Overview
Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore AI-driven observability solutions for massive AWS serverless workflows in this 17-minute conference talk from Conf42 Observability 2025. Discover how to implement adaptive parallelism insights to optimize serverless fan-out patterns and overcome common challenges in large-scale distributed systems. Learn about the technical implementation using telemetry data and graph neural networks to create intelligent monitoring solutions that can dynamically adjust to workflow demands. Examine real-world evaluation metrics and performance benchmarks that demonstrate the effectiveness of AI-powered observability approaches. Gain practical insights from lessons learned and best practices for implementing these solutions in production environments. Understand the current limitations of AI-driven observability tools and explore future research directions for advancing serverless workflow monitoring. Access comprehensive coverage of implementation workflows, from initial setup through deployment and optimization strategies for massive serverless architectures.
Syllabus
00:00 Introduction and Welcome
00:12 Understanding Adaptive Parallelism Insights
00:46 Challenges in Serverless Fan-Outs
01:28 Guiding Questions for Adaptive Parallelism
02:04 Technical Deep Dive: Telemetry and Graph Neural Networks
03:22 Implementation and Workflow
06:52 Evaluation and Metrics
08:21 Lessons Learned and Best Practices
09:26 Limitations and Future Work
10:17 Conclusion and Next Steps
10:50 Q&A and Closing Remarks
Taught by
Conf42