In spite of a sluggish economy, the Carrier Ethernet market continues to be one of the fastest growing segments in the telecom industry. This is due, in part, to the wholesale of mobile backhaul. Across the world, networks are experiencing unprecedented mobile data traffic growth thanks to the ever-escalating availability of innovative mobile devices, and end-users increasingly asking for access to social- and business-related applications while they’re on the move.

For mobile operators looking to take advantage of this unprecedented growth in mobile data traffic, one domain that requires critical attention is backhaul infrastructure, where legacy technology is struggling to cope with the data influx. This has left mobile operators with no choice but to look at cost-efficient and scalable Carrier Ethernet as a means to combat the viral traffic growth. Nearly every operator aspiring to launch Long Term Evolution (LTE) services is first “preparing the ground” by deploying an Ethernet-based backhaul network.

But Carrier Ethernet presents difficult challenges not encountered previously by mobile operators. These challenges include ensuring continued operational abilities to assure transport to cell sites, and the need to guarantee the same end-to-end quality as traditional TDM. Other challenges include the necessity to right size Ethernet backhaul infrastructures, and the requirement to manage the complexities associated with class of service and MPLS traffic engineering.

To successfully meet these challenges and deliver optimized bandwidth and end-to-end availability while also ensuring quality of service, mobile operators, or their wholesale network infrastructure providers, need Ethernet-based proactive performance management and assurance. Several key considerations should be taken into account when considering management and reporting capabilities that address a multi-vendor Carrier Ethernet environment and mobile domain.

Pseudowire Monitoring and Live Troubleshooting
Although Carrier Ethernet is the emerging technology of choice for next-generation networks (4G), operators still have to support their legacy 2G and even 3G radio networks, which use legacy transport standards like TDM or ATM. Pseudowire or circuit emulation technologies are enabling operators to seamlessly support legacy transport interfaces over a pure Carrier Ethernet network, but transport connections to individual cell sites are still prone to multiple problems ranging from weather to careless configuration changes.

Pseudowire monitoring empowers service providers to track service performance over virtual tunnels (pseudowires), Ethernet virtual circuits, interfaces and VLANs, all of which are used to transport traditional technologies like ATM, frame relay and TDM, via Carrier Ethernet networks. The operations teams that traditionally managed the tools for testing legacy connectivity now need similar tools to assure the availability and quality of emulated connections over Carrier Ethernet. Additionally, the capability to troubleshoot transport to each cell site in real-time is vital for comprehensive service management.

End-to-End Quality of Service Assurance
In today’s fiercely competitive marketplace, operators who deliver poor service quality to their customers are extremely likely to experience churn. Proactive assurance of the mobile backhaul is critical in addressing this challenge. End-to-end quality degradations in the mobile backhaul lead to local market/regional issues that affect service experience for an entire population of mobile users. With operators looking to push their top-line revenue by introducing more and more value-added applications that are streaming-oriented (e.g., video) and real-time (e.g., gaming), stringent end-to-end quality guarantees are essential and demand monitoring and measurement across the multiple classes of traffic separation.

The need to assure data service experience as well as the compulsion to adhere to strict timing synchronization make it imperative to consider end-to-end quality assurance techniques even before rolling out the Ethernet backhaul network. The recognition of this by the industry has led to the introduction of end-to-end Ethernet assurance standards like IEEE 802.1ag and ITU-T Y.1731, or more generically, Ethernet OAM. Popular Carrier Ethernet infrastructure vendors like Cisco, Alcatel-Lucent, ADVA and others have followed suit by introducing standards-based, "built-in" instrumentation within their offerings to measure end-to-end quality of Carrier Ethernet.

Leveraging such instrumentation is inherently advantageous in that service providers can avoid the costs of deploying explicit probes (that may be expensive as well as intrusive) across the entire Radio Access Network (RAN) backhaul. Furthermore, if the Ethernet backhaul being deployed consists of multiple vendors, some of which may not be standards-compliant or have proprietary instrumentation, operators need to be able to leverage the vendor instrumentation in a generic manner. This clearly implies the need for effective service assurance platforms that can leverage vendor-specific instrumentation to portray the end-to-end quality of the Ethernet backhaul using vendor-agnostic key quality indicators.

Balancing With Analytics and Forecasts
Oversizing a Carrier Ethernet network infrastructure to meet growing traffic demands is counterproductive because it defeats the advantages Ethernet provides: more flexibility and scalability. On the other hand, under-provisioning would compromise end-to-end transport quality and cause other problems such as sloppy synchronization in the RAN. Therefore, rightsizing of components of the Carrier Ethernet backhaul infrastructure is necessary to realize the advantages of Carrier Ethernet, while also ensuring end-user quality of experience.

Rightsizing is complicated, however, by the extremely temporal data traffic patterns generated by mobile consumers. For example, holidays and events have a significant influence on consumers' messaging and browsing patterns. Mobile device usage dramatically increases in New York City's Times Square on New Year's Eve. Additionally, demographic factors cause variation in data traffic spatially -- there is more data traffic in some geographic regions than others. An accurate comprehension of these fluctuating traffic patterns requires deep analytics of traffic utilization. Industry-proven concepts such as "busy day" and "busy hour" designations are crucial in determining the worst stress levels of the backhaul network. Furthermore, hourly baselines, engineering benchmarks such as 95th percentile, and accurate traffic forecasts based on historical traffic usage, are indispensable to achieving the objective of rightsizing the network and planning maintenance and engineering activities intelligently. Being able to accomplish this down to every sub-element (interface, class of service, and VLAN) requires industry-grade assurance tools that demonstrate proven scalability and performance. With rightsizing being an ongoing effort, the ability to continuously baseline end-to-end quality over long periods of time becomes necessary to ensure that the chosen infrastructure sizing can meet service quality expectations.

Advanced MPLS Reports Provide Insight
Carrier Ethernet introduces complexities not faced with T1 backhaul. T1 brought only channelization, but Ethernet backhaul brings challenges such as the management of VLANs (for traffic separation), classes of service (for traffic prioritization), pseudowires, Ethernet virtual lines, Ethernet virtual LANs (for broadcast TV), and MPLS tunnels. The same transport-engineering and transport-operations teams that managed the more deterministic TDM technology are being entrusted with a far more sophisticated transport medium in Carrier Ethernet, and they need the right set of assurance tools to measure and report on performance in a holistic manner. Engineering an end-to-end connection between a cell site and the upstream controller using Carrier Ethernet requires an orchestration of all of the aforementioned entities, as a modification of any one entity requires a deep and ongoing analysis of the other entities from a performance perspective.

Mobile operators and wholesale backhaul infrastructure providers are embracing the more cost-efficient and scalable Carrier Ethernet to future-proof for explosive data growth and 4G mobile networks -- and to maintain operating margins and profitability.

At the same time, mobile operators have to support their legacy 2G and even 3G radio networks. The same transport team that manages TDM backhaul needs to be equipped with the right new tools that can operate, assure, and engineer Carrier Ethernet holistically. The inevitable transition from a stable and legacy technology to a next-generation IP/Ethernet infrastructure demands a much deeper comprehension of the new technology. It is critical to have the right Ethernet-based performance management to ensure the quality experience end-users demand from their mobile services in a cost-effective way.

About the Author
Ranga Thittai is product manager for InfoVista, a service assurance management software company. InfoVista empowers telecom operators, services providers, and large enterprises to maximize business return and productivity by assuring the optimal performance of mission-critical networked services, applications, and infrastructures. Ranga can be reached at rthittai@infovista.com. For more information, visit www.infovista.com.

What is your experience with this? Tell your fellow readers now!