Utpal Mangla: VP & Senior Partner; Telco, Media, Entertainment Industry, IBM

Mathews Thomas: Distinguished Engineer; Telco, Media, Entertainment Industry, IBM

5G with edge computing is an emerging area which will revolutionize and transform various industries. The rapidly increasing number of edge nodes, management of sophisticated workloads running AI, variability of edge nodes, distribution of content at the edge and security implications provides challenges to creating end-to-end solutions. The diagram below provides an overview of the edge ecosystem.

5G and Edge Landscape

Some key components that form the edge ecosystem are the following:

  • Cloud: This could be a public or private cloud, which can be a repository for the media container-based workloads including applications and machine learning models.
  • Network Edge: This is generally part of the Communication Service Providers (CSP) core network which can host larger edge applications and data.
  • Edge cluster/gateway: An edge cluster/gateway is a multi-edge compute node that is in a remote operations facility such as a factory, retail store, hotel, distribution center, or bank.
  • Edge device: An edge device is a special-purpose piece of equipment that also has compute capacity integrated into the d

Troubleshooting, root-cause analysis and management of such a landscape is complex especially given the number of nodes involved. The  explosion of data available and the need to automate the process with minimal human intervention.

Closed-loop automation systems enable companies to transform network and IT operations by using AI-driven automation to detect anomalies, determine resolution, and implement the required changes within a continuous highly automated framework. Closed-loop automation helps solve many problems before they even become issues in a 5G Edge environment and many CSP’s are beginning to integrate it into their 5G Edge deployment.

There are different types of closed loop systems. A simple closed-loop implementation detects issues that could happen in the future. The appropriate data is analyzed by various predictive models, which then make a recommendation on the change to be made to the orchestration layer, which implements the change. In complex cases, closed-loop automation combines the predictive insights information with additional AI systems to determine a resolution which is then implemented by a Robotic Process Automation (RPA) system. The following diagram provides an overview of a closed-loop automation system that addresses issues of varying complexity.

Closed-loop automation ensure 5G Edge implementations function properly with minimal human interventions. Data is collected from the different edge nodes and anomalies are detected.  The data usually consists of large, real-time, time-series data to analyze networks applications, database metrics, operating systems, etc. This gives anomaly detection the capability to identify patterns and raise awareness towards appropriate actions. Machine learning models are used create the patterns for the series of alerts so that those can be bound to causes and known actions, and then be corrected accordingly. The machine learning algorithms also predict how application and network behaviors are dependent on seasonality and other factors to ensure that appropriate corrective actions are taken, thereby permitting systems to perform optimal. It makes use of various AI algorithms to ensure the accuracy of root-cause identifications and implements the required remediation steps.  The remediation includes invoking an RPA system which is integrated with an AI system which has been trained to resolve the identified anomaly. If the AI system determines it has a high confidence that the planned resolution is correct, the RPA will invoke a system such as an orchestration engine to implement the solution automatically. If not, a trouble ticket is generated, and an engineer works to resolve the issue.

In summary, edge computing built on 5G will soon be mainstream in solutions across many industries.  It is, however, a very complex environment to manage given the many systems involved and the need to automate the management of the environment as much as possible.  Close Loop Automation built on AI is an important step to address this issue and is being adopted by service providers, IT companies and research organizations to address this issue.

About the Authors

Utpal Mangla ( MBA, PEng, CMC, ITCP, PMP, ITIL, CSM ) is a Vice President and Senior Partner in IBM. He is the Global Leader of IBM’s Telecommunications, Media and  Entertainment (TME) Industry’s Center of Competency. In addition, he leads the ‘Innovation Competency’ focusing on AI, 5G EDGE, Hybrid Cloud and Blockchain Innovations for TME clients worldwide. In his role as senior executive in  the business and thought leader in emerging technologies, Utpal’s mission is  to fuel growth by building, selling and implementing differentiated competitive market service solution offerings  to meeting critical business imperatives of our customers.

Mathews Thomas is a Distinguished Engineer at IBM’s Telecom and Media labs. With 20+ years in consulting, system integration and industry experience working with many of the major Telecom and Media companies, Mathews brings in-depth expertise and knowledge in defining, building and running strategic IT projects.  He works with clients, partners, and standards organizations to develop innovative solutions built on IBM’s AI, analytics, blockchain and cloud platforms with current focus on 5G and edge computing. He holds over 50 patents, has over 40 publications and has presented at over 50 conferences in above areas..