-
Bug
-
Resolution: Done
-
Normal
-
rhos-18.0.0
-
None
-
1
-
False
-
-
False
-
Committed
-
None
-
-
-
Moderate
Planning an OVS-DPDK deployment
https://docs.redhat.com/en/documentation/red_hat_openstack_services_on_openshift/18.0/html/deploying_a_network_functions_virtualization_environment/plan-ovs-dpdk-deploy_rhoso-nfv#two-numa-node-example-ovsdpdk-deploy_plndpdk-nfv
All examples still use the old heat template syntax, e.g.:
NIC 1 for DPDK, with two physical cores for PMD In this use case, you allocate two physical cores on NUMA 0 for PMD. You must also allocate one physical core on NUMA 1, even though DPDK is not enabled on the NIC for that NUMA node. The remaining cores are allocated for guest instances. The resulting parameter settings are: OvsPmdCoreList: "2,3,4,5,10,11" NovaComputeCpuDedicatedSet: "6,7,12,13,14,15"
Shouldn't those use the epdm syntax?
edpm_ovs_dpdk_pmd_core_list: "2,3,4,5,10,11"
cpu_dedicated_set: "4,5,6,7,12,13,14,15
