Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.
Comment: Published from Jenkins build: https://jenkins.onosproject.org/job/generateReleaseTestONWiki%20-%20DRAFT/80/

System Env:

  • Server: Dual XeonE5-2670 v2 2.5GHz; 64GB DDR3; 512GB SSD
  • System clock precision is +/- 1 ms
  • 1Gbps NIC
  • JAVA_OPTS="${JAVA_OPTS:--Xms8G -Xmx8G}"

ONOS Apps:

  • drivers
  • null
  • demo

ONOS Config:

  • cfg set org.onosproject.provider.nil.NullProviders deviceCount 35
  • cfg set org.onosproject.provider.nil.NullProviders topoShape linear
  • cfg set org.onosproject.provider.nil.NullProviders enabled true

"Constant-Load" Test Conditions:

  • F = eventually consistent flows rule - 122500, 10000 with FlowObj - total # of flows installed
  • N: # of neighboring ONOS's for flows to be installed when ONOS1 is the server installing flows
  • S: #servers installing flows
  • SW = 35 - total # of switches (Null Devices) connected to ONOS cluster evenly distributed to active ONOS nodes
  • FL: # flows to be installed on each switch

Command:

  • python3 $ONOS_ROOT/tools/tests/bin/flow-tester.py -f FL -n N -s servers

Note: Error bars in the graphs below represent standard deviations. Only the upper error bars are shown.

Following graphs include results using flow rule stores with both strong consistency and eventual consistency models. The ONOS team is still working on performance improvements for flow rule store with strong consistency.