HCI data plane deployment with DPDK

You can deploy your NFV infrastructure with hyperconverged nodes, by co-locating and configuring Compute and Ceph Storage services for optimized resource usage.

NUMA node configuration example
Copy link

For increased performance, place the tenant network and Ceph object service daemon (OSD)s in one NUMA node, such as NUMA-0, and the VNF and any non-NFV VMs in another NUMA node, such as NUMA-1.

Table 1. CPU allocation
NUMA-0 NUMA-1

Number of Ceph OSDs * 4 HT

Guest vCPU for the VNF and non-NFV VMs

DPDK lcore - 2 HT

DPDK lcore - 2 HT

DPDK PMD - 2 HT

DPDK PMD - 2 HT
Table 2. Example of CPU allocation
NUMA-0 NUMA-1

Ceph OSD

32,34,36,38,40,42,76,78,80,82,84,86

DPDK-lcore

0,44

1,45

DPDK-pmd

2,46

3,47

nova

5,7,9,11,13,15,17,19,21,23,25,27,29,31,33,35,37,39,41,43,49,51,53,55,57,59,61,63,65,67,69,71,73,75,77,79,81,83,85,87

Recommended configuration for HCI-DPDK
Copy link

You can configure a standard set of parameters to tune your hyperconverged infrastructure (HCI) deployment.

Table 3. Tunable parameters for HCI deployments
Block Device Type OSDs, Memory, vCPUs per device

NVMe

Memory : 5GB per OSD OSDs per device: 4 vCPUs per device: 3

SSD

Memory : 5GB per OSD OSDs per device: 1 vCPUs per device: 4

HDD

Memory : 5GB per OSD OSDs per device: 1 vCPUs per device: 1

Use the same NUMA node for the following functions:

  • Disk controller
  • Storage networks
  • Storage CPU and memory

Allocate another NUMA node for the following functions of the DPDK provider network:

  • NIC
  • PMD CPUs
  • Socket memory