USENIX Courses

Case Study: A Thundering Herd in the Wild

Dive into a real-world investigation of an unexpected "thundering herd" variant, following the troubleshooting process from sampling to kernel mode analysis that revealed synchronized thread wakeups caused by absolute timer system calls.

• YouTube
• 35 minutes
• Self-Paced
• Free Video

Optimizing Machine Learning Training Infrastructure: A Governance Approach

Discover how Meta transformed ML training infrastructure management through governance strategies that drive ROI, maximize resources, and accelerate time-to-market, with real-world examples and lessons learned.

• YouTube
• 32 minutes
• Self-Paced
• Free Video

Distributed Tracing in Action: Our Journey with OpenTelemetry

Explore a journey with Distributed Tracing using OpenTelemetry and Istio in microservices. Learn about context propagation challenges, automatic instrumentation, and how tracing enhances application performance insights.

• YouTube
• 32 minutes
• Self-Paced
• Free Video

"On-Call Is Ruining My Life" and Other Tales about Holding the Pager as an SRE

Explore the human and technical aspects of SRE on-call duty, from survey findings to alternative models, with insights on improving the experience and questioning current practices.

• YouTube
• 35 minutes
• Self-Paced
• Free Video

Chaos Experiments - Datacenter Stress Testing

Explore how a financial services provider developed an automated chaos engineering program for entire datacenter testing, overcoming challenges of fear and uncertainty while building robust infrastructure.

• YouTube
• 46 minutes
• Self-Paced
• Free Video

Inclusive SRE: Best Practices for Working with a Visually Impaired Incident Analyst or Responder

Discover best practices for collaborating with visually impaired incident analysts, including screen reader demos and strategies to overcome dashboard visibility challenges while leveraging diverse perspectives.

• YouTube
• 32 minutes
• Self-Paced
• Free Video

The Search for Speed - Optimizing OpenSearch Performance

Discover how to tackle OpenSearch performance issues using scientific methods, monitoring, modeling, and experimentation to reduce latency and costs while creating a framework for continuous improvement.

• YouTube
• 29 minutes
• Self-Paced
• Free Video

Handling the Largest Domains Migration, Ever!

Discover the challenges and strategies behind Squarespace's unprecedented migration of over 10 million domains from Google Domains, exploring the technical complexities of this massive infrastructure project.

• YouTube
• 37 minutes
• Self-Paced
• Free Video

Beyond Sequential: A Recipe for Async Pipeline Observability and Alerting

Discover practical strategies for implementing Service Level Objectives (SLOs) in asynchronous microservices, from identifying key metrics to designing dashboards and alerts that effectively monitor customer experience.

• YouTube
• 36 minutes
• Self-Paced
• Free Video

Maturing Your Data Architecture in a Week: How Bluesky Survived

Discover how Bluesky's small engineering team rapidly evolved their data architecture to handle explosive growth of 1M+ users per day and 1,600+ events/second during an 11-day crisis period.

• YouTube
• 39 minutes
• Self-Paced
• Free Video

Running DRP Tabletop Exercises - Disaster Recovery Plan Testing

Discover how to develop, test, and maintain effective Disaster Recovery Plans through collaborative tabletop exercises. Learn best practices for designing reliable services and ensuring your organization is prepared for recovery after disasters.

• YouTube
• 38 minutes
• Self-Paced
• Free Video

Cattle vs. Pets - A Cost-Effective Elasticsearch Architecture to Scale-Out Beyond Petabytes

Discover a scalable, cost-effective multi-cluster Elasticsearch architecture that handles petabytes of data, featuring intelligent query routing, real-time auditing, and automated rate-limiting to reduce costs while maintaining performance.

• YouTube
• 43 minutes
• Self-Paced
• Free Video

Is the S in SRE for "Security"?

Explore the significant overlap between Cybersecurity and Site Reliability Engineering (SRE), learning how these disciplines can collaborate to improve organizational performance through shared capabilities and mutual goals.

• YouTube
• 32 minutes
• Self-Paced
• Free Video

An SRE Approach to Monitoring ML in Production

Explore the challenges of monitoring ML systems in production from an SRE perspective, learning how to bridge domain expertise gaps and implement effective observability practices for reliable ML operations at scale.

• YouTube
• 39 minutes
• Self-Paced
• Free Video

Techniques Netflix Uses to Weather Significant Demand Shifts

Discover how Netflix's streaming control plane architecture handles massive demand shifts through load balancing, failover strategies, intelligent scaling, and resilience techniques for maintaining service quality across global systems.

• YouTube
• 44 minutes
• Self-Paced
• Free Video