BLOG POST
Enhancing telecoms performance management with network and service data, enabled by DataOps
Telecoms performance management is an essential discipline, but legacy solutions will be challenged by future network and service evolution. How can we elevate telecoms performance management to meet the demands of new services and network architectures? By integrating network and service data, based on the innovations enabled by DevOps approaches to data management and ingestion.
Why should telecoms performance management evolve?
For network operators, telecoms performance management solutions have long been a valuable tool in their library of assets. They collect data and provide insightful statistics that help operators drive KPIs and optimize performance. However, they have typically accepted a limited range of source data, principally including 3 key inputs – OSS data, configuration data, and network inventory reference data.
All well and good — but today’s networks are not just evolving technically, towards cloud-native architectures based on 5G SA and soon, fixed 5G, they are also required to meet new expectations in terms of customer experience.
This means that legacy telecoms performance management solutions may be left behind, unable to deliver the insights and reporting required. In particular, they may not be able to deliver a true end-to-end view of services and the ways in which they are experienced, as services can span solutions that are outside the traditional OSS / BSS domains.
Further, such systems may also not be accessible by — or comprehensible to — the different users and roles (which we can define as personas) that need to access or to discover insights from the network and its assets, and the services it delivers.
So, we face new challenges and demands. In this context, let’s consider how performance management solutions should evolve to address them.
A growing range of data sources is essential
Well, first we need to consider other sources of data – both those available today and those to which we might have access in the future. We need a way to elegantly integrate – and sort – these additional data, so they can be made available for processing, interrogation and, ultimately, to deliver relevant insights to the right stakeholder personas.
Second, we need to consider what current data is missing from telecoms performance management solutions. One clear gap is real-time network and service data from the interfaces within the network – not simply output from OSS systems, for example, but also including transactions and signaling events from across control and media plane interfaces in the 5G core, RAN, IMS, 4G network, transmission layers, and so on.
Another gap is a view of the subscriber that is also included within control and media plane data. Network probes see information regarding subscribers in terms of their MSISDNs and other session information that uniquely relates to them and the services they consume (this video streaming session, at that time, for that user, with this quality). Performance management data cannot drill down to that level of information because data is aggregated.
In other words, we need to bring 5G and network performance monitoring data from probes into telecoms performance management solutions, so that we can combine it with traditional data sources, extending our focus.
Integrating new data into telecoms performance management
How can we integrate new 5G and other data into performance management solutions? The answer is via the adoption of DataOps processing. DataOps is a means of facilitating “the integration and automation of data flows between data managers and data consumers.” In other words, it provides a way to bring together both current and new sources of data in a way that makes it accessible and consumable.
That’s the approach we’ve taken at Elisa Polystar. By leveraging innovations in DataOps, we’ve combined our expertise in network and service performance monitoring to empower a newly enriched series of telecoms performance management solutions.
Our solution spans both data management and consumption, which means it caters for different kinds of user profiles — from technical to marketing, from customer service to planning — via portals such as those available from KALIX, that are tailored to the needs of different user communities.
As a result, users can leverage both the built-in telecoms expertise, combined with performance management data, and intuitive interfaces that guide them to obtain valuable insights. As a unified solution, it blends all data sources, spanning network domains, OSS and BSS layers – enabling correlation to deliver the end-to-end view today’s networks demand.
However, it is the level of data integration and ingestion that transforms the discipline of telecoms performance management. The truth is, we can speculate about future sources of data that may further enhance performance management and KPI reporting capabilities, but we don’t yet really know what these data sources might be.
They could come from external partners – a streaming content provider, a cloud hosting platform, or a 5G-enabled factory, perhaps – but we’re in the early days of extending our view to such sources. We can only definitively say that we know we need to extend data processing capabilities in novel directions.
How do we unlock the power of DevOps?
But we can prepare for this shift. Whereas data transfer in legacy telecoms performance management solutions has typically depended on file import (for a limited range of formats), network monitoring solutions have typically also leveraged real-time ingestion from active network interfaces. A combination of all of the above is now necessary – with an expanded range of formats to allow extension to future data sources.
That’s where DataOps really comes into play, because it provides a flexible architecture to accommodate any data source and format that can be extended as required in an agile manner. So, we can realize adaptive data pipelines and database structures to respond seamlessly to changes in input data – and its diversity.
And, we can go further. With the addition of open APIs, the totality of the captured network, service and performance management data can easily be exported from the solution to support other activities – such as automation and closed-loop operations.
Conclusion – towards flexible telecoms performance management
We already know that quality data is the foundation of telecom performance management. We also now know that we need to enhance this with real-time network and service data, to deliver an enriched view. However, we must also recognize that other data sources and inputs can enhance our understanding of performance and provide contextual information to support persona-driven analytics and insights discovery, and to fuel new levels of processing for automation.
By making this step-change, we can deliver telecoms performance management that’s adapted for the next generation of services – and equip our teams with the tools they need to make sense of increasingly complex interactions and flows, so they can deliver the experience and quality demanded, while enabling continuous optimization and new levels of automation.
