Circonus - Design Failures Case Study
Explore Circonus platform's evolution, including architectural changes and lessons learned from failures in telemetry management and SLO implementation for system reliability.
Keep Building Fresh - Shopify's Journey to Kubernetes
Shopify's transition to Kubernetes: challenges, solutions, and benefits of moving from traditional infrastructure to cloud-native. Insights on custom tooling and automation for efficient container orchestration.
The Myth of Cloud Agnosticism
Explores the practical challenges of cloud agnosticism, discussing financial, technical, and operational complexities of multi-cloud deployments and migrations between providers.
Capacity Planning in Four Parts - Telling the Future without a Crystal Ball
Learn proactive capacity planning techniques to avoid reactive scaling, covering monitoring, future planning, enforcement, re-evaluation, and common pitfall avoidance for SREs and engineers.
Know Your Kubernetes Deploys
Explore Shopify's approach to securing Kubernetes deployments through container signatures and admission controllers, ensuring a robust software supply chain.
Against On-Call - A Polemic
Challenges traditional on-call practices in tech, analyzing their impact on workers and proposing alternatives for a future without on-call shifts. Explores industry history and calls for change.
Bamboozling Certificate Authorities with BGP
Análisis de vulnerabilidades en la infraestructura de clave pública y demostración de ataques BGP para obtener certificados falsos, con propuestas de contramedidas para mejorar la seguridad.
Foreshadow - Extracting the Keys to the Intel SGX Kingdom with Transient Out-of-Order Execution
Unveiling Foreshadow: A microarchitectural attack exploiting speculative execution to extract secrets from Intel SGX enclaves, compromising trusted execution environments and remote attestation.
TeEther - Gnawing at Ethereum to Automatically Exploit Smart Contracts
Explore automatic vulnerability detection and exploit generation for Ethereum smart contracts using teEther, a tool that analyzes binary bytecode to identify and exploit security weaknesses.
Polisis - Automated Analysis and Presentation of Privacy Policies Using Deep Learning
Automated framework for analyzing privacy policies using deep learning, enabling scalable queries and applications like privacy icon assignment and question-answering systems.
BlackIoT - IoT Botnet of High Wattage Devices Can Disrupt the Power Grid
Explores potential attacks on power grids using IoT botnets of high-wattage devices, demonstrating vulnerabilities and impacts through simulations on real-world models. Highlights interdependency between IoT and power grid security.
Who Is Answering My Queries - Understanding and Characterizing Interception of the DNS Resolution Path
Explores DNS interception by on-path devices, revealing widespread occurrence and potential security risks. Presents novel detection methods and analyzes data from global IP addresses to uncover concerning practices.
Analogy Cyber Security - From 0101 to Mixed Signals
Explores threats to sensor integrity in smart devices and their impact on system security. Discusses EMI interference, vulnerabilities in voice-controlled systems, and potential defenses for improving sensor security.
Modelling and Analysis of a Hierarchy of Distance Bounding Attacks
Presentation on extending applied pi-calculus to model distance bounding protocols, analyzing security properties, and introducing a new property for protecting uncompromised devices like contactless payment cards.
Characterizing and Understanding Phases of SRE Practices
Explores the evolution of Site Reliability Engineering practices, offering insights into various stages of implementation across key areas like incident response, postmortems, and monitoring to help teams assess and advance their SRE journey.
Citron - Distributed Range Lock Management with One-sided RDMA
Innovative distributed range lock management system using one-sided RDMA, enhancing concurrent access to shared storage with improved throughput and reduced latency compared to CPU-based approaches.
SHADE - Enable Fundamental Cacheability for Distributed Deep Learning Training
Innovative caching system for distributed deep learning training, improving performance by prioritizing important data samples and enhancing cache hit ratios in I/O-intensive workloads.
Perseus - A Fail-Slow Detection Framework for Cloud Storage Systems
Innovative framework for detecting and analyzing fail-slow failures in cloud storage systems, reducing tail latency and providing insights into root causes through large-scale dataset analysis.
HadaFS - A File System Bridging the Local and Shared Burst Buffer for Exascale
Innovative file system for exascale supercomputers, bridging local and shared burst buffers. Enhances scalability, performance, and data sharing while addressing metadata management and I/O challenges.
ROLEX - A Scalable RDMA-oriented Learned Key-Value Store for Disaggregated Memory Systems
Scalable RDMA-oriented learned key-value store for disaggregated memory systems, improving performance on dynamic workloads and addressing challenges in data storage and retrieval.
Fisc - A Large-scale Cloud-native-oriented File System
Innovative cloud-native file system addressing challenges in container-based applications. Introduces lightweight client, DPU-based offloading, storage-aware mechanism, and full path QoS for improved performance and resource efficiency.