On Fri, Aug 21, 2015 at 07:37:51AM -0700, Ben Pfaff wrote:
> On Fri, Aug 07, 2015 at 04:57:36PM -0700, Jarno Rajahalme wrote:
> > Add a benchmark command for classifier lookup performance testing.
> >
> > Usage:
>
> This usage note is good, but putting it just in the commit log will mean
> that it gets lost. It should be in a --help message, or failing that in
> a source code comment.
>
> I'm not sure I believe in the realism of random priorities. Random
> priorities are are worst case for the optimization that skips subtables
> based on priorities. Our NSDI paper showed that subtables tend to have
> a small number of priorities (often just one) in practice.
>
> If n_rules < n_subtables, or if the random numbers come out just right,
> then I think that the classifier will have fewer than the requested
> number of subtables. Also, if the same rule is generated more than
> once, the classifier will have fewer than the requested number of rules.
>
> Acked-by: Ben Pfaff <b...@nicira.com>
Also a minor spelling fix:
diff --git a/tests/test-classifier.c b/tests/test-classifier.c
index 54b595f..b2d4afd 100644
--- a/tests/test-classifier.c
+++ b/tests/test-classifier.c
@@ -1302,7 +1302,7 @@ run_benchmarks(struct ovs_cmdl_context *ctx)
n_lookups = strtol(ctx->argv[5], NULL, 10);
printf("\nBenchmarking with:\n"
- "%d rules with %d prioritites in %d tables, "
+ "%d rules with %d priorities in %d tables, "
"%d threads doing %d lookups each\n",
n_rules, n_priorities, n_tables, n_threads, n_lookups);
_______________________________________________
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
