BGL - GPU-Efficient GNN Training by Optimizing Graph Data I/O and Preprocessing
Optimizing GPU-based GNN training for large graphs through improved data preprocessing, caching, and partitioning techniques, enhancing efficiency and performance in graph-based machine learning tasks.
Synthesizing Runtime Programmable Switch Updates
Explores FlexPlan, a tool for safe runtime updates of programmable switches. Discusses techniques for generating step-by-step plans that maintain correctness and memory constraints during live reconfigurations.
Enabling High-Quality Real-Time Communications with Adaptive Frame-Rate
Explore adaptive frame rate control for ultra-low latency in high-quality real-time communications, addressing decoder bottlenecks and network fluctuations to enhance user experience in cloud gaming.
SLNet - A Spectrogram Learning Neural Network for Deep Wireless Sensing
Innovative deep learning approach for wireless sensing, combining spectrogram generation and learning to enhance accuracy in various applications like gesture recognition and fall detection.
DOTE - Rethinking Predictive WAN Traffic Engineering
Innovative approach to WAN traffic engineering using historical data and deep learning, outperforming existing methods in quality and runtime while nearly matching an omniscient oracle.
OpenLoRa - Validating LoRa Implementations through an Extensible and Open-sourced Framework
Explore OpenLoRa, an open-source platform for evaluating LoRa packet collision resolution techniques. Learn about implementation challenges and performance benchmarks in low SNR environments.
OneWAN is Better Than Two - Unifying a Split WAN Architecture
Unifying split WAN architecture for cloud providers, addressing operational complexity and cost inefficiency. Introduces ONEWAN, a software-defined solution for Internet and inter-datacenter traffic, featuring innovative routing and traffic engineering t…
Sketchovsky - Enabling Ensembles of Sketches on Programmable Switches
Optimizing network measurement tasks on programmable switches using Sketchovsky, a framework for efficient cross-sketch optimization and composition to reduce hardware resource usage.
SelfTune - Tuning Cluster Managers
Automated framework for optimizing cluster manager parameters in large-scale cloud environments, improving throughput and performance through reinforcement learning algorithms.
SRNIC - A Scalable Architecture for RDMA NICs
Scalable RDMA NIC architecture improving connection and network scalability while maintaining high performance and low CPU overhead, outperforming commercial RNICs in large-scale data centers.
Bolt - Sub-RTT Congestion Control for Ultra-Low Latency
Innovative congestion control system for ultra-fast data center networks, reducing latency and improving flow completion time while maintaining high utilization at speeds up to 400Gbps.
Rearchitecting the TCP Stack for I-O-Offloaded Content Delivery
Explores IO-TCP, a split TCP stack design offloading disk I/O and packet transfer to SmartNIC, enhancing online content delivery performance by reducing CPU burden and improving scalability.
RingLeader - Efficiently Offloading Intra-Server Orchestration to NICs
Explores a novel approach to datacenter server orchestration by offloading tasks to NICs, improving scalability, efficiency, latency, and throughput compared to software-only solutions.
Dashlet - Taming Swipe Uncertainty for Robust Short Video Streaming
Innovative system for improving short video streaming quality by predicting user swipe patterns and pre-buffering content, outperforming TikTok while reducing wasted data downloads.
RECL - Responsive Resource-Efficient Continuous Learning for Video Analytics
Innovative video analytics framework integrating model reuse and online retraining for efficient, adaptive DNN models. Optimizes GPU allocation and leverages shared model zoo for improved performance across diverse vision tasks.
Ichthyology - Phishing as a Science
Explore real-world phishing attacks, psychological tactics, and advanced prevention techniques. Learn evidence-based strategies to protect against credential theft in this insightful cybersecurity presentation.
Cloak & Dagger - From Two Permissions to Complete Control of the UI Feedback Loop
Unveiling Android vulnerabilities: combining permissions for stealthy UI control attacks, exploring security mechanisms, and discussing potential countermeasures and Android team responses.
Breaking the x86 Instruction Set
Uncover hidden x86 instructions and hardware bugs through innovative processor fuzzing techniques and page fault analysis, revealing chipset secrets.
Revoke-Obfuscation - PowerShell Obfuscation Detection and Evasion Using Science
Explore PowerShell obfuscation detection techniques using scientific methods, including character frequency analysis, cosine similarity, and machine learning algorithms for improved cybersecurity.
How the Feds Caught Russian Mega-Carder Roman Seleznev
Uncover the intricate investigation that led to the capture of Russian hacker Roman Seleznev, responsible for massive credit card fraud and point-of-sale hacks, through electronic evidence and international cooperation.
Making Diversity a Priority in Security
Experts discuss strategies and initiatives to increase diversity in cybersecurity, addressing recruitment, retention, mentorship, and overcoming biases to create a more inclusive industry.