parallel-computing

Documents about parallel-computing

• Taming GPU Underutilization via Static Partitioning and Fine-grained CPU Offloading#2534arXiv CS
• NL-CPS: Reinforcement Learning-Based Kubernetes Control Plane Placement in Multi-Region Clusters#2535arXiv CS
• City-Scale Visibility Graph Analysis via GPU-Accelerated HyperBall#2536arXiv CS
• Asynchronous Quantum Distributed Computing: Causality, Snapshots, and Global Operations#2537arXiv CS
• Scheduling Coflows in Multi-Core OCS Networks with Performance Guarantee#2538arXiv CS
• Wattlytics: A Web Platform for Co-Optimizing Performance, Energy, and TCO in HPC Clusters#2539arXiv CS
• LegoDiffusion: Micro-Serving Text-to-Image Diffusion Workflows#2540arXiv CS
• LogAct: Enabling Agentic Reliability via Shared Logs#2541arXiv CS
• Administrative Decentralization in Edge-Cloud Multi-Agent for Mobile Automation#2542arXiv CS
• Reduced-Mass Orbital AI Inference via Integrated Solar, Compute, and Radiator Panels#2543arXiv CS
• Blink: CPU-Free LLM Inference by Delegating the Serving Stack to GPU and SmartNIC#2544arXiv CS
• Parallel Batch-Dynamic Maximal Independent Set#2545arXiv CS
• Measurement of Generative AI Workload Power Profiles for Whole-Facility Data Center Infrastructure Planning#2546arXiv CS
• Making Room for AI: Multi-GPU Molecular Dynamics with Deep Potentials in GROMACS#2547arXiv CS
• InfiniLoRA: Disaggregated Multi-LoRA Serving for Large Language Models#2548arXiv CS
• Autopoiesis: A Self-Evolving System Paradigm for LLM Serving Under Runtime Dynamics#2549arXiv CS
• Scheduling the Unschedulable: Taming Black-Box LLM Inference at Scale#2550arXiv CS
• NestPipe: Large-Scale Recommendation Training on 1,500+ Accelerators via Nested Pipelining#2551arXiv CS
• On the Decidability of Distributed Tasks with Output Sets under Asynchrony and Any Number of Crashes#2552arXiv CS
• Determinacy with Priorities up to Clocks#2553arXiv CS
• Exploiting Aggregate Programming in a Multi-Robot Service Prototype#2554arXiv CS
• Branching Out: Existential External Choice in Effpi#2555arXiv CS
• Beyond End-to-End: Dynamic Chain Optimization for Private LLM Adaptation on the Edge#2556arXiv CS
• Nexus: Transparent I/O Offloading for High-Density Serverless Computing#2557arXiv CS
• SwarmIO: Towards 100 Million IOPS SSD Emulation for Next-generation GPU-centric Storage Systems#2558arXiv CS
• Foundry: Template-Based CUDA Graph Context Materialization for Fast LLM Serving Cold Start#2559arXiv CS
• Sparsity-Aware Roofline Models for Sparse Matrix-Matrix Multiplication#2560arXiv CS
• DynLP: Parallel Dynamic Batch Update for Label Propagation in Semi-Supervised Learning#2561arXiv CS
• ForkKV: Scaling Multi-LoRA Agent Serving via Copy-on-Write Disaggregated KV Cache#2562arXiv CS
• Fine-Grained Power and Energy Attribution on AMD GPU/APU-Based Exascale Nodes#2563arXiv CS
• CodecSight: Leveraging Video Codec Signals for Efficient Streaming VLM Inference#2564arXiv CS
• GTaP: A GPU-Resident Fork-Join Task-Parallel Runtime with a Pragma-Based Interface#2565arXiv CS
• JZ-Tree: GPU friendly neighbour search and friends-of-friends with dual tree walks in JAX plus CUDA#2566arXiv CS
• Communication Requirements for Linearizable Registers#2567arXiv CS
• Optimizing OpenFaaS on Kubernetes: Comparative Analysis of Language Runtimes and Cluster Distributions#2568arXiv CS
• ALTO: Adaptive LoRA Tuning and Orchestration for Heterogeneous LoRA Training Workloads#2569arXiv CS
• Analyzing Persistent Alltoallv RMA Implementations for High-Performance MPI Communication#2570arXiv CS
• MegaTrain: Full Precision Training of 100B+ Parameter Large Language Models on a Single GPU#2571arXiv CS
• DeepStack: Scalable and Accurate Design Space Exploration for Distributed 3D-Stacked AI Accelerators#2572arXiv CS
• RegGuard: Legitimacy and Fairness Enforcement for Optimistic Rollups#2573arXiv CS
• The Energy Cost of Execution-Idle in GPU Clusters#2574arXiv CS
• Sampling Parallelism for Fast and Efficient Bayesian Learning#2575arXiv CS
• Communication-Efficient Collaborative LLM Inference over LEO Satellite Networks#2576arXiv CS
• Edge-Oriented Orchestration of Energy Services Using Graph-Driven Swarm Intelligence#2577arXiv CS
• Tight Bounds on Window Size and Time for Single-Agent Graph Exploration under T-Interval Connectivity#2578arXiv CS
• LP-GEMM: Integrating Layout Propagation into GEMM Operations#2579arXiv CS
• NBI-Slurm: Simplified submission of Slurm jobs with energy saving mode#2580arXiv CS
• Sustaining Exascale Performance: Lessons from HPL and HPL-MxP on Aurora#5933arXiv CS
• NOMAD: Generating Embeddings for Massive Distributed Graphs#5934arXiv CS
• MATCHA: Efficient Deployment of Deep Neural Networks on Multi-Accelerator Heterogeneous Edge SoCs#5935arXiv CS

Related topics