From data centre to cell tower
Whether it’s reliable, uninterrupted voice connectivity, or real-time access to content and data, smartphone and wireless PC users expect the utmost quality from their mobile operators. This is why ensuring high-quality, always available wireless voice and data services from one end of the network to the other – and back again – is a must, writes Marc Lippe.
With every dropped call, dropped packet, instance of jitter or failed download, a mobile service provider risks losing subscribers and high-value business customers to the competition. But delivering that level of unquestionable, end-to-end service quality is easier said than done due to the complexity and cost of managing today’s mobile data infrastructure and the explosion of data traffic crossing it. An inundated router, a downed line or a crashed server can leave hundreds or thousands of subscribers without service, while leaving the service provider vulnerable to disgruntled customers, churn and lost revenue. For these reasons, mobile operators must be able to quickly identify network trouble spots and the effect of bandwidth-hogging applications— before customers are impacted. They must do so in an increasingly complex delivery environment that consists of data centre, IP/MPLS, Ethernet backhaul mobile packet core, legacy voice core, mixed generation RAN and 4G technologies— from multiple equipment vendors, while keeping network and service operations costs as low as possible. To overcome these critical performance, quality and cost concerns, mobile service provider must adopt a unified application and network performance assurance approach that will enable them to address the underlying challenges of mobile data service delivery, the management of the IP transport architecture and the proliferation of service assurance tools. A key aspect often overlooked is that the actual transport of that data is done by the underlying IP wireline infrastructure rather than the wireless network facilities. By bringing together the quality of experience (QoE) perspectives from these two domains, providers can comprehensively assure end-to-end quality.
Mobile data service delivery challenges
Data services represent mobile service providers’ fastest growing revenue stream, accounting for up to 50% of growth compared with the fairly flat revenue growth from voice services. Simultaneously, the rapidly growing volume of bandwidth-intensive mobile data traffic has made it difficult for operators to ensure quality and availability on these data services. Service providers must be able to:
• Differentiate themselves from competitors by assuring a quality end-user experience on data services and applications
• Build trust with and retain high-value business customers, especially where service-level agreements are critical
• Hold third-party application providers andinternet service providers accountable for the quality of the applications travelling over the network
All-IP transport architecture challenges
The data traffic explosion has led operators to increasingly recognise the importance of IP/Ethernet technology as a cost-effective transport medium and place a significant amount of emphasis, effort and investment on the transport network as a network asset. This is best seen in the mobile backhaul domain where many mobile operators are using or deploying Carrier Ethernet. The all-IP network infrastructure provides the ease of deployment and bandwidth to deliver both packet based voice and data services over a common transport backbone while reducing the operators’ growing infrastructure costs. But with the dawning of this all-IP network infrastructure, a host of management challenges have emerged. These include:
• Providing the same quality of service/quality of experience that users are accustomed to receiving over legacy transport network technologies
• Equipping transport engineering and operations groups with the same level of operational and engineering capabilities they had with TDM
• Meeting the bandwidth demands of mobile data and application users without oversizing and overspending on network infrastructure
• Monitoring performance and capacity across the entire mobile network infrastructure, including the backhaul network, regardless of its multi-generation, multi-vendor, multidomain composition
The challenge of proliferating service assurance tools
The majority of today’s service assurance tools are designed to tackle either per-subscriber service monitoring, mobile infrastructure monitoring or transport monitoring, but not all three at once. This has resulted in a proliferation of tools and made assuring the end-user experience an increasingly difficult and expensive proposition. Hence, IT departments have struggled to:
• Manage down both capital and operational expenses
• Employ a single, vendor-agnostic toolset that can address both service and network management requirements, including transport monitoring, from end to end.
Delivering high-performance mobile data services
Mobile service providers need to be equipped with the reporting capabilities they need to offer service level guarantees. For mobile service providers, the biggest market is the consumer market, and ensuring that services are always available and meet end-user quality expectations is critical to avoiding churn. But as business from wireless VPNs and machine-to-machine (M2M) offerings increase, mobile operators must also protect their high-value enterprise accounts by delivering the level of service these corporate customers are paying for and by providing their customers with customised dashboards depicting such compliance. On the other side of the coin, mobile operators have an obligation to their subscribers to ensure that contracted application partners are delivering content of the highest quality. Subscribers look to their mobile provider to troubleshoot and deal with quality issues, even if the problem originates from a third-party content provider. Further complicating quality assuranceis that the infrastructure between the mobile operator and its third-party content providers is typically a leased ISP connection that the operator has no control over. Application deep packet inspection (DPI) offerings will help mobile operators remotely monitor the performance of applications coming from content partners, as well as further sectionalise the views into TCP delays and actual application response delays. Mobile service providers will then be able to distinguish between ISP connectivity issues and poor remote-application performance. A single, integrated network, service and application service assurance solution will allow service providers’ IT departments to manage end to-end mobile network and service quality in real time and assure the performance of data services and applications on a per subscriber basis, without the cost and delays associated with having to jump between OSS toolsets. Monitoring and reporting of every infrastructure entity along the service delivery path, from data centre to mobile packet core, to the IP/MPLS backbone, radio access network and the Ethernet backhaul, will dispense with the siloed approach of tools that can report on either network or service performance only. A unified service assurance solution is also to deliver direct business benefits, such as providing insight into new service opportunities, with a view into the types of applications each subscriber is using by offering DPI between the SGSN and GGSN nodes and in the data centre. This enables the service provider to not only more readily pinpoint problems but also understand subscriber activity and preferences. Thus, if the operator identifies a high level of requests for a specific application coming from the student area of a particular city, it can follow up by tailoring a product it can market to that segment and demographic. With this comprehensive and actionable visibility into the health of the network infrastructure and the services and applications traversing it, engineers and operations personnel can proactively ensure availability and consistent performance on the growing scope of bandwidth-hungry applications and services they deliver to customers, wherever those users may be. By immediately understanding the relationships between resources, the services they support and the respective performance indicators, they will be able proactively identify and focus on the network root causes of service degradation before customers are negatively impacted, dedicate bandwidth, and analyse and prioritise application traffic in a way that ensures a quality end-user experience.
