Anticipation of eventual widespread IMS (IP Multimedia Subsystem) rollouts, and even the current migration to IP among many networks, are changing the way service providers, operators, and solution vendors approach the issue of QoS. Granted, there is no lack of skeptics who question the inevitability of IMS adoption, but it really has the potential to answer many demands of both consumers and service providers. As long as IMS is deployed properly, we can assume it will account for a large portion of the communications market. For those specialized “best-of-breed” vendors who intend to live up to that classification, now is the time to ensure that a solution is available for operators to guarantee premium user experience from Day One of their IMS rollout. In the process of developing an appropriate solution, it has become clear that the new network and OSS architecture is only one of many changes involved. Ensuring QoS over IMS requires not only new software, but a new strategy.
Before the IMS network structure even enters the picture, the first issue to deal with must be the new market conditions and business model, and the goals of all players involved. Consumers are essentially looking for variety and flexibility. Naturally, this entails more varied subscription plans and service bundles, and the ability to tailor these options to fit their own personal needs. Furthermore, consumers expect flexibility in network access - both in terms of method (mobile, wired, or FMC) and location (roaming ability) - while receiving the same QoS as on their home network.
For providers, IMS first and foremost provides the ability to answer the consumer demands listed above. By bringing standardizing and centralizing control for all communications services as well as access methods, service providers will have the ability to rapidly deploy new content and services to keep pace with demand. In addition, the IMS structure allows operators to accomplish this over a converged IP network without redundant “silo” infrastructure, improving overall network efficiency and reducing costs. More importantly, IMS can enable service providers to build a new business model. With margins from legacy services flattened through tight competition, providers can leverage IMS to leave the role of low-value voice commodity and become a more high-value, one-stopshop communications service provider.
However, providers must be careful in the way they pursue their IMS rollout. Dependable QoS is central to the IMS business model, and any shortfalls in that respect can bring major setbacks in IMS adoption. Currently, many providers are planning to initially focus almost exclusively on provision and billing technology, while viewing service assurance as a priority for later stages. It can be very tempting to many service providers to prioritize their resources in this way, with the goal of being able to begin service deployment and billing as soon as possible to grab early market share. If QoS is not ensured from initial rollout, consumers may be slower to sign on, or the high-value image of IMS may be tarnished, reducing it to another low-value commodity.
So, if service providers need to guarantee QoS from day one, how can this be accomplished? Many service assurance technologies and methods which are currently deployed over legacy networks are not sufficient for the needs of an IMS provider. First and foremost, service assurance for IMS must be a proactive and customer- centric operation. That is, the goal is no longer simply reducing repair time following service degradation or a lapse in network availability; if subscribers are paying for a highvalue service, they expect better. In order to achieve the full value of IMS, providers will need a strategy to implement automatic corrective action and prevent potential degradation before the user experience is affected.
One cornerstone of a proactive service assurance strategy which has recently seen increased adoption among operators is the concept of service impact analysis. Essentially, this analysis tool allows operators to take any actual or potential network event, and map forward its effects over various users or services. Once this information is available, the OSS can issue recommendations to preserve QoS among the affected users. This means that rather than simply correcting problems according to network element, they can be addressed in a way that will have maximum impact on user experience.
In fact, this service impact analysis has another implication, which also applies to silo networks as well as IMS: a more business-oriented approach to QoS. Granted that any operator must suffice with finite resources, the information provided in this fashion allows an operator to prioritize their network resources for maximum impact on revenue and reputation, in addition to overall user experience. For example, if an operator sees that a given network event users of different subscriber classes, it will be possible to dedicate resources to first guarantee that premium SLAs are upheld, and then to address other priorities in a more deliberate and controlled manner.
Another important aspect of service assurance for IMS networks is to provide a unified network view for “endto- end” service assurance. Under legacy silo networks, it was sufficient to simply monitor a series of disparate Element Management Systems or Network Management Systems, which would indirectly result in satisfactory service. However, the complexity of IMS networks would overwhelm this method, with too many variables contributing to the actual service quality. A preferable approach would be to manage operations across multiple domains under a single platform, and in particular to unite fault and performance management operations, thereby providing a more coherent picture of the network and connecting user experience more directly to network management.
A true end-to-end service assurance solution does more than just unite fault and performance management. It also captures a network view beyond its internal operations, providing data that reflects the real service quality as experienced by the end-user. In IMS networks, this can be accomplished by leveraging data contained in Diameter CDRs. Diameter CDRs were developed to provide data for both policy and charging purposes and therefore in IMS networks the service assurance system should be able to take advantage of this resource as well. In fact, even among currently deployed IP networks, the more advanced service assurance systems have managed to incorporate data from a variety of sources, such as SNMP agents on network resources, testing probes, EMSs and NMSs, and various xDRs, such as IPDRs and SIP CDRs, yielding positive results. In both cases, this data can be aggregated and correlated to measure any number of customizable KPI/KQIs, which directly reflect the user experience.
Any service assurance strategy for IMS networks must also include an increased emphasis on traffic monitoring, trend analysis and planning. Among legacy networks, the silo structure kept traffic simple enough that almost all network faults were hardware-related in their root-cause. However, as previously mentioned, one of the key benefits of IMS is the flexibility for subscribers to access and use their network services wherever and however they wish. This new level of flexibility, combined with the increased bandwidth demands of many services, magnifies the impact of traffic on overall network performance. Even in today’s pre-IMS environment, we already see that congestion accounts for a significant portion of service degradation.
Operators need a proactive strategy to maintain QoS in the midst of such complex traffic; retroactively managing the network to “unclog” congestion after service degradation is already felt will not be enough for the IMS subscriber. In this case, historical data on past traffic patterns must be analyzed through advanced correlation techniques in order to anticipate traffic- related faults before their effects reach the end user.
Finally, an important feature of any service assurance solution among IMS networks must be bidirectional mediation, which can almost give the effect of an automated Network Operations Center. Given the complex network operations that will fall under centralized control following IMS rollout, engineers’ manual corrections may be too slow, even if aided by OSS recommendations. Once the data from service- impact, traffic, fault or performance monitoring systems reveal potential service degradation, the ability for the service assurance OSS to initiate automatic corrective action can greatly enhance end-user experience. In fact, even if such corrective action is not yet necessary, such an automated solution could trigger complementary active service monitoring to pinpoint the malfunctioning phase within the whole IMS service delivery chain, adding to the proactive nature of the operation.
Premium service quality is an integral part of the user experience that consumers expect from IMS. A major incentive for service providers to deploy their own IMS networks is to offer a high-value, all-inclusive communications service. If concept of IMS is to truly succeed, and avoid becoming another commoditized bit-pipe, then it must be provided with a high enough quality to justify its high value. If service assurance is left for a later priority, providers may still deploy IMS networks, but its market share may be narrower or simply at a lower value. Only a proactive and customer- centric service assurance solution, incorporated in the initial stages of rollout, can uphold QoS at a level which will allow successful market adoption under a high-value framework for IMS networks.
Duby Yoely is VP Solutions Engineering, TTI Telecom. (news - alert) For more information, go to the company online at http://www.tti-telecom.com.