PDF of this tutorialThe next step toward building a service-driven metro network focuses on technology considerations within the context of the network architecture. In this step, service providers consider the pros and cons of different technologies, such as MPLS or service interworking, to deliver a specific service. During this step, service providers should determine how to network their existing installed bases of services with any new Ethernet WAN services. Some of the other important requirements to consider include QoS, network security, network availability, and redundancy needs.
Following are brief overviews of some of the many technology issues that providers should consider. Each of these technology deployment mechanisms is driven by the services, SLAs, and network architecture that the service provider has selected.
Service Interworking
Most service providers have a large installed base of ATM and Frame Relay services that they want to combine with new Ethernet WAN services. In this mixed environment, service interworking is critical to expanding the reach of Ethernet services.
Encapsulation of Ethernet into different frame formats is one part of service interworking, and several standards are responsible for Ethernet encapsulation. For example, RFC 2427 deals with Frame Relay, whereas RFC 2684 deals with ATM. Service interworking at Layer 2 can be complex, because each Layer 2 protocol has a different frame format and uses different address resolution mechanisms.
For example, consider different addressing schemes-Ethernet has Layer 2 source and destination addresses, while Frame Relay and ATM have only a destination address. And High-Level Data Link Control (HDLC) and Point-to-Point Protocol (PPP) have no addresses. Moreover, each Layer 2 protocol has different address resolution processes. The service interworking function must be able to mediate between the different address resolution mechanisms by "spoofing" the appropriate Address Resolution Protocols (ARPs) between different services, and translating between the different frame formats. Cisco provides numerous interworking solutions, using technologies such as routed bridge encapsulation (RBE) or integrated routing and bridging (IRB).
Security
Security is a primary consideration in any public switched network. Service providers need to be able to ensure that different customers on a common infrastructure cannot affect each other, and that denial-of-service (DoS) attacks or other malicious actions cannot interfere with SLA compliance.
In addition, service providers are likely to offer network security as a value-added service, protecting customers from security attacks. Most security attacks originate from within the customer's network-not from outside. Service providers can provide protection from attacks such as ARP spoofing, Dynamic Host Control Protocol (DHCP) attacks, and other internally-based threats.
Although Ethernet brings tremendous flexibility to the service portfolio, service providers should also consider any security changes that it requires. Traditional point-to-point WAN connections are easier to secure than multipoint-to-multipoint networks based on switched Ethernet technologies. Publicly available hacker software can enable users to exploit standard Ethernet switch mechanisms without any expert knowledge, so service providers should choose a solution that includes support for many robust security features.
Consistent, End-to-End QoS
QoS is another important technology consideration. Both service providers and customers implement QoS to achieve tighter bandwidth control and to implement stringent SLAs. QoS mechanisms can control traffic attributes such as delay, jitter, or packet loss to efficiently use network resources and ensure consistent application performance during congestion events. Effective QoS enables customers match their networks to their business needs, and helps service providers to differentiate themselves from competitors and to scale capacity more effectively.
In point-to-point connections, traffic attributes can be enforced at each network ingress point, and traffic profiles can easily be defined, implemented, controlled, and measured. Layer 3 devices, such as routers, can offer a rich set of QoS features.
For multipoint-to-multipoint services, a meaningful QoS definition is more complex, and service providers need to enforce traffic contracts both at the network ingress and egress.
QoS not only enables clearly defined SLAs, it also increases the transport efficiency of the network. Using intelligent packet processing together with QoS, service providers can oversubscribe their networks to make better use of their existing interfaces and bandwidth.
Redundancy and Availability
Redundancy and availability are important network requirements, and can be applied in several ways. Equipment redundancy can improve overall network availability, reducing the average time between failures. Using the appropriate network protocols, such as IP routing protocols, MPLS fast reroute, Rapid Spanning Tree, Sub-Network Connection Protection (SNCP), Multiplex Section Shared Protection Ring (MS-SPRing), or bidirectional line switched ring (BLSR) can also improve the availability of a network service. Because each protocol includes tools and capabilities to help ensure maximum network availability, protocols and protocol capabilities need to be carefully aligned with the projected service offerings.
