SREcon Talks & Presentations

USENIX

Cross Continent Infrastructure Scaling at Instagram

Explore Instagram's global infrastructure scaling challenges and solutions, focusing on cross-continent deployment, latency issues, and strategies for managing stateful services.

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

USENIX

Linux Memory Management at Scale - Under the Hood

Explore Linux memory management for reliable, large-scale systems. Learn about swap, cgroups, PSI, and memory protection from a Facebook SRE, debunking myths and offering expert insights.

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

USENIX

Leading without Managing - Becoming an SRE Technical Leader

Explore technical leadership in SRE, focusing on strategic guidance without people management. Learn tactics for career growth, mentoring, and identifying impactful work opportunities.

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

USENIX

Building Blocks of Distributed Systems - Parts 1 & 2

Interactive class exploring orchestration, load balancing, databases, and storage in distributed systems. Covers key concepts and design principles for large data processing pipelines through collaborative exercises.

• YouTube
• 1 hour 23 minutes
• On-Demand
• Conference Talk

USENIX

Implementing Distributed Consensus

Explore distributed consensus through "Skinny," a lock service. Learn Paxos protocol, design a simple consensus protocol, and tackle implementation challenges from theory to coding.

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

USENIX

Tuning Java's G1 Garbage Collector for Realtime Services

Explore Java's G1 Garbage Collector tuning for realtime services, focusing on performance measurement, latency optimization, and effective settings for large heaps.

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

USENIX

A Tale of Two Postmortems - A Human Factors View

Explore the contrast between Safety I and Safety II approaches in incident reviews, using a fictional company's experience to demonstrate more effective postmortem practices for improved system reliability.

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

USENIX

How We Used Kafka to Scale Database Infrastructure

Explore LinkedIn's NoSQL database scaling journey, from MySQL replication challenges to Kafka-based solutions, offering insights on reliable data delivery and infrastructure optimization.

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

USENIX

How to Trade off Server Utilization and Tail Latency

Explore the trade-offs between server utilization and tail latency in large-scale systems. Learn queueing theory basics, analyze typical outcomes, and gain practical insights for optimizing system performance.

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

USENIX

Let's Build a Distributed File System

Explore the architecture of typical and distributed file systems, including key components and operations. Learn through a simple Python implementation of a basic distributed file system.

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

USENIX

TCP - Architecture, Enhancements, and Tuning

Comprehensive exploration of TCP architecture, enhancements, and tuning, covering core concepts, connection establishment, flow control, congestion avoidance, and performance optimization techniques for network engineers.

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

USENIX

Operating Elasticsearch with Ease at Scale

Discover strategies for operating Elasticsearch at scale, covering common issues, optimization techniques, and best practices for handling massive deployments with billions of ingests and millions of searches daily.

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

USENIX

Yes, No, Maybe? Error Handling with gRPC Examples

Explore error handling in gRPC with practical examples, focusing on client-server discrepancies and effective code snippets for managing various response scenarios.

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

USENIX

Latency SLOs Done Right

Explore three approaches to accurately calculate latency SLOs using histograms for correct quantiles. Learn to frame SLOs effectively for improved service level management and monitoring.

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

USENIX

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

Explore Twitter's Aperture algorithm for client-side load balancing, addressing scalability challenges in large service clusters and improving RPC framework efficiency.

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