Page History

Summary: Edge Scoping 

Distributed Cloud Infrastructure

...

Object Hierarchy, Cloud-agnostic Placement Policy & Homing Policies

Value:

• 5G VNF Placement
• Improve ONAP Deployability through Cloud-Agnostic Intent for Intra-DC Placement

References: ONAP R3+ Cloud Infrastructure Modeling; Cloud Infrastructure Aggregate Representation Classes 

...

• Cloud agnostic placement attributes are targeted to abstract the following cloud specific placement attributes
  • HPA attributes (e.g. Smart NIC Family, GPU Family) based on Cloud specific HW/SW support
  • Normalized CPU capacity for VMs/Containers based on Cloud specific HW support
    
    • Reference for CPU Normalization:  https://d1.awsstatic.com/whitepapers/Demystifying_vCPUs.df200b766578b75009ad8d15c72e493d6408c68a.pdf
  • Fine-grained Placement attributes based on Cloud specific SW support
    • e.g., Rack-level Anti-affinity-> Azure: Fault-Domain, AWS: Placement-Group 
    • e.g., Exclusivity ->                   Azure: Isolated VM, AWS: Dedicated Host 
    • e.g., Fine grained QoS ->       VMware Minimum guarantee, Kubernetes Burstable Class
• Cloud Agnostic Intent-based API – Example 1
  • Intent: Rack-level Anti-affinity for VMs within a VNF
  • Different Realization for different clouds
    • OpenStack-based (open source, Wind River, VMware VIO)
      • Heat Template – policy-group, anti-affinity policy
    • Azure
      • Fault-domain
    • AWS
      • Placement group 
• Cloud Agnostic Intent-based API with networking – Example 2
  • Intent: Rack-level Anti-affinity for VMs within a VNF with High Performance intra-DC data plane networking
  • Different Placement realization for different clouds
    • OpenStack-based (open source, Wind River, VMware VIO)
      • Heat Template – policy-group, anti-affinity policy
    • Azure
      • Fault-domain
    • AWS
      • Placement group 
  • Networking
    • Several Realization Options for Networking – not all options may be available in all clouds – the final realization choice could be based on least current resource usage or other criteria
      
      • Overlay in DPDK-based switch/router
      • Gateway in DPDK-based switch/router
      • Overlay in SmartNIC - implies SR-IOV is enabled in VMs
      • Gateway in SmartNIC
      • Overlay in ToR - implies SR-IOV is enabled in VMs
      • Gateway in ToR
      • Gateway in a HW appliance
    • The final choice is translated to cloud specific attributes
• For the 5G Service/VNF placement example in Note 3
  
  • Multi-Cloud interprets cloud agnostic intent as Physical DC Endpoint and translates to cloud-specific placement attribute such as Availability Zone
    • For this example, each distributed physical DC is in a separate Availability Zone for a OpenStack-based Cloud

...

Intra-DC networking with Intent-based Cloud Agnostic API (SO → MC → Cloud Region)

References: https://wiki.onap.org/download/attachments/28379482/ONAP-mc-sdn-v2.pptx?api=v2; https://wiki.onap.org/download/attachments/11928197/ONAP-mc-sdn.pdf?version=1&modificationDate=1506518708000&api=v2

Value:

• 5G High Performance Networking 

• Improve ONAP Deployability through Cloud-Agnostic Intent for Intra-DC network interconnect
  
  

...

• Cloud Agnostic Intent-based API example 2
  • Intent: High Performance intra-DC data plane networking
  • Several Realization Options – the final realization choice could be based on least current resource usage or other criteria
    
    • Overlay in DPDK-based switch/router
    • Gateway in DPDK-based switch/router
    • Overlay in ToR
    • Gateway in ToR
    • Gateway in a HW appliance
  • Realizations which are fixed
    • Underlay maps to ToR/Network Fabric
    • HW-assisted data plane maps to VMs/Containers with SR-IOV support 

Cloud

...

Resource Partitioning for Differentiated QoS

Value:

• All use cases

• 5G deployment on constrained Edge DCs

Edge Automation Requirement:

...

Aggregated Infrastructure Telemetry Streams

ONAP, as in R2, collects the statistics/alarms/events from workloads (VMs) and take any close loop control actions such as Heal a process, scale-out, restart etc.. In R3, infrastructure related statistics/alarms/events will be collected, generate actionable insights and take life cycle actions on the workloads.  Infrastructure statistics normally include performance counters, NIC counters, IPMI information on per physical server node basis.  To reduce the load on the ONAP, it is necessary that aggregated (summarized) information is sent to the ONAP from edge-clouds. 

...