USENIX Courses

Cloud, Kubernetes, and Service Networking - Taming the Turtles

Demystify Kubernetes networking, exploring CNI, kube-proxy, VPCs, and service meshes. Gain insights into eBPF usage and learn to debug complex networking issues in cloud-native environments.

• YouTube
• 40 minutes
• On-Demand
• Free Video

A Dual Approach to Accountability Engineering in SRE

Explore strategies for effective SRE team structuring, focusing on agency and accountability in microservices environments. Learn to balance philosophical and pragmatic approaches for optimal performance.

• YouTube
• 22 minutes
• On-Demand
• Free Video

Reliable Data for Large ML Models: Principles and Practices

Explore best practices for ensuring reliable data in large ML models, focusing on challenges in LLMs and applying SRE principles to manage data risks in production systems.

• YouTube
• 31 minutes
• On-Demand
• Free Video

How to Use Prometheus's Native Histograms

Explore Prometheus's native histograms for improved SRE tooling. Learn practical usage, SLO tracking, and troubleshooting techniques to enhance your monitoring capabilities.

• YouTube
• 36 minutes
• On-Demand
• Free Video

Continuous Profiling in the Cloud-Native Era

Explore continuous profiling techniques to optimize resource usage, reduce cloud costs, and enhance software performance across multiple languages in cloud-native environments.

• YouTube
• 36 minutes
• On-Demand
• Free Video

Overcoming Challenges in Serving Large Language Models - SREcon23 Europe/Middle East/Africa

Explore hosting GPT models in Kubernetes, covering GPU sharding, tensor parallelism, and model optimization. Learn trade-offs between latency, accuracy, and resource allocation, with a live demo showcasing performance.

• YouTube
• 31 minutes
• On-Demand
• Free Video

How We Managed a Massive-scale Incident at Datadog

Explore Datadog's response to a massive global outage, including technical challenges, solutions, and incident management strategies for coordinating 500+ engineers over 2+ days.

• YouTube
• 40 minutes
• On-Demand
• Free Video

Building a 5-Exaflop Supercomputer for Meta-AI Research and Large-Scale Model Training

Explore the architecture, construction, and operation of Meta's 5-exaflop AI Research SuperCluster, supporting large-scale model training with insights on monitoring and observability for cutting-edge AI development.

• YouTube
• 35 minutes
• On-Demand
• Free Video

Dark Matter and Deep State: The Unseen Majority of Everything

Explore the importance of noticing the ordinary in diverse fields like government, astronomy, and SRE. Learn to identify automation opportunities and balance human intervention in predictable scenarios.

• YouTube
• 38 minutes
• On-Demand
• Free Video

When One Line Took Thousands of Websites Offline - Lessons from a Kubernetes Incident

Explore a critical incident where a single configuration change caused widespread website outages, examining the recovery process, lessons learned, and subsequent improvements in Kubernetes infrastructure.

• YouTube
• 27 minutes
• On-Demand
• Free Video

Symptom-based Alerting for Machine Learning - Lessons from 30+ Use Cases

Practical guide for ML monitoring: essential metrics, prioritization, and tool selection. Insights from 30+ ML use cases to detect issues in real-time and overcome traditional software monitoring limitations.

• YouTube
• 25 minutes
• On-Demand
• Free Video

Deconstructing an Abstraction to Reconstruct a Database Outage

Explore a complex Postgres cluster outage, learn to reproduce failures locally, and gain insights into debugging complex systems and understanding database internals.

• YouTube
• 33 minutes
• On-Demand
• Free Video

Do Not Thrash the Node.js Event Loop - Deploying at Scale

Discover strategies to optimize Node.js deployments, prevent resource exhaustion, and improve system efficiency. Learn mathematical insights and practical solutions for scaling Node.js applications effectively.

• YouTube
• 30 minutes
• On-Demand
• Free Video

Cache Me If You Can: How Grafana Labs Scaled Up Their Memcached 42x and Improved Reliability

Discover how Grafana Labs scaled their memcached 42x, improving reliability and cost-effectiveness for large-scale data querying using cloud-based local SSDs and innovative caching strategies.

• YouTube
• 26 minutes
• On-Demand
• Free Video

Deploying and Debugging HTTP/3 - Challenges and Solutions

Explore QUIC and HTTP/3 deployment challenges, debugging techniques, and real-world case studies. Learn about encryption, load balancing, and troubleshooting tools for modern networking protocols.

• YouTube
• 43 minutes
• On-Demand
• Free Video