IP-MPLS (Multiprotocol Label Switching)

IP-MPLS Overview

• IP-MPLS, or Multiprotocol Label Switching, serves as a routing mechanism in telecommunications networks.
• It accelerates communications using short labels for data routing in place of longer network addresses.
• IP-MPLS encapsulates various individual streams into packets, giving rise to its designation as a "multiprotocol" routing technique.
• The Internet Engineering Task Force (IETF) provides the specified MPLS framework that enables efficient forwarding, routing, and traffic switching.
• MPLS overcomes limitations of IP-based forwarding for VPNs and is part of the packet switching network family.
• MPLS relies on packet labels instead of inspecting the packet itself to make forwarding choices.
• MPLS operates between the OSI data link and network layers, functioning as a Layer 2.5 networking protocol.
• It employs a label-based forwarding model.
• The technology is chosen for Indian Railways to support L2 and L3 services enabling IP-MPLS as the future transport technology.
• IP-MPLS meets the requirements of core, aggregation, and access networks, and it can operate on a common NMS for all three network components' OAM.
• Simplified WAN management and operation is achievable through IP-MPLS and SD-WAN by decoupling network hardware from its control system.
• The introduction of MPLS allows the forwarding to be done through hardware, increasing its speed compared to normal routing.
• MPLS paths are unidirectional, with forward and return paths that typically differ.
• To serve transport needs, congruent bidirectional paths are definable.

Technology and Terminology

• MPLS enhances performance and efficiency through an overlay technique not tied to a specific technology.
• A forwarding equivalence class (FEC) value is assigned when a packet enters the MPLS network, using a small label.
• The process negates the need for header analysis each time by indexing FEC.
• Predetermined paths are used for traffic routing based on the FEC label.
• Routers can use low-latency routes for applications, including live video streaming.
• A label is a part of the MPLS header, positioned between the data-link and IP headers, defining the path a packet takes comprising of:
• 20 bits for the label (label stack).
• 3 experimental bits for class of service.
• 1 bit reserved for the bottom of the stack.
• 8 bits for time-to-live (TTL).
• Label Forwarding Information Base: This table on label switch-capable devices shows how to forward frames using label values.
• Label Switched Path (LSP): Represents a unidirectional tunnel between routers within an MPLS network.
• Label Edge Router/Ingress Router (LER): Encapsulates packets within an MPLS LSP and selects the initial path.
• Label Switched Router (LSR): Performs MPLS switching within an LSP.
• Egress Router: Indicates the final router in an LSP, responsible for removing the label.
• Label switched: Relates to when an LSR uses a frame label to make forwarding choices.
• Label Switch Controller (LSC): The LSR communicates with ATM switch and provides label information within the switch.
• Label Distribution Protocol (LDP): Signalling protocol that distributes labels in an MPLS network.
• Enables Label Switched Routers (LSR) to establish Label Switched Paths (LSP) by mapping network layer routing data to data link layer switched paths.
• Through the Label Distribution Protocol, LSRs can collect, distribute, and release label binding information to other LSRs, enabling hop-by-hop packet delivery.
• Forwarding Equivalence Class (FEC): Refers to a group of forwarded IP packets that are equivalent in terms of forwarding which assigns a LSP.

MPLS Operation

• Operation is conducted through the control and data planes:
• Control Plane: Exchanges routing information and distributes labels between adjacent devices using protocols such as OSPF, IS-IS, and BGP, creating label bindings, then distributes the label-binding information among LSRs using LDP.
• Data Plane: Forwards packets using LFIB, which is managed by the control plane based on MPLS capable routers the destination IP address or label relying on a label based forwarding engine.
• Internal gateway protocols automatically creates routing tables using routers such as OSPF and IS-IS.
• MPLS capable routers rely on label distribution protocol to sets up mappings between labels and destination networks.
• Routing topology and tables are used for label values between adjacent devices creating pre-configured maps using Label Switching Paths between endpoints.
• Packets entering the ingress edge label switching router are assessed for layer-3 service needs like QoS and bandwidth management.
• The edge LSR then adds a label to the packet header before forwarding.
• The LSR reads the label, replaces it, and forwards according to the table.
• The Egress Edge Router removes the label, reads the packet header, and sends it to its final destination.

MPLS Services

• MPLS Traffic Engineering (MPLS-TE) balances traffic load, improving cost efficiency of IP Traffic.
• Traffic engineering is enabled by MPLS.
• MPLS TE uses an independent Label Switched Path (LSP) to carry traffic that differs from normal destination-based routing.
• IP networks have multiple pathways, but some become congested while others are underused.
• By supporting route optimization and avoiding packet loss, Voice over IP (VoIP) traffic relies on MPLS.
• Link utilization is upheld by MPLS.
• In order to ensure successful trasmission, Qos requirements are put on video traffic.
• Through subscriber's network policy devices in an IP network can intelligently prioritize traffic.
• QoS is used for bandwidth, delay, jitter, and packet loss controls.
• MPLS network ingress turns from Internet Protocol (IP) precedence information into Class of Service (CoS) bits or becoming an MPLS CoS value .
• IP QoS is different from MPLS QoS because IP QoS is based on IP precedence field and MPLS QoS relies on CoS bits.
• End-to-end IP QoS can happen across the network, with Differentiated services offered by MPLS CoS information,.
• Multiple customer sites can connect with the use of MPLS VPNs. via a Service Provider (SP) backbone instead of dedicated leased lines.
• Customer sites and the SP backbone have direct connections between them.
• SP offers VPN service with cost effective as they share backbone network resources between multiple customers.
• outsources planning, provisioning, and management using an SP.
• MPLS-enabled IP VPNs use connectionless IP networks, frame relay, and multiple IP service classes.
• MPLS VPNs are efficient and can classify traffic, without configuring complex meshes.
• MPLS networks offer privacy without tunnelling/encryption unlike VPN solutions because of the labels on a network-by-network-basis.
• MPLS offers supports services in a way that frame relay VPN does not with VPN awareness to switched or routed networks.
• Quick and cost effective for VPN with MPLS leveragingLSP tunnels.

Advantages and Disadvantages

• MPLS enhances bandwidth, flexibility, and performance.
• MPLS is Scalable:
• Ensuring high-performance telecomunication networks can be provided.
• Networks can be mainteded and engineered.
• MPLS grows network inter because MPLS facilitates growth of network inter-connectivity by using hardware.
• MPLS connects data centers with branch offices.
• New remote connections can occur with MPLS with cloud functions.
• By using MPLE the serivce provider handles WAN routing.
• Through MPLS's Quality of Service (QoS), latency-sensitive traffic like VoIP can be enabled.
• Multiple traffic types as Enhanced bandwidth.
• Alternative network options ensures uptime with MPLS.
• Network congestion is also reduced.
• Disadvantages:
• Total control is lacked because of what your service provider provides without total control of it.
• It's cost with Ethernet less than T1 however it's still expanse as if you're more advanced.

Implementation Considerations

• On Indian Railways where activities are managed, a Division is the main operational communications unit
• Circuits link Divisional HQ with stations, adjacent divisional HQ, Zonal HQ, and the internet gateway.
• Activities are monitored by the Divisional HQ.
• Circuits such as UTS/FOIS are centralized.
• Within the Division, services and applications like IP exchanges, VOIP, and application software are managed.
• Locating servers for division-related services and applications in Divisional HQs will help solve latency and bandwidth utilization issues.

Implementation Steps

• Replace legacy equipment.
• Train staff in divisions.
• Create monitoring for administration of the unified network.
• Zone HQ NOC capabilities include:
• Single point of contact for interdivisional and inter-zonal issues.
• Node installations, troubleshooting and updating for zonal nodes.
• Service provisioning for zonal nodes.
• Internet Policy Control.
• Reporting and recommendations for overall performance.
• Patch management and whitelisting.
• Backup management.
• Division HQ NOC capabilities include:
• Troubleshooting and updating.
• Field support.
• Node installations, troubleshooting and updating.
• Service provisioning.
• Performance reporting and improvement recommendations.
• Patch management and whitelisting.
• Backup management.