HN800: Root Cause Analysis for the Entire Stack (Sponsored)

In this episode of Heavy Networking, Ethan and Banks host John Capobianco and Vreesha Sriram from Selector to discuss comprehensive root cause analysis across the full IT stack. They emphasize that Selector's approach is not a one-size-fits-all software but a customized solution tailored to each client's environment through workshops and deep engagement with their network architecture and operations teams. The discussion covers how Selector uses an ensemble of data-centric AI models, including LLMs, regression, and clustering models, to ingest, normalize, and correlate data from various sources, providing actionable insights and reducing alert fatigue. They also explore real-world examples of ticket reduction, Kubernetes cluster monitoring, and integration with automation platforms, highlighting the ROI and the shift towards natural language-based interactions for network operations.

Outlines

Sign in to continue reading, translating and more.

Continue

Heavy Networking

Introduction to Full-Stack Root Cause Analysis with Selector

Meta Tags, AI Models, and the Network Language Model

Engagement Stages and Pre-Day Zero Workshops

Embedding with Customer Teams and Kubernetes Monitoring Example

Distinguishing Network Issues from Application and Infrastructure Problems

Data Ingestion, Model Location, and Tuning

Data Normalization, Team Engagement, and Time to Production

Interacting with the Platform and Adding New Data Sources

Data Volume, Licensing, and Root Cause Analysis

Integration with Automation Workflows and Dashboard Summarization

ROI, AI Education, and Closing Remarks

HN800: Root Cause Analysis for the Entire Stack (Sponsored)

Heavy Networking

00:00Introduction to Full-Stack Root Cause Analysis with Selector

Introduction to Full-Stack Root Cause Analysis with Selector

04:05Meta Tags, AI Models, and the Network Language Model

Meta Tags, AI Models, and the Network Language Model

07:47Engagement Stages and Pre-Day Zero Workshops

Engagement Stages and Pre-Day Zero Workshops

12:29Embedding with Customer Teams and Kubernetes Monitoring Example

Embedding with Customer Teams and Kubernetes Monitoring Example

17:52Distinguishing Network Issues from Application and Infrastructure Problems

Distinguishing Network Issues from Application and Infrastructure Problems

25:23Data Ingestion, Model Location, and Tuning

Data Ingestion, Model Location, and Tuning

33:04Data Normalization, Team Engagement, and Time to Production

Data Normalization, Team Engagement, and Time to Production

39:30Interacting with the Platform and Adding New Data Sources

Interacting with the Platform and Adding New Data Sources

47:52Data Volume, Licensing, and Root Cause Analysis

Data Volume, Licensing, and Root Cause Analysis

55:36Integration with Automation Workflows and Dashboard Summarization

Integration with Automation Workflows and Dashboard Summarization

1:02:04ROI, AI Education, and Closing Remarks

ROI, AI Education, and Closing Remarks