> > Disabling the EMC on master I have measured a baseline performance > > (in+out) of ~1.32 Mpps (64 bytes, 1000 L4 flows). The average number > > of subtable lookups per megaflow match is 2.5. > > Just running parallel ping between the tunnel end-point IPs on the two > servers increases the number of subtables for br-prv and increases the > average number of subtable lookups from 2.5 to 3.5 (or even 4.5 during > lifetime of the ARP megaflows) with a corresponding decrease of throughput > to 1.14 Mpps. > > > With the patch the average number of subtable lookups per megaflow > > match goes down to 1.25 (Apparently there are still two ports of > > different nature hashed to the same vector, otherwise it should be exactly > one). > > Even so the forwarding performance grows by ~30% to 1.72 Mpps. > > With the patch the parallel ping has no impact on average number of > subtable lookups and performance. The performance gain is then ~50%. >
I really like this approach. Ordering of the tables is also done to some extent in the ofproto classifier but what you are describing here is something that is agnostic to the actual protocols being matched on. +1 on the approach _______________________________________________ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev
