Some measurements and some questions

Brecht Vermeulen Brecht.Vermeulen at rug.ac.be
Mon Aug 14 23:48:04 EDT 2000


Hello,

I've done some measurements to compare a standard linux router to a Click
router both on throughput and delay of which the results are attached.

I've done this only on a AMD K6 processor, because my primary concern was
the difference, not the absolute performance.

Some questions :
* it seems that the routing table is a very consuming component, see the
document between click 3 (entries for 2 subnets) and click 4 (entries for
5 subnets) which makes the latency a lot higher than the standard linux.
Is this a known issue ?

* If there is given a parameter which is bigger than 2100000 bytes per
second (16Mbit/s), the /proc/click/meter/meters gives a minus rate and the
component doesn't do what is expected.

* Is there an upcoming meter component which has the behaviour as needed
in diffserv, this is all traffic under the limit goes through and only the
out of rate packets are dropped or remarked. Now the meter component sends
all traffic to one output according to the rate. (e.g. a 2 Mbit stream
through a 1 Mbit meter goes all trhough output 1, but normally only
half of the traffic should be going through output 1)

* When will click-1.1 be published as described in the TOCS paper ?

* The use of the optimisation (fastclassifier, devirtualise) doesn't seem
spectacular, is this normal ?

* When the time parameter to UDP gen is givenfor e.g. 10 seconds, it sends
only traffic for 5 seconds ?

* In the help of UDP gen is stated that the data in the UDP packet is :
UDP data, but the actual data is 'UDP packet !' (minor bug :-) )

* Are the syslog messages like : 'tulip_start_xmit while interrupts off
and eth2: Tx hung, 966228 vs. 966227' alarming ?

* Is there a plan to implement the mixed interrupt driven/polling method
as in the paper 'Eliminating Receive Livelock' (because now the PC is
continuously for 100% loaded, which produces a lot of heat day and night)

* Is it possible to print the tulip stats as in the function
tulip_print_stats ?

thanks in advance,
best regards,
Brecht Vermeulen

---------------------------------
Brecht Vermeulen

Department of Information Technology (INTEC)
Ghent University - IMEC
Sint-Pietersnieuwstraat 41
B-9000 Gent
Belgium

Phone : +32 9 267 35 86
Fax : +32 9 267 35 99
Email : brecht.vermeulen at intec2.rug.ac.be
-------------- next part --------------
A non-text attachment was scrubbed...
Name: workdoc_linux_vs_click.pdf
Type: application/pdf
Size: 487504 bytes
Desc: 
Url : https://amsterdam.lcs.mit.edu/pipermail/click/attachments/20000814/c7e39356/workdoc_linux_vs_click.pdf


More information about the click mailing list