Segment Routing
In today's interconnected world, where data traffic is growing exponentially, network operators face numerous challenges regarding scalability, flexibility, and efficiency. To address these concerns, segment routing has emerged as a powerful networking paradigm that offers a simplified and programmable approach to traffic engineering. In this blog post, we will explore the concept of segment routing, its benefits, and its applications in modern networks.
Segment routing is a forwarding paradigm that leverages source routing principles to steer packets along a predetermined path through a network. Instead of relying on complex routing protocols and their associated overhead, segment routing enables the network to be programmed with predetermined instructions, known as segments, to define the path packets should traverse. These segments can represent various network resources, such as links, nodes, or services, and are encoded in the packet's header.
Enhanced Network Scalability: Segment routing enables network operators to scale their networks effortlessly. By leveraging existing routing mechanisms and avoiding the need for extensive protocol exchanges, segment routing simplifies network operations, reduces overhead, and enhances scalability.
Traffic Engineering and Optimization: With segment routing, network operators gain unparalleled control over traffic engineering. By specifying explicit paths for packets, they can optimize network utilization, avoid congestion, and prioritize critical applications, ensuring a seamless user experience.
Fast and Efficient Network Restoration: Segment routing's inherent flexibility allows for rapid network restoration in the event of failures. By dynamically rerouting traffic along precomputed alternate paths, segment routing minimizes downtime and enhances network resilience.
Highlights: Segment Routing
The complexity and scale of MPLS networks have grown over the past several years. Segment routing simplifies the propagation of tags through an extensive network by reducing overhead communication or control-plane traffic. By ensuring accurate and timely control-plane communication, traffic can reach its destination properly and efficiently. As a second-order effect, this reduces the likelihood of user error, order of operation errors, and control-plane sync issues in the network.
Segment Routing’s architecture allows the software to define traffic flow proactive rather than reactively responding to network issues. With these optimizations, Segment Routing has become a hot topic for service providers and enterprises alike, and many are migrating from MPLS to Segment Routing.
What is segment routing?
Segment routing is a forwarding paradigm that allows network operators to define packet paths by specifying a series of segments. These segments represent instructions that guide the packet’s journey through the network. Network engineers can optimize traffic flow, improve network scalability, and provide advanced services by leveraging segment routing.
Segment routing is a flexible and scalable traffic engineering solution that simplifies network operations. At its core, segment routing leverages the concept of source routing, where the sender of a packet determines the complete path it will take through the network. Routers can effortlessly steer packets along a predefined path by assigning a unique identifier called a segment to each network hop, avoiding complex routing protocols and reducing network overhead.
Key Concepts and Components
To fully grasp segment routing, it’s essential to familiarize yourself with its key concepts and components. One fundamental element is the Segment Identifier (SID), which represents a specific network node or function. SIDs are used to construct explicit paths and enable traffic engineering. Another essential concept is the label stack, which allows routers to stack multiple SIDs together to form a forwarding path. Understanding these concepts is crucial for effectively implementing segment routing in network architectures.
The Building Blocks
- Segment IDs
Segment IDs are fundamental elements in segment routing. They uniquely identify a specific segment within the network. Depending on the network infrastructure, these IDs can be represented by various formats, such as IPv6 addresses or MPLS labels.
- Segment Routing Headers
Segment routing headers contain the segment instructions for packets. They are added to the packet’s header, indicating the sequence of segments to traverse. These headers provide the necessary information for routers to make forwarding decisions based on the defined segments.
Traffic Engineering with Segment Routing
- Traffic Steering
Segment routing enables precise traffic steering capabilities, allowing network operators to direct packets along specific paths based on their requirements. This fine-grained control enhances network efficiency and enables better utilization of network resources.
- Fast Reroute
Fast Reroute (FRR) is a crucial feature of segment routing that enhances network resiliency. By leveraging backup paths and pre-calculated segments, segment routing enables rapid traffic rerouting in case of link failures or congestion. This ensures minimal disruption and improved quality of service for critical applications.
Integration with Network Services
- Service Chaining
Segment routing seamlessly integrates with network services, enabling efficient service chaining. Service chaining directs traffic through a series of service functions, such as firewalls or load balancers, in a predefined order. With segment routing, this process becomes more streamlined and flexible.
- Network slicing
Network slicing leverages the capabilities of segment routing to create virtualized networks within a shared physical infrastructure. It enables the provisioning of isolated network slices tailored to specific requirements, guaranteeing performance, security, and resource isolation for different applications or tenants.
Understanding MPLS
MPLS, a versatile protocol, has been a stalwart in the networking industry for decades. It enables efficient packet forwarding by leveraging labels attached to packets for routing decisions. MPLS provides benefits such as traffic engineering, Quality of Service (QoS) control, and Virtual Private Network (VPN) support. Understanding the fundamental concepts of label switching and label distribution protocols is critical to grasping MPLS.
The Rise of Segment Routing
Segment Routing, on the other hand, is a relatively newer paradigm that simplifies network architectures and enhances flexibility. It leverages the concept of source routing, where the source node explicitly defines the path that packets should traverse through the network. By incorporating this approach, segment routing eliminates the need to maintain per-flow state information in network nodes, leading to scalability improvements and more accessible network management.
Key Differences and Synergies
While MPLS and Segment Routing have unique characteristics, they can also complement each other in various scenarios. Understanding the differences and synergies between these technologies is crucial for network architects and operators. MPLS offers a wide range of capabilities, including Traffic Engineering (MPLS-TE) and VPN services, while Segment Routing simplifies network operations and offers inherent traffic engineering capabilities.
MPLS and BGP-free Core
So, what is segment routing? Before discussing a segment routing solution and the details of segment routing vs. MPLS, let us recap how MPLS works and the protocols used. MPLS environments have both control and data plane elements.
A BGP-free core operates at network edges, participating in full mesh or route reflection design. BGP is used to pass customer routes, Interior Gateway Protocol (IGP) to pass loopbacks, and Label Distribution Protocol (LDP) to label the loopback.
Labels and BGP next hops
LDP or RSVP establishes MPLS label-switched paths ( LSPs ) throughout the network domain. Labels are assigned to the BGP next hops on every router where the IGP in the core provides reachability for remote PE BGP next hops.
As you can see, several control plane elements interact to provide complete end-to-end reachability. Unfortunately, the control plane is performed hop-by-hop, creating a network state and the potential for synchronization problems between LDP and IGP.
Related: Before you proceed, you may find the following post helpful:
- Observability vs Monitoring
- Network Traffic Engineering
- What Is VXLAN
- Technology Insight for Microsegmentation
- WAN SDN
What is Segment Routing |
|
Back to basics with Segment Routing Solution
Keep complexity to edges.
In 2002, the IETF published RFC 3439, an Internet Architectural Guideline and Philosophy. It states, “In short, the complexity of the Internet belongs at the edges, and the IP layer of the Internet should remain as simple as possible.” When applying this concept to traditional MPLS-based networks, we must bring additional network intelligence and enhanced decision-making to network edges. Segment Routing is a way to get intelligence to the edge and Software-Defined Networking (SDN) concepts to MPLS-based architectures.
MPLS-based architectures
MPLS, or Multiprotocol Label Switching, is a versatile networking technology that enables the efficient forwarding of data packets. Unlike traditional IP routing, MPLS utilizes labels to direct traffic along predetermined paths, known as Label Switched Paths (LSPs). This label-based approach offers enhanced speed, flexibility, and traffic engineering capabilities, making it a popular choice for modern network infrastructures.
Components of MPLS-Based Architectures
It is crucial to understand the workings of MPLS-based architectures’ key components. These include:
1. Label Edge Routers (LERs): LERs assign labels to incoming packets and forward them into the MPLS network.
2. Label Switch Routers (LSRs): LSRs form the core of the MPLS network, efficiently switching labeled packets along the predetermined LSPs.
3. Label Distribution Protocol (LDP): LDP facilitates the exchange of label information between routers, ensuring the proper establishment of LSPs.
1st Lab Guide on a BGP-free core.
Here, we have a typically pre-MPLS setup. The main point is that the P node is only running OSPF. It does not know the CE routers or any other BGP routes. Then, BGP runs across a GRE tunnel to the CE nodes. The GRE tunnel we are running is point-to-point.
When we run a traceroute from CE1 to CE2, the packets traverse the GRE tunnel, and no P node interfaces are in the trace. The main goal here is to free up resources in the core, which is the starting point of MPLS networking. In the lab guide below, we will upgrade this to MPLS.
Source Packet Routing
Segment routing is a development of the Source Packet Routing in the Network (SPRING) working group of the IETF. The fundamental idea is the same as Service Function Chaining (SFC). Still, rather than assuming the processes along the path will manage the service chain, Segment Routing considers the routing control plane to handle the flow path through a network.
Segment routing (SR) is a source-based routing technique that streamlines traffic engineering across network domains. It removes network state information from transit routers and nodes and puts the path state information into packet headers at an ingress node.
Benefits of Segment Routing:
1. Simplified Network Operations: Segment routing simplifies network operations and reduces the complexity of traditional routing protocols by decoupling the control plane from the forwarding plane. Network operators can define explicit paths for specific traffic flows, eliminating the need for complex and dynamic routing algorithms.
2. Enhanced Scalability: Segment routing offers improved scalability by enabling network operators to leverage the existing routing infrastructure while avoiding the scalability issues associated with traditional routing protocols. By leveraging a distributed control plane and existing MPLS (Multi-Protocol Label Switching) infrastructure, segment routing allows for efficient forwarding of packets across large-scale networks.
3. Traffic Engineering Flexibility: With segment routing, network operators have fine-grained control over the path packets take through the network. This flexibility allows for efficient traffic engineering, enabling operators to optimize network resources, prioritize specific traffic flows, and adjust the path based on real-time network conditions.
MPLS Traffic Engineering
MPLS TE is an extension of MPLS, a protocol for efficiently routing data packets across networks. It provides a mechanism for network operators to control and manipulate traffic flow, allowing them to allocate network resources effectively. MPLS TE utilizes traffic engineering to optimize network paths and allocate bandwidth based on specific requirements.
It allows network operators to set up explicit paths for traffic, ensuring that critical applications receive the necessary resources and are not affected by congestion or network failures. MPLS TE achieves this by establishing Label Switched Paths (LSPs) that bypass potential bottlenecks and follow pre-determined routes, resulting in a more efficient and predictable network.
2nd Lab Guide on MPLS TE
In this lab, we will examine MPLS TE with ISIS configuration. Our MPLS core network comprises PE1, P1, P2, P3, and PE2 routers. The CE1 and CE2 routers use regular IP routing. All routers are configured to use IS-IS L2.
There are four main items we have to configure:
- Enable MPLS TE support:
- Globally
- Interfaces
- Configure IS-IS to support MPLS TE.
- Configure RSVP.
- Configure a tunnel interface.
Synchronization Problems
Packet loss can occur in two scenarios when the actions of IGP and LDP are not synchronized. Firstly, when an IGP adjacency is established, the router begins to forward packets using the new adjacency before the actual LDP exchange occurs between peers on that link.
Secondly, when an LDP session terminates, the router forwards traffic using the existing LDP peer link. This issue is resolved by implementing network kludges and turning on auto-synchronization between IGP and LDP. Additional configurations are needed to get these two control planes operating safely.
Solution – Segment Routing
Segment Routing is a new architecture built with SDN in mind. Separating data from the control plane is all about network simplification. SDN is a great concept; we must integrate it into today’s networks. The SDN concept of simplification is a driver for introducing Segment Routing.
Segment routing vs MPLS
Segment routing utilizes the basics of MPLS but with fewer protocols, less protocol interaction, and less state. It is also applied to MPLS architecture with no change to the forwarding plane. Existing devices switching based on labels may only need a software upgrade. The virtual overlay network concept is based on source routing. The source chooses the path you take through the network. It steers a packet through an ordered list of instructions called segments.
Like MPLS, Segment Routing is based on label switching without LDP or RSVP. Labels are called segments, and we still have push, swap, and pop actions. You do not keep the state in the middle of the network, as the state is in the packet instead. In the packet header, you put a list of segments. A segment is an instruction – if you want to go to C, use A-B-C.
- With Segment Routing, the Per-flow state is only maintained at the ingress node to the domain.
It is all about getting a flow concept, mapping it to a segment, and putting that segment on a true path. It keeps the properties of resilience ( fast reroute) but simplifies the approach with fewer protocols. As a result, it provides enhanced packet forwarding behavior while minimizing the need to maintain the network state.
3rd Lab guide on MPLS forwarding.
The previous lab guide can easily be upgraded to MPLS. We removed the GRE tunnel and the iBGP neighbors. MPLS is enabled with the mpls ip command on all interfaces on the P node and the PE node interfaces facing the P node. Now, we have MPLS forwarding based on labels while maintaining a BGP-free core. Notice how the two CEs can ping each other, and there is no route for 5.5.5.5 in the P node.
Two types of initial segments are defined
Node and Adjacency
Nodel label: Nodel label is globally unique to each node. For example, a node labeled “Dest” has label 65 assigned to it, so any ingress network traffic with label 65 goes straight to Dest. By default, it will take the best path. Then we have the Adjacency label: a locally significant label that takes packets to an adjacent path. It forces packets through a specific link and offers more specific path forwarding than a nodel label.
Segment routing: A new business model
Segment Routing addresses current issues and brings a new business model. It aims to address the pain points of existing MPLS/IP networks in terms of simplicity, scale, and ease of operation. Preparing the network with an SDN approach allows application integration directly on top of it.
Segment Routing allows you to take certain traffic types and make a routing decision based on that traffic class. It permits you to bring traffic that you think is important, such as Video or Voice, to go a different way than best efforts traffic.
Traffic paths can be programmed end-to-end for a specific class of customer. It moves away from the best-path model by looking at the network and deciding on the source. It is very similar to MPLS, but you use the labels differently.
SDN controller & network intelligence
Controller-based networks sit perfectly with this technology. It’s a very centralized and controller application methodology. The SDN controller gathers network telemetry information, decides based on a predefined policy, and pushes information to nodes to implement data path forwarding. Network intelligence such as link utilization, path response time, packet drops, latency, and jitter are extracted from the network and analyzed by the controller.
The intelligence now sits at the edges. The packet takes a path based on the network telemetry information extracted by the controller. The result is that the ingress node can push a label stack to the destination to take a specific path.
- Your chosen path at the network’s edge is based on telemetry information.
Applications of Segment Routing:
1. Traffic Engineering and Load Balancing: Segment routing enables network operators to dynamically steer traffic along specific paths to optimize network resource utilization. This capability is handy in scenarios where certain links or nodes experience congestion, enabling network operators to balance the load and efficiently utilize available resources.
2. Service Chaining: Segment routing allows for the seamless insertion of network services, such as firewalls, load balancers, or WAN optimization appliances, into the packet’s path. By specifying the desired service segments, network operators can ensure traffic flows through the necessary services while maintaining optimal performance and security.
3. Network Slicing: With the advent of 5G and the proliferation of the Internet of Things (IoT) devices, segment routing can enable efficient network slicing. Network slicing allows for virtualized networks, each tailored to the specific requirements of different applications or user groups. Segment routing provides the flexibility to define and manage these virtualized networks, ensuring efficient resource allocation and isolation.
Segment Routing: Closing Points
Segment routing offers a promising solution to the challenges faced by modern network operators. Segment routing enables efficient and optimized utilization of network resources by providing simplified network operations, enhanced scalability, and traffic engineering flexibility. With its applications ranging from traffic engineering to service chaining and network slicing, segment routing is poised to play a crucial role in the evolution of modern networks. As the demand for more flexible and efficient networks grows, segment routing emerges as a powerful tool for network operators to meet these demands and deliver a seamless and reliable user experience.
Summary: Segment Routing
Segment Routing, also known as SR, is a cutting-edge technology that has revolutionized network routing in recent years. This innovative approach offers numerous benefits, including enhanced scalability, simplified network management, and efficient traffic engineering. This blog post delved into Segment Routing and explored its key features and advantages.
Understanding Segment Routing
Segment Routing is a flexible and scalable routing paradigm that leverages source routing techniques. It allows network operators to define a predetermined packet path by encoding it in the packet header. This eliminates the need for complex routing protocols and enables simplified network operations.
Key Features of Segment Routing
Traffic Engineering:
Segment Routing provides granular control over traffic paths, allowing network operators to steer traffic along specific paths based on various parameters. This enables efficient utilization of network resources and optimized traffic flows.
Fast Rerouting:
One notable advantage of Segment Routing is its ability to quickly reroute traffic in case of link or node failures. With the predefined paths encoded in the packet headers, the network can dynamically reroute traffic without relying on time-consuming protocol convergence.
Network Scalability:
Segment Routing offers excellent scalability by leveraging a hierarchical addressing structure. It allows network operators to segment the network into smaller domains, simplifying management and reducing the overhead associated with traditional routing protocols.
Use Cases and Benefits
Service Provider Networks:
Segment Routing is particularly beneficial for service provider networks. It enables efficient traffic engineering, seamless service provisioning, and simplified network operations, leading to improved quality of service and reduced operational costs.
Data Center Networks:
In data center environments, Segment Routing offers enhanced flexibility and scalability. It enables optimal traffic steering, efficient workload balancing, and simplified network automation, making it an ideal choice for modern data centers.
Conclusion:
In conclusion, Segment Routing is a powerful and flexible technology that brings numerous benefits to modern networks. Its ability to provide granular control over traffic paths, fast rerouting, and network scalability makes it an attractive choice for network operators. As Segment Routing continues to evolve and gain wider adoption, we can expect to see even more innovative use cases and benefits in the future.
