Networks unchained: the shift toward intent-based autonomous operations

The telecommunications industry, which plays a crucial role in global connectivity, has been experiencing a technological revolution driven by innovations like 5G, IoT, cloud computing, and AI. However, managing these networks has become increasingly challenging. Automation is needed to handle routine tasks, monitor network health, and respond to issues in real-time. Unfortunately, the skill sets of communication service providers (CSPs) may not align with the evolving demands of this landscape. To succeed, CSPs require versatile teams, including data scientists, software developers, and service assurance engineers. Fortunately, programming languages have evolved to low-code/no-code paradigms, and generative AI can generate formal code based on natural language descriptions of tasks. This has given rise to intent-based networking (IBN), where human administrators express network objectives in natural language, which are then translated into network policies and configurations. IBN has the potential to address the talent gap within telcos and improve network management. Autonomous networks (AN) take IBN a step further by self-configuring, self-optimizing, and self-healing networks based on human intents. However, there are concerns about the feasibility and application of IBN and AN, including intent expression, accurate translation into network configuration, system transparency, and complexity. In this blog, we explore the potential and challenges of IBN and AN in practical applications. We use the example of deploying a new service to illustrate the need for streamlining interactions between CSP teams and the network. We introduce the concept of intents, which are high-level objectives that CSPs want to achieve in their networks. Instead of dealing with complex network configurations, engineering teams express objectives with intents, which are then translated into the required network configurations. IBN and AN have the potential to revolutionize various aspects of network management, including service preparation, introduction, service ordering, feasibility checks, and dynamic service assurance. However, there are challenges in expressing and conveying intents, as well as executing on intents. TM Forum has introduced the TMF921 Intent-based Networking API to define high-level network intents, but there is a concern that network engineers would need to familiarize themselves with a formal language to fully utilize intents. In addition, formalizing intent specifications may not reduce the number of parameters required. The intent handler, the core component of IBN, also needs to evolve into an autonomous system to transition from automated to autonomous operation successfully. In future blogs, we will delve deeper into the challenges and opportunities of IBN and AN.



Source link