The concept of edge assurance represents a new approach to how telcos can overcome the challenges of today’s telecommunication networks. By decentralizing intelligence and harnessing the power of ML and AI, service assurance will become more efficient, responsive, and effective.
Nearly 200 million years ago, orb-web spiders began constructing the 2D webs that are still familiar today. Instead of proactively hunting for prey, spiders build a web, retreat, and wait for prey to become tangled in the web. Within the web, the insect struggles to escape, which results in vibrations that inform the spider that an insect has been captured. Further vibrations tell the spider of the insect’s exact location, prompting it to head in that direction to consume its prey.
Much like a spider that builds a web and waits for prey to come to it, the early days of telecommunications service assurance mandated that telecom operators and mobile network operators (MNOs) operate in a reactive mode. It was only after receiving an alarm (similar to vibrations within the web) that the data deposited to a centralized location was analyzed, and when warranted, a more comprehensive investigation of the data on edge (on probes) took place. Once this laborious and time-consuming process was complete, the issue was finally resolved. While initially, this multi-step approach delivered the desired outcomes, the exponential growth of telecom data and increased data speeds has led to a surge in User Plane traffic, resulting in significant data storage challenges and delays in the mean time to resolution (MTTR) of issues.
Telcos and MNOs are now struggling with numerous challenges, including insufficient data storage capacity or the budget to store the vast amounts of data needed for troubleshooting and analysis. Today’s complex data protection requirements add to these issues. Alleviating these challenges requires a new approach that necessitates a radical change in thinking. Coined the Octopus Model, it is being hailed as the future of both service assurance and network autonomy.
See also: Harnessing Real-time Data: Transforming Data Management with Artificial Intelligence
Why Octopus Assurance?
To understand Octopus Assurance, one needs to delve into the brain(s) of the octopus. Octopuses have both a central brain and a brachial plexus – a network of nerves within and between the arms – which serves as a second brain. While both brains can communicate with each other, they can also think and behave independently. For instance, when an octopus is hunting prey, its arms move in and out of various holes in the reef. When it encounters a potential food source, it doesn’t access the central brain to determine the best course of action. Instead, its brachial plexus feels, tastes, and becomes spatially aware, and the arm makes the decision – independently from the central brain. Essentially, the octopus’ arm makes an independent decision on the edge.
Using this octopus analogy, we can determine that centralizing data for troubleshooting use cases is inefficient and unnecessary. The concept of edge assurance requires systems to be capable of automating decisions closer to the root of the problem. However, independent, collaborative decision-making requires advanced artificial intelligence (AI) and machine learning (ML). Complex pattern recognition and a new form of communication between central assurance and edge assurance will be necessary for the success of telecommunication operators and MNOs using the Octopus Model. Leveraging a combination of technologies (AI, ML algorithms, real-time analytics, and edge computing) and innovative processes will enable issue detection, root cause analysis, and automated issue resolution to take place quickly.
Revolutionizing service assurance with edge agents
Edge assurance could revolutionize the current service assurance approach for more effective and efficient troubleshooting and resolution. Like the octopus’ arms, edge assurance requires agents (edge agents) located not within the core but closer to network boundaries.
Using near real-time data to validate their findings, edge agents proactively look for anomalies and issues on the network’s edge. Take, for example, the use of edge agents for a problem with call success ratio (CSR). Instead of being a slower centralized key performance indicator (KPI), an edge agent can quickly track where calls are failing in the radio access network (RAN). Suppose the CSR in the access network falls below a predefined threshold (AKA a rule). In that case, the edge agent can trigger an alarm or independently try to resolve the issue automatically.
Edge agents can handle even more complex issues. Imagine an edge agent tracking anomalous behavior that simple rules can’t resolve. In this scenario, the edge agent can cross reference what it has recorded about the behavior against a library of previous anomalous incidents using incident fingerprinting. If it finds a match, it can access the required automated resolution data, resolving the issue faster than waiting for the aggregation, mediation, and correlation process of centralized service assurance. Once the problem has been resolved, the edge agent will share its findings with other edge agents and crowdsource from them. This ensures that each agent shares the same level of intelligence. This data is referred to as evidential data.
The criticality of evidential data
As telecommunication operators and MNOs embrace varying levels of autonomous networks, evidential data will become increasingly indispensable. Resolution, automation, and orchestration systems will require evidence that a fix has been possible in the past. The evidence in question resides in two categories – 1) generic fixes and automations for generic issues, 2) operator-specific fixes and automations. While the evidence itself is incident-specific, it needs to, at a minimum, include:
- A fingerprint (description) of the incident that can be cross-referenced against an ongoing issue.
- A template for the fix, such as reboot, redirect, configuration update, etc.
- A reference that allows the action system to understand if the fix, once applied, has resolved the issue.
Will you remain a spider or become an octopus?
The concept of edge assurance represents a new approach to how telcos and MNOs can overcome the challenges of today’s telecommunication networks. By decentralizing intelligence like the octopus’s brachial plexus and harnessing the power of ML and AI, service assurance will become more efficient, responsive, and effective.
While this evolution will require retooling current practices, a commitment to innovation, and a willingness to embrace new technologies and methodologies, the future of service assurance lies not in centralizing data but in using edge assurance agents for better network performance. These innovations empower operators to manage and thrive amidst the data deluge, ensuring robust service assurance and enhanced customer satisfaction.
The question telcos and MNOs must ask themselves is, will you remain a spider, or will you evolve to be more proactive and nimbler, fixing issues faster and more autonomously with edge agents?
