Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

ONAP ComponentLife Cycle PhaseEnhancements
PolicyDesign

Define Distributed Cloud Infrastructure Placement Policies (Note 3)

  • Leverage Standardized Distributed Cloud Infrastructure Object Hierarchy & Capabilities from A&AI
Multi-CloudDeploy

Support Distributed Cloud Infrastructure Capability Discovery (Note 1, Note 2)

For workload instantiation, translate cloud-agnostic Opaque TLVs intent from SO into cloud-specific placement attributes (Note 7)

A&AIDeploy

Support Standardized Distributed Cloud Infrastructure Object Hierarchy & Capability Database

  • Loose coupling between HW objects (private cloud) and SW objects (private and public clouds)
  • Includes Standardized Capabilities across clouds & Capabilities unique to certain clouds
  • Note:
    • Multi-Cloud Distributed Cloud Infrastructure Capability Discovery process will populate the aforementioned database
OOFDeployExecute Distributed Cloud Infrastructure Placement Policies for Optimized Service/VNF Placement across Cloud Regions (Note 4)
SODeployExtend SO↔OOF API to support opaque TLVs cloud agnostic intent (Note 5)

Assumption for Policy, SO, OOF:

...

  • For the 5G Service/VNF placement example in Note 3
    • OOF will pass the Physical DC End Point to SO as a opaque TLVcloud agnostic intent

Note 6:

  • For the 5G Service/VNF placement example in Note 3
    • SO passes the Physical DC End Point to Multi-Cloud as a opaque TLVcloud agnostic intent, besides the Cloud Region

...

  • 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
    • 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
  • For the 5G Service/VNF placement example in Note 3
    • Multi-Cloud interprets opaque TLV cloud agnostic intent as Physical DC Endpoint as 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

Cloud Infrastructure for Distributed Clouds (5G etc.) – Networking Focused (only the unique aspects from the previous section are brought out)

ONAP ComponentLife Cycle PhaseEnhancements
PolicyDesign

Define Distributed Cloud Infrastructure Network Realization Policies Per Cloud Region (Note 1)

  • Leverage Standardized Distributed Cloud Infrastructure Object Hierarchy & Capabilities from A&AI
Multi-Cloud (MC)Deploy

MC <-> SDN-DC (Private Cloud) or MC <-> Public Cloud interaction

  • Translate Cloud Agnostic intent to Cloud specific intent or imperative APIs
  • Return network endpoint (used for VNF placement)


SODeploy

Option 1: (SO -> MC -> Private/Public SDN-DC) -- desired

  • SO <-> MC IaaS Intent-based API (BesidesOne or more of Compute/Network/Storage/Analytics)

Option 2: (SO -> SDN-C -> MC → Private/Public SDN-DC)

  • SO <-> SDN-C interaction
    • SO uses the Generic Resource API (SDN-C NB) to SDN-C to determine the right Directed Graph to invoke
    • The Generic Resource API can be used to pass the cloud region and other cloud agnostic intent
  • SDN-C <-> MC IaaS Intent-based API
    • By looking up the directed graph, SDN-C calls MC with a programmable intent-based API which includes cloud region and other cloud agnostic intent

Note 1:

  • Intent-based API example 1
    • Intent: High Performance intra-DC data plane networking with no Host CPU usage

•Realization Possibilities
•Overlay in SmartNIC
•Gateway in SmartNIC
•Overlay in ToR
•Gateway in a ToR
•Gateway in a HW appliance
•Realizations which are fixed
•Underlay maps to ToR/Network Fabric
•No CPU usage for data plane networking maps to VMs/Containers with SR-IOV support

Cloud Infrastructure Impact – Definition, Creation & Management of Network Slice for 5G

...

ONAP ComponentLife cycle phaseActivities
AAI and ESRDeploy & Run time
  • Add/Modify/Delete recording and alerting rules
AAI and ESRRun time
  • Add/Modify/Delete Edge-cloud information
Multi-CloudRun time
  • Get Edge information from A&AI whenever Edge-Cloud is added or removed.
  • Prepare to wait for information from that Edge-cloud
  • Receive information from edge-cloud and put it in the time series DB.
  • Summation based on recording & alerting rules
  • Export information to DCAE via DMAPP or VES

Intent-based Cloud Agnostic IaaS API for Cloud Infrastructure

ONAP ComponentLife Cycle PhaseEnhancementsPolicyDesign