惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

D
DataBreaches.Net
Y
Y Combinator Blog
I
InfoQ
Recent Announcements
Recent Announcements
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
博客园 - Franky
IT之家
IT之家
H
Help Net Security
月光博客
月光博客
S
SegmentFault 最新的问题
B
Blog
aimingoo的专栏
aimingoo的专栏
GbyAI
GbyAI
P
Proofpoint News Feed
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
G
Google Developers Blog
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
U
Unit 42
Vercel News
Vercel News
博客园 - 叶小钗
Microsoft Security Blog
Microsoft Security Blog
腾讯CDC
Jina AI
Jina AI
T
The Blog of Author Tim Ferriss

cs.DC updates on arXiv.org

DUAL-BLADE: Dual-Path NVMe-Direct KV-Cache Offloading for Edge LLM Inference Progressive Semantic Communication for Efficient Edge-Cloud Vision-Language Models Efficient, VRAM-Constrained xLM Inference on Clients Folding Tensor and Sequence Parallelism for Memory-Efficient Transformer Training & Inference DORA: A Scalable Asynchronous Reinforcement Learning System for Language Model Training AMMA: A Multi-Chiplet Memory-Centric Architecture for Low-Latency 1M Context Attention Serving RaMP: Runtime-Aware Megakernel Polymorphism for Mixture-of-Experts Spark Policy Toolkit: Semantic Contracts and Scalable Execution for Policy Learning in Spark Internet of Everything in the 6G Era: Paradigms, Enablers, Potentials and Future Directions PolyKV: A Shared Asymmetrically-Compressed KV Cache Pool for Multi-Agent LLM Inference A Survey on Split Learning for LLM Fine-Tuning: Models, Systems, and Privacy Optimizations ITAS: A Multi-Agent Architecture for LLM-Based Intelligent Tutoring Latency and Cost of Multi-Agent Intelligent Tutoring at Scale TACO: Efficient Communication Compression of Intermediate Tensors for Scalable Tensor-Parallel LLM Training FreeScale: Distributed Training for Sequence Recommendation Models with Minimal Scaling Cost CommFuse: Hiding Tail Latency via Communication Decomposition and Fusion for Distributed LLM Training A Taxonomy and Resolution Strategy for Client-Level Disagreements in Federated Learning Usable Agent Discovery for Decentralized AI Systems Cloud to Edge: Benchmarking LLM Inference On Hardware-Accelerated Single-Board Computers Data-Free Contribution Estimation in Federated Learning using Gradient von Neumann Entropy Shard the Gradient, Scale the Model: Serverless Federated Aggregation via Gradient Partitioning Promoting Simple Agents: Ensemble Methods for Event-Log Prediction GraphLeap: Decoupling Graph Construction and Convolution for Vision GNN Acceleration on FPGA AGNT2: Autonomous Agent Economies on Interaction-Optimized Layer 2 Infrastructure FedSIR: Spectral Client Identification and Relabeling for Federated Learning with Noisy Labels Stream-CQSA: Avoiding Out-of-Memory in Attention Computation via Flexible Workload Scheduling A Delta-Aware Orchestration Framework for Scalable Multi-Agent Edge Computing Federated Learning over Blockchain-Enabled Cloud Infrastructure Optimal Routing for Federated Learning over Dynamic Satellite Networks: Tractable or Not? Sherpa.ai Privacy-Preserving Multi-Party Entity Alignment without Intersection Disclosure for Noisy Identifiers
Exploiting Parallelism in Optical Network Systems: A Case...
Anna Engelmann, Wolfgang Bziuk, Admela Jukan, Muriel Medard · 2017-07-10 · via cs.DC updates on arXiv.org

As parallelism becomes critically important in the semiconductor technology, high-performance computing, and cloud applications, parallel network systems will increasingly follow suit. Today, parallelism is an essential architectural feature of 40/100/400 Gigabit Ethernet standards, whereby high speed Ethernet systems are equipped with multiple parallel network interfaces. This creates new network topology abstractions and new technology requirements: instead of a single high capacity network link, multiple Ethernet end-points and interfaces need to be considered together with multiple links in form of discrete parallel paths. This new paradigm is enabling implementations of various new features to improve overall system performance. In this paper, we analyze the performance of parallel network systems with network coding. In particular, by using random LNC (RLNC), - a code without the need for decoding, we can make use of the fact that we have codes that are both distributed (removing the need for coordination or optimization of resources) and composable (without the need to exchange code information), leading to a fully stateless operation. We propose a novel theoretical modeling framework, including derivation of the upper and lower bounds as well as an expected value of the differential delay of parallel paths, and the resulting queue size at the receiver. The results show a great promise of network system parallelism in combination with RLNC: with a proper set of design parameters, the differential delay and the buffer size at the Ethernet receiver can be reduced significantly, while the cross-layer design and routing can be greatly simplified.