USENIX Courses

Enabling OS Research by Inferring Interactions in the Black-Box GPU Stack

Uncover GPU stack interactions to enable OS-level resource management. Methodology creates state machines capturing meaningful transitions, allowing kernel interception and control of GPU requests.

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

Packetdrill - Scriptable Network Stack Testing, from Sockets to Packets

Explore packetdrill, a scripting tool for testing TCP/UDP/IP network stack implementations. Learn its design, usage, and impact on developing Linux TCP features and bug fixing.

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

Unioning of the Buffer Cache and Journaling Layers with Non-volatile Memory

Innovative buffer cache architecture using non-volatile memory to improve file system performance by unifying caching and journaling, resulting in significant I/O performance gains without compromising reliability.

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

ROP is Still Dangerous - Breaking Modern Defenses

Explores advanced Return Oriented Programming attacks, demonstrating vulnerabilities in modern defense mechanisms and proposing strategies to overcome security measures like kBouncer and ROPecker.

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

A Full GPU Virtualization Solution with Mediated Pass-Through

Explore GPU virtualization with gVirt, achieving 95% native performance and scalability up to 7 VMs through mediated pass-through technology for secure isolation and full feature support.

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

Bringing Gesture Recognition to All Devices

Innovative gesture recognition system for low-power devices, utilizing existing wireless signals to enable always-on functionality across various platforms, including battery-free devices.

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

The Origins of USAA's Postmortem of the Week

Discover USAA's journey in establishing an IT-wide Postmortem Review meeting, with tips for implementing your own large-scale review process. Learn from their experiences and insights.

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

How Our SREs Safeguard Nanosecond Performance - at Scale - in an Environment Built to Fail

Explore SRE principles in high-frequency trading, balancing innovation with risk management. Learn strategies for safeguarding nanosecond performance in a regulated, failure-prone environment at scale.

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

SRE "Power Words" - The Lexicon of SRE as an Industry

Explore the impact of language in Site Reliability Engineering, examining key terms and their role in shaping industry practices and communication with stakeholders.

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

When Systems Flatline - Enhancing Incident Response with Learnings from the Medical Field

Explore how medical field strategies can enhance incident response in tech systems, from algorithm-guided decisions to standardized protocols for effective management and resolution.

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

Ceci N'est Pas un CPU Load

Explore the limitations of digital troubleshooting compared to analog electronics, and discover why our mental models for digital systems are constrained.

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

Spike Detection in Alert Correlation at LinkedIn

Discover how LinkedIn uses spike detection in alert correlation to quickly identify root causes of outages in their complex microservices architecture, reducing false positives and engineer toil.

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

Risky Business - Investigating the Security Practices of Vendors on an Online Anonymous Market Using Ground-Truth Data

Investigates security practices of vendors on dark web markets, revealing surprising vulnerabilities among digital item sellers compared to drug vendors. Analyzes password strength, 2FA usage, and PGP adoption.

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

Argus - Debugging Performance Issues in Modern Desktop Applications with Annotated Causal Tracing

Innovative tool for debugging performance issues in desktop applications, using annotated causal tracing to model ambiguities and efficiently diagnose root causes of slowdowns.

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

Octo - INT8 Training with Loss-aware Compensation and Backward Quantization for Tiny On-device Learning

Explores INT8 quantization for on-device learning, introducing Loss-aware Compensation and Parameterized Range Clipping to optimize forward and backward passes while preserving training quality.

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