庞玉栋博客
//

分布式计算

20 篇 · 资讯流
  2. arXiv cs.DC（分布式并行）

    DySCo: Dynamic Sharding for Collaborative Edge-Cloud LLM Inference with Depth-Synchronized Batching

    Pervasive intelligent applications are increasingly deployed on mobile and Internet of Things (IoT) edge devices. Conseq

    原文 ↗
  3. arXiv cs.DC（分布式并行）

    Persistent Memory in Multi-Agent LLM Inference: What It Costs, What It Buys, and When You Can Tell

    Decomposing long-context inference across cooperating agents bounds the active KV cache per call rather than total evide

    原文 ↗
  5. arXiv cs.DC（分布式并行）

    EdgeAgent: Orchestrating On-Device LLM inference for End-User Multi-Agent Systems on CPU-GPU Unified Memory Architectures

    Emerging multi-agent LLMs demand privacy-preserving edge deployment, yet current inference systems struggle with these c

    原文 ↗
  6. arXiv cs.DC（分布式并行）

    ServeTwin: A Benchmark-Validated Simulator for Distributed LLM Architecture Exploration

    Evaluating distributed LLM serving designs on physical clusters is costly. Yet existing simulators provide only subsets

    原文 ↗
  8. arXiv cs.DB（数据库）

    HakiCC: LLM-Driven Multi-Agent Design and Optimization of Concurrency Control Protocols

    Large language models (LLMs) have recently been applied in systems research as a tool to reduce human-intensive engineer

    原文 ↗
  9. arXiv cs.DC（分布式并行）

    Leto: Fast In-Place Recovery for LLM Training on Surviving Hardware

    Hardware-operable failures (HOFs) interrupt large language model (LLM) training but permit recovery on the same hardware

    原文 ↗
  11. arXiv cs.DC（分布式并行）

    Characterizing High Bandwidth Flash for LLM Serving

    Large language model (LLM) serving requires substantial memory to store model weights and KV caches. As models grow larg

    原文 ↗
  13. arXiv cs.DC（分布式并行）

    SPLASH: Switching Parallel Layouts of Attention with Seamless Handoff for LLM Serving

    No single way of parallelizing attention serves large language models well under all loads. Low concurrency favors tenso

    原文 ↗
  14. arXiv cs.DC（分布式并行）

    Janus: Evidence-Before-Effect Sagas and Offline-Verifiable Provenance for Agentic LLMs

    Agentic large language models (LLMs) now move money through tools, yet the record of what they did is usually a trace th

    原文 ↗
  16. arXiv cs.DC（分布式并行）

    TempoKV: Timely Staging of LLM KV Caches for Memory-Semantic Flash

    Reusable prefix key-value (KV) caches can outgrow GPU memory in large language model (LLM) serving. A memory-semantic fl

    原文 ↗