[ovs-dev] [PATCH v3 0/3] DPDK performance improvement: exact match cache

Daniele Di Proietto ddiproietto at vmware.com
Mon Aug 18 20:29:02 UTC 2014

This series introduces in dpif-netdev an exact match cache for frequently
used flows. It uses the RSS hash (currently implemented only for netdev-dpdk)
to search the miniflow in an hash table.

While there might be still some tuning left to do, these patch provides
significant performance improvements for the following simple testcase:

CPU: Intel(R) Xeon(R) CPU E5-2650 0 @ 2.00GHz
NIC: Intel Corporation Ethernet Controller 10-Gigabit X540-AT2 (rev 01)
     dual port (dpdk0, dpdk1)

    Bridge "br0"
        Port "dpdk1"
            Interface "dpdk1"
                type: dpdk
        Port "br0"
            Interface "br0"
                type: internal
        Port "dpdk0"
            Interface "dpdk0"
                type: dpdk

Source connected to dpdk0, sink to dpdk1

1 flow 64 bytes UDP packets going from source to sink

zero drop throughput:

   master: ~ 5.6 Mpps
new patch: ~13.0 Mpps

    - Ben suggestions:
      + fixed several coding style issues
    - Implemented Pravin suggestions:
        + dp_netdev_rss()
        + reverse order of batch lookup in dp_netdev_queue_batches()
    - Split dp_netdev_input() in two functions to save memory in exact match
      cache processing
    - Use variable length arrays to reduce cache footprint
    - Rebased

Daniele Di Proietto (3):
  ofpbuf: Add ofpbuf_rss()
  dpif-netdev: Add dp_netdev_packet_rss()
  dpif-netdev: Exact match cache

 lib/dpif-netdev.c | 451 +++++++++++++++++++++++++++++++++++++++++++++---------
 lib/netdev-dpdk.c |   3 +-
 lib/ofpbuf.h      |  14 ++
 3 files changed, 395 insertions(+), 73 deletions(-)


More information about the dev mailing list