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

Last Week in AI
Last Week in AI
U
Unit 42
博客园 - 【当耐特】
Y
Y Combinator Blog
H
Hackread – Cybersecurity News, Data Breaches, AI and More
Microsoft Security Blog
Microsoft Security Blog
Recent Announcements
Recent Announcements
P
Proofpoint News Feed
Martin Fowler
Martin Fowler
量子位
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
有赞技术团队
有赞技术团队
aimingoo的专栏
aimingoo的专栏
博客园 - 司徒正美
美团技术团队
雷峰网
雷峰网
小众软件
小众软件
G
Google Developers Blog
GbyAI
GbyAI
Jina AI
Jina AI
爱范儿
爱范儿
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
F
Fortinet All Blogs
Vercel News
Vercel News

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
Notes on Bit-reversal Broadcast Scheduling
Marcin Kik · 2012-01-17 · via cs.DC updates on arXiv.org

This report contains revision and extension of some results about RBO [arXiv:1108.5095]. RBO is a simple and efficient broadcast scheduling of $n = 2^k$ uniform frames for battery powered radio receivers. Each frame contains a key from some arbitrary linearly ordered universe. The broadcast cycle -- a sequence of frames sorted by the keys and permuted by $k$-bit reversal -- is transmitted in a round robin fashion by the broadcaster. At arbitrary time during the transmission, the receiver may start a simple protocol that reports to him all the frames with the keys that are contained in a specified interval of the key values $[K', K"]$. RBO receives at most $2 k + 1$ other frames' keys before receiving the first key from $[K', K"]$ or noticing that there are no such keys in the broadcast cycle. As a simple corollary, $4 k + 2$ is upper bound the number of keys outside $[K', K"]$ that will ever be received. In unreliable network the expected number of efforts to receive such frames is bounded by $(8 k + 4) / p + 2 (1 - p) / p^2$, where $p$ is probability of successful reception, and the reception rate of the requested frames is $p$ -- the highest possible. The receiver's protocol state consists of the values $k$, $K'$ and $K"$, one wake-up timer and two other $k$-bit variables. Its only nontrivial computation -- the computation of the next wake-up time slot -- can be performed in $O (k)$ simple operations, such as arithmetic/bit-wise operations on $k$-bit numbers, using only constant number of $k$-bit variables.