1.000  13028.73
> >
> > Recv    Send    Send                                      Utilization
> >     Service Demand
> > Socket Socket  Message  Elapsed                    Send     Recv
> > Send    Recv
> > Size     Size     Size        Time      Throughput  local    remote
> > local   remote
> > bytes    bytes   bytes       secs.      10^6bits/s    % S      % U
> > us/KB   us/KB
> >
> > 87380  16384   1024    121.18        0.08          21.98    -1.00
> > 22194.559  -1.000
>
> OK, next test :)  Allow the bitrate to be much higher.

I generated two more results for UDP_STREAM with throughputs of 0.8
Mbps and 8 Mbps

Socket     Message  Elapsed      Messages                        CPU
   Service
Size        Size         Time         Okay Errors   Throughput    Util
      Demand
bytes       bytes        secs            #      #       10^6bits/sec
% US     us/KB

110592    1024       10.10        1000      0        0.8
-1.00    -1.000
110592                  10.10        1000                0.8
   21.72    2193.020


Socket     Message  Elapsed      Messages                        CPU
   Service
Size        Size         Time         Okay Errors   Throughput    Util
      Demand
bytes       bytes        secs            #      #       10^6bits/sec
% US     us/KB

110592    1024       10.20       10000      0     8.0
-1.00    -1.000
110592                 10.20       10000              8.0
  26.11    266.328
 Still the numbers dont seem good. While carrying the above test, I
ran TOP as well. The numbers looked quite realistic in there (ard 1 -
2 %).
>
> thanks,
>
> rick
>