HyGra: Accelerating Network-State Simulation for LLM Training in DCNs via Adaptive Packet-Flow Granularity
- 1 December 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 75 (12) , 4264-4276
Abstract
No abstract availableAll Related Versions
Funding Information
- Major Scientific Instruments and Equipment Development Project of the National Natural Science Foundation of China (62427809)
- Key Program of the Natural Science Foundation of Jiangsu Province (BK20243053)
- Natural Science Foundation for Excellent Young Scholars of Jiangsu Province (BK20250143)
- Open Project of the State Key Laboratory of Internet Architecture (HLW2025MS24)
- State Key Laboratory of Processors, Institute of Computing Technology, Chinese Academy of Sciences (CLQ202506)
- Fundamental Research (Natural Science) Project of Jiangsu Higher Education Institutions (25KJB520030)
- Nanjing Science and Technology Innovation Program for Overseas Returnees
- National Natural Science Foundation of China (62602292)
This publication has 24 references indexed in Scilit:
- Advances of Pipeline Model Parallelism for Deep Learning Training: An OverviewJournal of Computer Science and Technology, 2024
- ASHL: An Adaptive Multi-Stage Distributed Deep Learning Training Scheme for Heterogeneous EnvironmentsIEEE Transactions on Computers, 2023
- FlexBlock: A Flexible DNN Training Accelerator With Multi-Mode Block Floating Point SupportIEEE Transactions on Computers, 2023
- A Practical Introduction to the OMNeT++ Simulation Framework2019
- Parallel discrete event simulation: The making of a field2017
- Upward Max Min Fairness2012
- The gem5 simulatorACM SIGARCH Computer Architecture News, 2011
- Data center TCP (DCTCP)2010
- The ns-3 Network Simulator2010
- SimGrid: A Generic Framework for Large-Scale Distributed Experiments2008