USENIX Courses

Automatically Detect the Top Performance & Scalability Issues in Distributed Architectures

Explore common performance patterns in distributed systems and learn automated detection techniques using CNCF's Keptn project for CI/CD pipelines, focusing on SLI/SLO-based quality gates.

• YouTube
• 37 minutes
• Self-Paced
• Conference Talk

Implementing Distributed Consensus

Explore distributed consensus through "Skinny," a lock service. Learn Paxos protocol, design a simple consensus protocol, and tackle implementation challenges in distributed systems.

• YouTube
• 40 minutes
• Self-Paced
• Conference Talk

Challenges of Starting an SRE Team from Scratch in an Enterprise

Insights on building an SRE team in a large enterprise, addressing challenges, establishing standards, and focusing on business priorities like security and cloud cost control.

• YouTube
• 33 minutes
• Self-Paced
• Conference Talk

When - Bin - Sh Attacks - Revisiting Automate All the Things

Explores challenges and impacts of pervasive automation in SRE, discussing human factors, potential harm, and strategies for effective implementation in complex socio-technical systems.

• YouTube
• 41 minutes
• Self-Paced
• Conference Talk

The Smallest Possible SRE Team

Strategies for maximizing SRE impact with limited resources. Learn to leverage data, tools, and communication for effective small-scale SRE implementation and organizational influence.

• YouTube
• 38 minutes
• Self-Paced
• Conference Talk

Confessions of a Systems Engineer - Learning from My 20+ Years of Failure

Veteran systems engineer shares invaluable lessons from two decades of failures, offering insights on designing, running, and troubleshooting large-scale online services to help others avoid costly mistakes.

• YouTube
• 39 minutes
• Self-Paced
• Conference Talk

Achieving Mutual TLS - Secure Pod-to-Pod Communication Without the Hassle

Explore secure pod-to-pod communication using Mutual TLS, focusing on certificate management, implementation in Golang and Java applications, and integration with Kubernetes for streamlined security.

• YouTube
• 32 minutes
• Self-Paced
• Conference Talk

Theseus - An Experiment in Operating System Structure and State Management

Redesigning OS modularity using Rust, Theseus reduces inter-component state dependencies, enabling live evolution and fault recovery for core components beyond existing systems.

• YouTube
• 20 minutes
• Self-Paced
• Conference Talk

RedLeaf - Isolation and Communication in a Safe Operating System

Explore RedLeaf, a Rust-based OS leveraging language safety for fine-grained isolation. Learn about innovative approaches to fault isolation, driver recovery, and performance in operating system design.

• YouTube
• 22 minutes
• Self-Paced
• Conference Talk

Microsecond Consensus for Microsecond Applications

Innovative consensus protocol leveraging RDMA for microsecond-scale replication and rapid fail-over in latency-sensitive applications, significantly outperforming existing systems.

• YouTube
• 20 minutes
• Self-Paced
• Conference Talk

KungFu - Making Training in Distributed Machine Learning Adaptive

KungFu: adaptive distributed machine learning library for TensorFlow. Enables real-time parameter tuning using high-level policies, efficient monitoring, and asynchronous communication for improved training performance.

• YouTube
• 20 minutes
• Self-Paced
• Conference Talk

Caladan - Mitigating Interference at Microsecond Timescales

Explore a new CPU scheduler that achieves better quality of service through fast core allocation, outperforming resource partitioning systems in managing microsecond-scale interference between tasks.

• YouTube
• 20 minutes
• Self-Paced
• Conference Talk

PANIC - A High-Performance Programmable NIC for Multi-tenant Networks

Explores Frenzy, a novel programmable NIC design with high-performance switching interconnect and hybrid push/pull packet scheduler, addressing limitations in multi-tenant networks and diverse offloads.

• YouTube
• 20 minutes
• Self-Paced
• Conference Talk

AIFM - High-Performance, Application-Integrated Far Memory

Explore AIFM, a high-performance system enabling efficient use of remote memory in datacenters. Learn about its API, runtime, and benefits for applications facing memory constraints.

• YouTube
• 20 minutes
• Self-Paced
• Conference Talk

Ansor - Generating High-Performance Tensor Programs for Deep Learning

Ansor: A framework for generating high-performance tensor programs in deep learning, improving execution speed across various hardware platforms through innovative search strategies and optimization techniques.

• YouTube
• 20 minutes
• Self-Paced
• Conference Talk