Loading...

Virtual Wan Traffic Flow Patterns

Virtual Wan Traffic Flow Patterns

Virtual Wan Traffic Flow Scenarios

 

This guide offers an exploration of the essential elements related to vWAN traffic flows, including their significance in shaping flow topology patterns. At the heart of the vWAN branch architecture lie pivotal components such as VPN, SD-WAN, and ExpressRoute. Within the vHub ecosystem, the routing instances hold the crucial role in managing traffic flows, employing BGP for communication. These routing instances, concealed from the end user within the vHub, serve as the linchpin of the vWAN system. The routing instances bear the responsibility of intercepting and directing traffic flows, ensuring the smooth transmission of data packets. If packet flow inspection is also required, these can be aided by an Azure Firewall (AzFW) or Network Virtual Appliance (NVA) inside the vHub. The moment an NVA or AzFW is incorporated into the vHub, they gain control over the data plane packet interception. To enable this inspection, Routing Intent must be enabled on multi-vhub scenarios, or Private/Internet Traffic must be enabled via Firewall policy if an Azure Firewall is deployed within the vHub(s). The subsequent diagrams offer a pictorial depiction of the various flow patterns observed for single and multiple hubs within the vWAN infrastructure. The intent of this article is to provide a more detailed depiction of flow patterns than currently available in the Azure public documents, vWAN Flow PatternsUnderstanding the flows and components involved in vWAN pricing is also essential, and this can help explain how vWAN pricing works. For details on vWAN pricing, please refer to the vWAN Pricing Page

 

Single vWAN Hub

 

Single Region.png

Traffic Flows  Traffic Flow Paths
Flow A Spoke VM-->Routing Instances-->Spoke VM 
Flow B Spoke VM-->vHub VPN-->Branch VPN Concentrator
Flow C Spoke VM-->MSEE IP-->Branch Customer Edge (CE) (Egress flows for ExR bypass the ExR GW!)
Flow D Branch Customer Edge (CE)-->MSEE IP-->ExpressRoute GW-->Spoke VM
Flow E Branch Customer Edge (CE)-->MSEE IP-->ExpressRoute GW-->vHub VPN Gateway-->Branch VPN Concentrator
Flow F Branch VPN Concentrator-->vHub VPN--->MSEE IP-->Branch Customer Edge (CE) (Egress flows for ExR bypass the ExR GW!)

 

Quick Take-Aways

 

In the aforementioned flow paths, it's pertinent to note that the Virtual Private Network (VPN) can be substituted with a Software-Defined Wide Area Network (SD-WAN) tunnels to yield identical results in the diagram above. If either an Azure Firewall or Network Virtual Appliance (NVA) is deployed within the virtual hub (vHub), they will intercept packets in lieu of the route service's Virtual IP (VIP), granted that Routing Intent is activated, or Private/Internet security via Azure Firewall policy on a single vHub!

During operations within a single vHub, it's important to realize that flows to and from branches, whether via IPSEC, SD-WAN, or ExpressRoute, do not traverse the route service instances. Instead, this pattern is only observed in Spoke to Spoke flows, which contributes towards the vHub infrastructure limit, currently capped at 50Gbps per vHub, For additional details, please refer to the linked resource. Lastly, it should be mentioned that Branch to Branch flows also bypass the route service instances.

 

Multiple vWAN Hubs

 

Multi-Hub.png

Traffic Flows  Traffic Flow Paths
Flow A Spoke VM-->Routing Instances-->Remote Routing Instances-->Remote Spoke VM
Flow B Spoke VM-->Routing Instances--->Remote vhub VPN-->Remote Branch VPN Concentrator
Flow C Spoke VM-->Routing Instances-->Remote MSEE IP--->Remote Branch Customer Edge (CE) (Egress flows for ExR bypass the ExR GW!)
Flow D Branch Customer Edge (CE)-->MSEE IP-->ExpressRoute GW-->Remote Routing Instances-->Remote Spoke VM
Flow E Branch Customer Edge (CE)-->MSEE IP-->ExpressRoute GW-->Remote vHub VPN GW-->Remote Branch VPN Concentrator
Flow F Branch VPN Concentrator-->Remote MSEE IP-->Remote Branch Customer Edge (CE) (Egress flows for ExR bypass the ExR GW!)

 

Quick Take-Aways

The same principles apply to multiple vHubs in terms of SD-WAN tunnels and Azure Firewall/NVA behavior as they do a single vHub.

In the flow patterns for multiple vHubs, we note that Spoke-to-Spoke communication across hubs invariably involves the routing instances. Traffic moving from Spoke to Branch and vice versa also traverses a single set of routing instances. However, Branch-to-Branch traffic does not pass through the routing instances, just like single vHub behavior.

 

Note:

It's important to acknowledge that Azure Virtual WAN does not natively provide transit for ExpressRoute-to-ExpressRoute flows. To facilitate this, Global Reach is required. For Global Reach details, please refer to the linked resource. Alternatively, an Azure Firewall or NVA can be deployed in each vHub, combined with activation of routing intent and engagement with Microsoft Support. Information on Routing Intent can be found here to facilitate transit between ExR CircuitsUnder this configuration, we effectively push down RFC1918 prefixes to each ExpressRoute branch, thereby providing a supernet of transit connectivity.

 

Conclusion

I hoped this guide helped shed more light on the comprehension of packet flow dynamics within the services employed in vWAN. Armed with this understanding, users are enabled to make informed decisions regarding the infrastructure units, in terms of (Gbps), that should be allocated to each respective vHub, and also the size of their branch gateways (VPN/ExpressRoute) and SD-WAN devices. 

Published on:

Learn more
Azure Networking Blog articles
Azure Networking Blog articles

Azure Networking Blog articles

Share post:

Related posts

Power Automate vs Click2Export: Which Is Better to Export Reports in Dynamics 365?

Organizations using Dynamics 365 often need more than just workflow automation. They need a reliable way to export reports, generate documents...

1 day ago

Microsoft Power Automate – Run Python scripts using current Python versions in desktop flows

We are announcing the ability to run Python scripts using current Python versions in desktop flows in Microsoft Power Automate. This feature w...

2 days ago

Microsoft Power Automate – Connect to local AI models from desktop flows

We are announcing the ability to connect to local AI models from desktop flows in Microsoft Power Automate. This feature will reach public pre...

2 days ago

Power Automate – Enable Process Intelligence Experience in object-centric process mining

We are announcing the availability of Process Intelligence Experience in object-centric process mining in Power Automate Process Mining. This ...

2 days ago

Dataverse Plugins vs Power Automate: What Microsoft Experts Actually Use in Real Projects

Introduction One of the most common questions I hear from Power Platform developers and Dynamics 365 professionals is: “Should I use a D...

3 days ago

Power Automate| AZ Key Vault | 403 firewall error

💡 Power Platform  ·  Azure Key Vault Fix “Connector IP Not Authorized”in Azure Key Vault + Power Automate A...

3 days ago

Microsoft Purview | Data Lifecycle Management – Power Automate integration with records management

Microsoft Purview Records Management will integrate with Power Automate to enable custom workflows when retention-labeled items reach the end ...

4 days ago

Microsoft Power Automate – Measure time and cost savings for desktop flows

We are announcing the ability to measure time and cost savings for desktop flows in Microsoft Power Automate. This feature will reach general ...

8 days ago

How to stop an infinite trigger loop in Power Automate?

Stop a flow from re-triggering itself with trigger conditions, flag columns, or a create-only trigger.

9 days ago

25 Power Automate Flows That Can Save You 10 Hours Every Week

Introduction Imagine arriving at work on Monday morning and discovering that most of your repetitive tasks have already been completed automat...

9 days ago
Stay up to date with latest Microsoft Dynamics 365 and Power Platform news!
* Yes, I agree to the privacy policy