Marius: Machine Learning over Billion-Edge Graphs - 10x Faster and 5x Cheaper
Discover how Marius optimizes machine learning on billion-edge graphs, achieving 10x faster and 5x cheaper training through innovative data flow architecture and resource utilization.
CXL and the Art of Hierarchical Memories: Their Management and Use
Explore hierarchical memory systems with CXL technology, from fundamental concepts to advanced optimization techniques for enhanced system performance and efficiency.
Optimizing Complex Hierarchical Memory Systems Through Simulations
Discover how modular simulation techniques can optimize complex memory systems, evaluate configurations, and achieve significant performance gains in a fraction of traditional testing time.
Mill vs. Spectre - Performance and Security
Explore CPU architecture vulnerabilities, focusing on Meltdown and Spectre attacks. Learn about the Mill architecture's unique approach to performance and security without hardware speculation.
What Lies Beneath
Explore the journey of Java code execution, from source to microcode, uncovering optimizations and hardware interactions for a deeper understanding of program behavior.
Not Your Father's Complexity - Performance in a New Machine Age
Explore memory layout's impact on Java program performance, challenging traditional complexity analysis and showcasing alternative collection implementations for optimized code execution.
Effectively Scheduling Computational Graphs of Deep Neural Networks toward Their Domain-Specific Accelerators
Innovative approach for scheduling DNN computational graphs on domain-specific accelerators, optimizing resource utilization and enhancing performance through synergy between network and hardware architectures.
Heterogeneous Memory Management
Explore heterogeneous computing challenges, focusing on unified address spaces, memory management, and data placement across CPUs and parallel devices like GPUs.
Unified Memory Framework: A Unified API for Diverse Memory Technologies
Explore Intel's Unified Memory Framework (UMF), an open-source solution for managing diverse memory technologies across CPU and GPU domains through consistent middleware mechanisms.
Memory Fabric Technologies for AI Infrastructure Growth
Explore the evolving memory fabric technologies crucial for AI infrastructure growth, including key innovations and future predictions for addressing expanding AI model requirements.
HMSDK: 2-Tier Memory System Optimization Solution for DRAM and CXL Memory
Discover HMSDK의 2-Tier 메모리 시스템 최적화 솔루션과 CXL 메모리 활용을 위한 API 및 라이브러리 구현 방법을 통해 데이터센터 성능 향상 방안을 탐구합니다.
UniMem - Redesigning Disaggregated Memory within a Unified Local-Remote Memory Hierarchy
Explores UniMem, a novel disaggregated memory system that optimizes remote memory access, local cache utilization, and page migration for improved performance in large-scale computing environments.
FlexMem - Adaptive Page Profiling and Migration for Tiered Memory
Adaptive page profiling and migration system for tiered memory, improving performance through dynamic decision-making and combining profiling methods for enhanced memory management.
What Lies Beneath
Explore the journey of Java code from source to hardware execution, uncovering optimization techniques and performance insights along the way.
Effective Data Tiering and Memory Allocation Techniques in Heterogeneous Memory System
Discover advanced techniques for optimizing data tiering and memory allocation in heterogeneous systems, focusing on efficient resource management and performance enhancement.
The Path to Autonomous Storage Systems - Challenges and Solutions
Discover breakthrough research in autonomous storage systems, exploring dynamic workload adaptation, efficiency optimization, and performance management for next-generation storage hierarchies.
SSD with CXL Interfaces - Exploring Brilliance and Innovation in Storage Technology
Explore the potential of CXL-based NAND solutions, examining practical applications, technical hurdles, and emerging opportunities in memory expansion and data-centric computing technologies.
Data Placement at Scale - Landscape, Trade-Offs, and Direction
Explore data placement strategies for large-scale systems, analyzing trade-offs and future trends in optimizing storage and retrieval efficiency.
Computer Architecture 101 for Scientists and Its Impact on Cluster Jobs
Explore core concepts of parallel computer architecture and their impact on cluster job performance. Gain insights to optimize resource allocation and improve efficiency in supercomputing tasks.
C++ Performance Optimization: Algorithmic Complexity, Data Locality, Parallelism, and Compiler Optimizations
Explore C++ performance optimization through algorithmic complexity, data locality, parallelism, and compiler optimizations. Gain insights into efficient coding practices and trade-offs for improved runtime efficiency.
Unlocking Modern CPU Power - Next-Gen C++ Optimization Techniques
Explore cutting-edge C++ optimization techniques for modern CPUs, covering architecture changes, benchmarking challenges, and practical strategies to enhance code performance on latest processors.