In NSX-T, external routing is handled by the Tier-0 router, the gateway that connects the virtual network to external networks. It manages North-South traffic and supports static and dynamic routing (BGP/OSPF), defining how traffic exits and enters the data center while keeping East-West traffic internal.

Multiple Choice

Which component is primarily used for external routing in NSX-T?

The Tier-0 router is primarily used for external routing in NSX-T. Its main function is to interface with external networks and manage the flow of traffic between the virtual networking environment created by NSX-T and the outside world. It handles North-South traffic, which is the traffic entering and leaving the data center, as opposed to East-West traffic that typically occurs between virtual machines within the same data center. The Tier-0 router supports features like Static Routing and Dynamic Routing protocols (such as BGP and OSPF), enabling it to effectively communicate with external routers and networks. This makes it a critical component for facilitating external connectivity and routing policies for the entire NSX-T environment. While the Edge Node is part of the NSX infrastructure and provides various services including routing and load balancing, it is the Tier-0 router itself that is responsible for the external routing functions. The Tier-1 router mainly serves as an internal routing layer and connects with Tier-0 routers for accessing external networks, thus reinforcing the role of the Tier-0 router in this context. The Service Router and Tier-1 router serve different purposes within the architecture, such as managing East-West traffic or specific services, rather than functioning as the main interface for external routing.

When you’re mapping the NSX-T landscape, you quickly discover a clean separation of duties between the different routing components. Each piece plays a distinct role, but one stands out when it comes to bridging your virtual network with the outside world. If you’re wondering where external connectivity gets its footing, the answer isn’t a mystery—it's the Tier-0 router.

Think of NSX-T like a city’s transportation system. Inside the city, you have local streets and neighborhoods connected by internal routes that keep traffic flowing between servers, applications, and microservices. That’s where Tier-1 routers and various service components shine, making East-West traffic smooth and reliable. But for the big exit ramp to the wider internet or to upstream data centers, you need a main gateway—the Tier-0 router. It’s the primary interface to external networks, the high-capacity conduit that handles North-South traffic: incoming requests from clients and outbound replies traveling to partner networks, cloud providers, or any upstream connectivity you rely on.

Let’s break down the pieces so you can see how they fit together in everyday operations.

Tier-0 router: the external gateway

  • The Tier-0 router is designed to connect the NSX-T virtual world to the outside networks. It’s the main surface through which traffic enters and leaves your data center.

  • North-South traffic—the kind that crosses the boundary of your private cloud—benefits from the Tier-0’s centralized routing and policy enforcement.

  • It supports both static and dynamic routing. Static routes are straightforward, predictable paths you define, while dynamic protocols like BGP and OSPF let the router learn routes from connected peers and adjust on the fly. This is especially handy in multi-homed setups or when you’re working with a patchwork of partner networks and internet egress points.

  • In many deployments, the Tier-0 sits at the edge of the NSX-T fabric, often paired with high-performance physical or virtual devices that can take advantage of features like route redistribution, network address translation (NAT) in certain scenarios, and robust fault tolerance.

Edge Node and edge services: a complementary workhorse

  • You’ll hear about Edge Nodes as well, and they’re no small players in the NSX-T ecosystem. Edge Nodes host a collection of services—routing, NAT, load balancing, VPN capabilities, and sometimes firewalling—but they’re not the primary external gateway.

  • The Edge Node is excellent for extending NSX-T capabilities closer to the network edge or for providing specialized services that require close interaction with external networks. It’s like having a regional hub that handles a subset of external connectivity and advanced edge services, while the Tier-0 handles the main external ingress and egress.

  • In practice, Edge Nodes can offload some external routing responsibilities or provide additional redundancy and scale, especially in large environments or distributed deployments. But the core external routing function—the big door to the outside world—remains the Tier-0 router.

Tier-1 router: the internal workhorse

  • Inside the NSX-T fabric, Tier-1 routers sit one notch down in the hierarchy. They’re the workhorses for East-West traffic, connecting virtual networks to each other and to the Tier-0 for any North-South needs.

  • Tier-1 routers simplify internal segmentation and policy enforcement. They’re essential for micro-segmentation, service chaining, and ensuring that internal traffic follows the right paths with the right security measures.

  • If you imagine your network as a city, Tier-1 is the neighborhood connector—the roads and cul-de-sacs that keep people moving between apartments, offices, and community centers. They rely on the Tier-0 to reach places beyond the city limits, which is exactly where the Tier-0’s external reach matters.

Service Router: the specialization layer

  • The Service Router (formerly part of NSX-T in various forms) plays a more specialized role. It can handle features like scalable overlay services, some NAT and load balancing tasks, and tight integration with certain service chains.

  • It’s not the primary gateway to external networks. Instead, it acts as a facilitator for specific services that need to be close to the traffic they’re handling, or as a component in larger service infrastructures.

  • When you’re architecting a deployment, you’ll often decide how much of the load for external-facing services you want to push to the Tier-0 edge versus how much you want to keep within the internal fabric using Tier-1 and Service Router capabilities.

Why the Tier-0 choice makes sense from a design perspective

  • Centralized external connectivity: Having a single, robust gateway for external routing simplifies management, monitoring, and policy enforcement. It’s easier to maintain consistent egress/ingress rules and to apply global routing policies at the edge of the NSX-T fabric.

  • Policy coherence: The Tier-0 router provides a clear boundary for security and reachability policies that apply to external networks. You can design your security posture to span the edge consistently, without having to sprinkle external rules across multiple components.

  • Scalability and resilience: Tier-0 is typically built to handle substantial North-South traffic and to participate in resilient routing topologies with redundant links and peers. This makes it a natural home for the external routing function, with the Edge Nodes and Tier-1 routers filling the nuanced roles inside the data center.

A few practical pointers to keep the big picture in view

  • Plan your transit: If you’re connecting to an upstream ISP, a corporate MPLS cloud, or other data centers, align those connections with Tier-0 interfaces. Think of Tier-0 as the primary gateway, with the outside world’s routes learned and distributed there.

  • Consider routing protocols: If your environment has partners or external networks that require dynamic routing, BGP is a common choice for Tier-0 to exchange routes. OSPF can come into play in certain greenfield or more intricate network designs. Static routes still have a place for straightforward, predictable paths or for specific failover plans.

  • Use Edge Nodes strategically: If you need specialized external services close to your network edge, Edge Nodes can be the right fit. They’re particularly useful when you want to offload particular tasks from Tier-0 or when you’re implementing services that benefit from physical proximity to the edge or closer integration with external devices.

  • Map East-West versus North-South: Keep in mind that East-West traffic—internal data-center traffic—should be efficiently handled by Tier-1 routers and the internal NSX-T fabric. The Tier-0 is your global doorway, and keeping those duties well-defined helps with performance and troubleshooting.

Common misconceptions and clarifications

  • People sometimes assume Edge Nodes are the primary external gateway. It’s a tempting view because Edge Nodes do a lot and sit near the edge, but the main external routing surface remains the Tier-0 router. The Edge Node complements, rather than replaces, Tier-0 in the external routing story.

  • It’s not all about one feature, either. Yes, Tier-0 supports static and dynamic routing, but the real advantage comes from its central position in the routing topology. It’s the anchor that ties external reachability to the virtual world, while the other components handle internal segmentation, service chaining, and edge services.

A quick mental model you can rely on

  • Tier-0 router = the main external gateway. It’s the “front door” to your data center in the NSX-T fabric.

  • Tier-1 router = the internal highway system, connecting virtual networks and enforcing internal policies.

  • Edge Node = the specialized satellite office at the network edge, hosting services and extending capabilities where you need them most.

  • Service Router = the service-leaning helper that enables particular features and flows, without being the door to the outside.

Bringing it together with a practical example

Imagine you’re running a mid-size keystone application platform. Your workloads sit in a few virtual networks, each with its own security rules and service chains. When users from the internet come in—whether they’re customers, partners, or automated clients—the traffic lands at the Tier-0 gateway. The Tier-0 routes that traffic toward the appropriate internal networks via Tier-1 routers, applying the right policies along the way. If a service needs a bit of extra handling, like a dedicated NAT translation or a specialized load-balancer path, the Service Router steps in to support that workflow. For anything that’s staying inside the data center, East-West traffic cruises along the internal fabric, guided by Tier-1 routing and NSX-T policy enforcement. Edge Nodes can be leveraged to handle edge-specific tasks or to extend services to an external location, but the big doorway—the Tier-0 router—remains the backbone of external connectivity.

A closing thought: elegance in simplicity

In many ways, the clean separation mirrors good software design: a focused, central authority for external reach, complemented by specialized components that handle the rest. It’s not about having more pieces; it’s about having the right pieces in the right places. When you look at it like that, external routing in NSX-T isn’t mystifying so much as a deliberate, well-thought-out architecture. The Tier-0 router isn’t just a component; it’s the anchor that keeps your hybrid world connected, predictable, and secure as traffic flows in and out of the data center.

If you’re curious to see how a real-world NSX-T deployment maps out these roles, you’ll notice the same pattern across vendor blueprints and reference architectures. The Tier-0 router stands tall as the external gateway, with Tier-1 routers handling internal pathways, Edge Nodes offering edge services where needed, and the Service Router providing targeted capabilities that keep complex service chains running smoothly. It’s a symphony of parts designed to keep the data moving where it should, with policy, performance, and reliability all in harmony.