Imagen de Portada

DynaFlow: Transparent and Flexible Intra-Device Parallelism via Programmable Operator Scheduling

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Pan, Yi, Gu, Yile, Luo, Jinbin, Wu, Yibo, Wang, Ziren, Zhang, Hongtao, Xu, Ziyi, Lin, Shengkai, Kasikci, Baris, Wang, Stephanie
Formato:	Preprint
Publicado:	2026
Materias:	Distributed, Parallel, and Cluster Computing
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

NanoFlow: Towards Optimal Large Language Model Serving Throughput
por: Zhu, Kan, et al.
Publicado: (2024)

ConsumerBench: Benchmarking Generative AI Applications on End-User Devices
por: Gu, Yile, et al.
Publicado: (2025)

Fiddler: CPU-GPU Orchestration for Fast Inference of Mixture-of-Experts Models
por: Kamahori, Keisuke, et al.
Publicado: (2024)

Argos: Agentic Time-Series Anomaly Detection with Autonomous Rule Generation via Large Language Models
por: Gu, Yile, et al.
Publicado: (2025)

Unleashing Scalable Context Parallelism for Foundation Models Pre-Training via FCP
por: Zhao, Yilong, et al.
Publicado: (2026)

CFP: Efficient Optimization of Intra-Operator Parallelism Plans for Large Model Training
por: Hu, Weifang, et al.
Publicado: (2025)

A Flexible Programmable Pipeline Parallelism Framework for Efficient DNN Training
por: Jiang, Lijuan, et al.
Publicado: (2025)

VibeServe: Can AI Agents Build Bespoke LLM Serving Systems?
por: Kamahori, Keisuke, et al.
Publicado: (2026)

Accelerating Heterogeneous Tensor Parallelism via Flexible Workload Control
por: Wang, Zhigang, et al.
Publicado: (2024)

TeleRAG: Efficient Retrieval-Augmented Generation Inference with Lookahead Retrieval
por: Lin, Chien-Yu, et al.
Publicado: (2025)

DynaTrain: Fast Online Parallelism Switching for Elastic LLM Training
por: Wang, Yuanqing, et al.
Publicado: (2026)

Fake Runs, Real Fixes -- Analyzing xPU Performance Through Simulation
por: Zarkadas, Ioannis, et al.
Publicado: (2025)

Magneton: Optimizing Energy Efficiency of ML Systems via Differential Energy Debugging
por: Pan, Yi, et al.
Publicado: (2025)

JanusPipe: Efficient Pipeline Parallel Training for Machine Learning Interatomic Potentials
por: Wang, Hongyu, et al.
Publicado: (2026)

PolyServe: Efficient Multi-SLO Serving at Scale
por: Zhu, Kan, et al.
Publicado: (2025)

Efficient Task Graph Scheduling for Parallel QR Factorization in SLSQP
por: Chatterjee, Soumyajit, et al.
Publicado: (2025)

A Simulated Annealing Approach to Identical Parallel Machine Scheduling
por: Li, Jiaxing, et al.
Publicado: (2024)

Hyperion: Hierarchical Scheduling for Parallel LLM Acceleration in Multi-tier Networks
por: Ma, Mulei, et al.
Publicado: (2025)

Concurrent Scheduling of High-Level Parallel Programs on Multi-GPU Systems
por: Knorr, Fabian, et al.
Publicado: (2025)

DreamDDP: Accelerating Data Parallel Distributed LLM Training with Layer-wise Scheduled Partial Synchronization
por: Tang, Zhenheng, et al.
Publicado: (2025)

Prioritizing Modalities: Flexible Importance Scheduling in Federated Multimodal Learning
por: Bian, Jieming, et al.
Publicado: (2024)

Optimal Parallel Scheduling under Concave Speedup Functions
por: Li, Chengzhang, et al.
Publicado: (2025)

Aryl: An Elastic Cluster Scheduler for Deep Learning
por: Li, Jiamin, et al.
Publicado: (2022)

Efficient MoE Inference with Fine-Grained Scheduling of Disaggregated Expert Parallelism
por: Pan, Xinglin, et al.
Publicado: (2025)

SLICE: SLO-Driven Scheduling for LLM Inference on Edge Computing Devices
por: Chow, Will
Publicado: (2025)

FlexPie: Accelerate Distributed Inference on Edge Devices with Flexible Combinatorial Optimization[Technical Report]
por: Zhang, Runhua, et al.
Publicado: (2025)

Bandwidth-Aware and Cost-Efficient Pipeline Parallel Scheduling in Geo-Distributed LLM Training
por: Zhang, Han, et al.
Publicado: (2026)

DeFT: Mitigating Data Dependencies for Flexible Communication Scheduling in Distributed Training
por: Meng, Lin, et al.
Publicado: (2025)

A Taxonomy of Schedulers -- Operating Systems, Clusters and Big Data Frameworks
por: Sliwko, Leszek
Publicado: (2025)

Communication-Efficient Device Scheduling for Federated Learning Using Lyapunov Optimization
por: Perazzone, Jake B., et al.
Publicado: (2025)

DynaServe: Unified and Elastic Execution for Dynamic Disaggregated LLM Serving
por: Ruan, Chaoyi, et al.
Publicado: (2025)

VoxServe: Streaming-Centric Serving System for Speech Language Models
por: Kamahori, Keisuke, et al.
Publicado: (2026)

FlashInfer: Efficient and Customizable Attention Engine for LLM Inference Serving
por: Ye, Zihao, et al.
Publicado: (2025)

RingAda: Pipelining Large Model Fine-Tuning on Edge Devices with Scheduled Layer Unfreezing
por: Li, Liang, et al.
Publicado: (2025)

GPUOS: A GPU Operating System Primitive for Transparent Operation Fusion
por: Yang, Yiwei, et al.
Publicado: (2026)

Efficient Data Labeling and Optimal Device Scheduling in HWNs Using Clustered Federated Semi-Supervised Learning
por: Hamood, Moqbel, et al.
Publicado: (2024)

SP-Chain: Boosting Intra-Shard and Cross-Shard Security and Performance in Blockchain Sharding
por: Li, Mingzhe, et al.
Publicado: (2024)

FlexSP: Accelerating Large Language Model Training via Flexible Sequence Parallelism
por: Wang, Yujie, et al.
Publicado: (2024)

Scheduling Coflows in Multi-Core OCS Networks with Performance Guarantee
por: Wang, Xin, et al.
Publicado: (2026)

Taming Request Imbalance: SLO-Aware Scheduling for Disaggregated LLM Inference
por: Wang, Qipeng
Publicado: (2026)