Sparring Partners | Rethinking AIOps – It’s all about the data

In network operations, we are always looking for areas for improvement. How should we manage applications to best serve end customers? How can we use end-to-end visibility of application and service performance to end customers, anomaly detection for automated pattern discovery and prediction to improve quality of experience? How should we utilize software-driven networks to deliver new services and support monetization strategies?

In this Sparring Partners with Rick Fulwiler, Chief Solutions Architect at NETSCOUT, we discussed how AIOps is becoming the focal point driving the transformation to improve network operations across all lines of business—IT, enterprise, cable, security and wireless. However, all these efforts will fail without being able to feed rich and reliable network data to AIOps.

• Network data is hidden gold we need to mine to drive the journey from traditional operations to AIOps. How do we meet the challenges of extracting this gold for AIOps?
• The quality of data determines the quality of the analytics results. How do we curate the rich data collected at the source to make complex analytics successful at scale?
• How can scalable AI sensors at the source reduce data blot by feeding consumable, actionable data to 3rd-party AIOps?
• AI and ML are no silver bullets; every model has a specific behavior. How can we develop the right model for a task?
• How can we integrate the mined data into the AIOps ecosystem?

• 00:05 Introduction, AIOps, NETSCOUT
• 03:30 Rick introduces himself
• 04:18 What is AIOps and why it matters, learning from other industries
• 07:02 Are AI and AIOps ready to be deployed commercially? Use cases
• 10:47 Data requirements for AIOps, hidden gold to be mined
• 13:59 What data is relevant? How to curate data, how data requirements are changing with 5G standalone
• 19:50 Data and AIOps in a multivendor environment, role of standardization, KPIs definition
• 25:23 Best practices for data curation, advice from NETSCOUT
• 29:37 AI sensors to capture real-time packet-level data, use cases, NWDAF
• 39:37 More use cases, monetization, value creation
• 43:32 Business case, quantifying the benefits, APIs
• 46:24 Role of GenAI, telco-specific models
• 52:00 Why is NETSCOUT supporting AIOps, need for real-time data, interest and investment in AIOps
• 57:51 Reliability, what the future stores