MLSynth: Towards Synthetic ML Traces
Open Access
- 8 September 2025
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 98-104
Abstract
No abstract availableThis publication has 16 references indexed in Scilit:
- MLTCP: A Distributed Technique to Approximate Centralized Flow Scheduling For Machine Learning2024
- Rail-only: A Low-Cost High-Performance Network for Training LLMs with Trillion Parameters2024
- RDMA over Ethernet for Distributed Training at Meta Scale2024
- Network Load Balancing with Parallel Flowlets for AI Training Clusters2024
- Chakra: Advancing Performance Benchmarking and Co-design using Standardized Execution TracesarXiv, 2023
- ASTRA-sim2.0: Modeling Hierarchical Networks and Disaggregated Systems for Large-model Training at Scale2023
- GPT-4 Technical ReportarXiv, 2023
- Efficient Large-Scale Language Model Training on GPU Clusters Using Megatron-LMarXiv, 2021
- Megatron-LM: Training Multi-Billion Parameter Language Models Using Model ParallelismarXiv, 2020
- Congestion Control for Large-Scale RDMA DeploymentsACM SIGCOMM Computer Communication Review, 2015