SREcon Talks & Presentations

USENIX

Emerging from Burnout

Exploring burnout in SRE: causes, impacts, and strategies for recovery. Learn to apply system reliability principles to personal resilience and mental health management.

• YouTube
• 41 minutes
• On-Demand
• Conference Talk

USENIX

I'm Putting Sloths on the Map

Explore Sloth, a Go tool for simulating network failures in cloud infrastructure. Learn how it helps discover and fix issues in monitoring, graceful degradation, and usability through controlled network disruptions.

• YouTube
• 26 minutes
• On-Demand
• Conference Talk

USENIX

A Practical Guide to Monitoring and Alerting with Time Series at Scale

Explore time series-based alerting methods and practical examples in Prometheus to reduce alert spam and improve production hygiene for more efficient monitoring at scale.

• YouTube
• 41 minutes
• On-Demand
• Conference Talk

USENIX

Reliability When Everything Is a Platform - Why You Need to SRE Your Customers

Explore the concept of SREs taking joint operational responsibility for customer-built systems on platforms, addressing challenges and discussing future implications for SRE as a discipline.

• YouTube
• 49 minutes
• On-Demand
• Conference Talk

USENIX

Aperture - A Non-Cooperative, Client-Side Load Balancing Algorithm

Explore Twitter's Aperture algorithm for non-cooperative, client-side load balancing in large-scale RPC frameworks, addressing scalability challenges in service clusters with thousands of instances.

• YouTube
• 26 minutes
• On-Demand
• Conference Talk

USENIX

The Map Is Not the Territory - How SLOs Lead Us Astray, and What We Can Do about It

Explore SLO limitations in complex services, understand failure modes, and learn best practices for robust SLO construction to improve reliability and decision-making in modern systems.

• YouTube
• 45 minutes
• On-Demand
• Conference Talk

USENIX

Pushing through Friction

Strategies for engineering leaders to identify and overcome organizational friction, enhancing team efficiency and driving positive change in complex technical environments.

• YouTube
• 35 minutes
• On-Demand
• Conference Talk

USENIX

SRE in the Third Age

Explore the evolution of Site Reliability Engineering (SRE) from its Google origins to widespread adoption, and ponder its future as the field matures and faces new challenges in the tech industry.

• YouTube
• 25 minutes
• On-Demand
• Conference Talk

USENIX

Zero-Downtime Rebalancing and Data Migration of a Mature Multi-Shard Platform

Explore zero-downtime techniques for rebalancing and migrating data in multi-shard platforms, focusing on MySQL binlog-based tools for load distribution and tenant isolation.

• YouTube
• 39 minutes
• On-Demand
• Conference Talk

USENIX

The SRE I Aspire to Be

Explore the importance of high-quality reliability measurements in SRE, learn tips for effective implementation, and understand why it's crucial for engineering work in SRE organizations.

• YouTube
• 43 minutes
• On-Demand
• Conference Talk

USENIX

Building a Scalable Monitoring System

Learn how to overhaul and centralize a complex monitoring system, creating a user-friendly solution that encourages developer engagement and continuous improvement.

• YouTube
• 27 minutes
• On-Demand
• Conference Talk

USENIX

Latency SLOs Done Right

Explore effective methods for implementing Latency Service Level Objectives (SLOs), addressing common pitfalls in latency measurement and aggregation for improved service quality tracking.

• YouTube
• 27 minutes
• On-Demand
• Conference Talk

USENIX

A Dashboard Is Worth a Thousand Words - Better Monitoring for Better Ops

Explore how CERN's new monitoring system improved service operations and fostered SRE practices, discussing design decisions, operational challenges, and the benefits of implementing SLIs/SLOs.

• YouTube
• 52 minutes
• On-Demand
• Conference Talk

USENIX

HBase Internals and Operations

Explore HBase architecture, components, and key metrics for efficient high-availability operations in this comprehensive overview of the Hadoop-based key-value datastore.

• YouTube
• 33 minutes
• On-Demand
• Conference Talk

USENIX

Software Networking and Interfaces on Linux

Comprehensive overview of Linux network interfaces, from traditional to modern cloud-native types. Explores VMs, containers, and service meshes, aiding in debugging cloud platforms.

• YouTube
• 57 minutes
• On-Demand
• Conference Talk