Chapter 10. Business Opportunities
While the previous chapters provide detailed technical information about segment routing (SR), this chapter explores the business opportunities that emerge from adopting this innovative technology. No company will invest in a technology if its business cannot benefit in some way. This chapter discusses various topics to help build a bridge between engineering, operations, marketing, finance, and product management. It can help stakeholders and leadership identify relevant benefits in their areas, effectively communicate, and justify the investment in a transformation where silos are torn down and barriers are identified and addressed in a timely manner for a successful transition to SR.
Networks and associated IT systems are implemented differently across network service providers and large enterprises, tailored to offer the desired services to internal and external customers. Likewise, network service providers adhere to different standards and frameworks, and each organization’s history and decisions influence to what extent standards and frameworks are adopted. We appreciate that organizational structures, processes, terminologies, services, and other aspects vary significantly across network service providers, and this chapter attempts to be descriptive without using any specific standard, framework, or terminology.
Echoing the insights from earlier technical chapters—such as Chapter 2, “What Is Segment Routing over MPLS (SR-MPLS)?” and Chapter 3, “What Is Segment Routing over IPv6 (SRv6)?” —this chapter highlights substantial opportunities and advantages of SRv6 compared to SR-MPLS.
The introduction of SRv6 technology presents a unique opportunity for network service providers. By harnessing its full potential, providers can gain a significant market edge or at least ensure that they remain competitive with their network service offerings over the coming decade. The improved performance, greater scale, network simplification, and new service options associated with SRv6 allow for the convergence of multiple networks into one. This consolidation ideally results in just one network to purchase, build, operate, support, power, cool, and host, thereby leading to substantial reductions in capital expenditures (CapEx) and operational expenditures (OpEx). The benefits extend to potential organizational optimizations, avoided costs for redundancy and scale-related spare capacity in multiple networks, simpler and improved service-level management (SLM), fewer integration points, and the convergence of operation support systems (OSS), business support systems (BSS), and IT systems in general. These factors can multiply the business opportunity related to the introduction of SRv6.
However, the advantages of SRv6 are not limited to consolidation and optimization. You may recall the transition from Asynchronous Transfer Mode (ATM)–based services to IP technology and its profound market impact two to three decades ago. Similarly, SR protocol options enable the transportation of new services over IP. Leased line and optical point-to-point services can now be offered over IP, reducing the need for optical network-based services to be exposed to customers. This shift simplifies the optical network stack, reducing the requirements for its OSS, BSS, and related IT systems to implement, test, support, and maintain. Instead, the IP services stack can incorporate these as additional service flavors, potentially reducing the cost of offering traditional leased line or optical services and enabling providers to offer these services at more competitive prices. Network service providers currently relying on third-party leased line or optical point-to-point services may even consider offering such services themselves, using their SR network. SRv6 allows for simplified chaining of network connectivity services with additional services such as Network Address Translation (NAT), firewalls, deep packet inspection (DPI), intrusion prevention systems (IPSs), and services offered on virtual machines or containers within data centers or clouds, reducing complexity and costs.
Network service providers operating single IP networks and requiring relatively simple IP or VPN connectivity services can also benefit from the introduction of SRv6. Although traditional MPLS VPN transport networks have been around for two decades, the networking industry is less likely to invest in significant developments or address known limitations due to its maturity, and so the transition to SRv6 is the next logical step. With device generations increasingly focusing their feature support around SRv6 and reducing legacy features to remove complexity and costs, it's worth evaluating investment in SRv6 for the coming decade.
The introduction of SRv6 represents a tremendous opportunity for network service providers. Those who thoroughly analyze and leverage its full potential can undergo a true transformation, benefiting their business and customers for years to come. A detailed analysis may even reveal the business benefits associated with SRv6 as justification for an early network lifecycle. The following sections delve into how simplification, convergence, and standardization can lead to new business and enrich existing services.
Technological Opportunities and Benefits
Before we dive into CapEx and OpEx opportunities, this section provides background information on selected opportunities. Some of these opportunities are not directly related to SR as a technology but rather to the fact that a new network may be built, or a new technology may be introduced.
Fewer Protocols
Figure 10-1 shows network technology–specific protocol stacks (though it omits protocols that are in common across all technologies).
)
Figure 10.1 Comparison of MPLS, SR-MPLS, and SRv6 Protocols
Legacy MPLS networks require around six (6) specific protocols to provide VPN services. Label Distribution Protocol (LDP) is used to exchange labels for endpoints participating in L2VPN services, and Multi-Protocol BGP (MP-BGP) handles the exchange of L3VPN services–related network information. Transport across networks or network domains is facilitated by another BGP flavor, called Border Gateway Protocol Labeled Unicast (BGP-LU). L2VPN and L3VPN services with bandwidth commitments rely on the capabilities of Resource Reservation Protocol Traffic Engineering (RSVP-TE), which reserves bandwidth on all links along a desired traffic path. Interior gateway protocols (IGPs) such as Intermediate System-to-Intermediate System (IS-IS) and Open Shortest Path First (OSPF) are used to exchange topology information within the transport network. While Multiprotocol Label Switching (MPLS) is used to transport VPN traffic, most control protocols leverage pure IP transport.
SR-MPLS–based networks have a smaller protocol stack (4). LDP is not required on the services layer in next-generation networks because MP-BGP (EVPN) can be used to provide L2VPN services. SR-MPLS–specific IGP extensions serve traffic engineering, fast reroute, and intra-domain paths; in legacy MPLS networks, these functions were provided through LDP and RSVP-TE. Optionally, IGP extension can be leveraged for inter-domain routing to replace BGP-LU—for instance, using static or controller-based SR-TE policies.
The protocol stack of SRv6-based networks is significantly simpler, with only three protocols (3) required to provide VPN services. MP-BGP exchanges information related to Layer 2 and Layer 3 VPN services across all participating endpoints. No BGP-LU is required for inter-domain routing. The IPv6 data plane and IPv6 segment routing header (SRH) allow for the interconnection of networks through the use of route summarization and traffic engineering via additional segments in the IPv6 and SRH packet headers. An important operational advantage of deploying SRv6 in a network, as opposed to MPLS, lies in the elimination of label stacks. This move prevents label stack overflows in P nodes and enhances the scalability of PE nodes, thereby mitigating the operational risks involved with scaling up existing network segments and integrating new ones.
Fewer protocols also means the following:
Less code running
Fewer configurations to create and change
Less knowledge required to build and operate a network
Fewer potential defects and interoperability issues
Therefore, fewer protocols can positively influence network stability, streamline troubleshooting, and lead to shorter mean time to repair (MTTR) when incidents occur.
Network service providers implementing a multi-vendor strategy in their network can benefit from fewer interoperability issues because fewer protocols are required to peer between different vendors. However, SRv6 is early in its lifecycle, and related standards are maturing still. Vendors have made differing levels of progress in implementing SRv6 across platforms and potentially may have interpreted standards slightly differently. Consequently, network service providers implementing a multi-vendor SRv6 network need to carefully assess the status of the preferred vendors, compare available features, and plan a thorough integration test phase.