click_chatter v/s performance

Petros Zerfos pzerfos at CS.UCLA.EDU
Wed Apr 17 19:14:22 EDT 2002


We also noticed such a performance degradation on our wireless testbed
using click (running in userspace).

Since our application is delay-critical, we noticed that by removing the
logging from click, our performance improved by about 80% (from ~70msecs
to ~10msecs).

Also, click_chatter uses the fprintf() function when run in userspace,
which performes a buffered output, adding to the delay of logging.

direct write() to the filesystem as suggested by Douglas is better, or
external logging (through tcpdump for example to track a protocol)

Petros


On Wed, 17 Apr 2002, Arvind Venkatesan wrote:

> Hello everybody,
>
> I was just wondering about how much do
> print outputs through click_chatter (say one click_chatter for every 10
> data packets for statistics gathering) affect a router
> performance/introduce noise, since it does take system resources to do
> so.
>
> 		Actually, I tried using StringAccum object and the
> append methods it provides, so that I can do one click_chatter for every
> 100 data packets(instead of once every 10 packets) and printout the
> buffered statistics. For some reason it prints out the buffered
> statistics and some junk(that confuse the awk scripts) after every
> cycle. I am taking care to clear this StringAccum object after every
> cycle of 100 data packets. How do I get rid of that junk? Am I missing
> anything here?
>
> 	Is there any better way of collecting statistics (in case
> click_chatter has some kind of impact
> on the router performance) ?
>
> Thanks for the response!!!
>
> -arvind
>
>
>
>
>
>

--
Petros Th. Zerfos
UCLA - Computer Science Department
email : pzerfos at cs.ucla.edu




More information about the click mailing list