USENIX Courses

SparTA - Deep-Learning Model Sparsity via Tensor-with-Sparsity-Attribute

Explores deep learning model sparsity via Tensor-with-Sparsity-Attribute, enabling efficient sparse models across various patterns and optimization techniques, delivering significant speedups compared to state-of-the-art solutions.

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

Carbink - Fault-Tolerant Far Memory

Fault-tolerant far memory system using erasure-coding and remote memory compaction for efficient datacenter applications, offering improved latency and performance over existing solutions.

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

CAP-VMs - Capability-Based Isolation and Sharing in the Cloud

Explore capability-based isolation and efficient data sharing in cloud environments using cVMs, a novel VM-like abstraction leveraging memory capabilities for enhanced security and performance.

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

Orca - A Distributed Serving System for Transformer-Based Generative Models

Explore Orca, a distributed serving system for transformer-based generative models, featuring iteration-level scheduling and selective batching to enhance performance and efficiency in large-scale language model inference.

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

MemLiner - Lining up Tracing and Application for a Far-Memory-Friendly Runtime

MemLiner: Runtime technique improving far-memory systems by aligning application and garbage collection memory accesses, reducing local-memory working set and enhancing remote-memory prefetching for better performance.

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

Owl - Scale and Flexibility in Distribution of Hot Content

Explore Owl, Meta's high-scale content distribution system combining centralized control with peer-to-peer data transfer, offering improved speed and flexibility over traditional methods.

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

Unity - Accelerating DNN Training Through Joint Optimization of Algebraic Transformations and Parallelization

Explore Unity, a system optimizing distributed DNN training through joint algebraic transformations and parallelization. Learn about its unified parallel computation graph and hierarchical search algorithm for improved performance.

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

Metastable Failures in the Wild

Explores metastable failures in distributed systems, analyzing real-world incidents to categorize triggers and amplification mechanisms, aiding understanding and potential solutions for this prevalent issue.

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

KSplit - Automating Device Driver Isolation

Automating device driver isolation to protect kernels from vulnerabilities, using CPU extensions for low-overhead implementation and advanced analyses to identify shared state and synchronization requirements.

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

ROLLER - Fast and Efficient Tensor Compilation for Deep Learning

Introduces ROLLER, a fast tensor compilation approach for deep learning using rTile abstraction and recursive construction, generating efficient kernels in seconds for various accelerators.

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

Verifying Hardware Security Modules with Information-Preserving Refinement

Explore Knox, a framework for building verified hardware security modules (HSMs) that eliminates bugs and side channels through formal verification and information-preserving refinement.

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

Zero-Change Object Transmission for Distributed Big Data Analytics

Explore zero-change object transmission for efficient data exchange in distributed big data analytics, improving performance without code modifications.

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

KRCORE - A Microsecond-scale RDMA Control Plane for Elastic Computing

Microsecond-scale RDMA control plane for elastic computing, enabling rapid connection establishment and efficient data transfer in distributed systems and serverless environments.

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

ECOS - Practical Mobile Application Offloading for Enterprises

Explore ECOS, an enterprise-centric offloading system for mobile applications, addressing privacy concerns and resource management to improve handheld device performance and energy efficiency.

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

Runtime Programmable Switches

Runtime programmable switches enabling partial reconfiguration of network data planes without service disruption. FlexCore design allows flexible updates with consistency guarantees, demonstrating feasibility and benefits for future networks.

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