mlsanityData to 14 Oct 2026

MLSYSIM: First-Principles Infrastructure Modeling for Machine Learning Systems

arXiv 2607.02558 ↗First posted 28 Jun 2026cs.DC cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

As machine learning shifts from laboratory curiosity to critical infrastructure, the systems that sustain it span an extraordinary range, from sub-milliwatt microcontrollers to multi-gigawatt datacenter fleets. Reasoning across this range is hard: empirical profiling requires the target hardware in hand, while cycle-accurate simulation costs hours per configuration, leaving no tool for rapid, full-stack architectural reasoning. We present MLSYSIM (Machine Learning Systems Infrastructure Modeling), a first-principles analytical framework that formalizes the "physics of systems" into a dimensionally-strict Python engine. MLSysim is built on a demand-supply abstraction that decouples computational demand from silicon supply and environmental context, and it enforces unit integrity at runtime so the silent conversion errors that plague ad-hoc modeling cannot occur. Every input is drawn from a typed, provenance-tracked registry, so no number enters an analysis without a documented source. On this engine we codify a taxonomy of 22 "Systems Walls" resolved by 28 composable models and solvers, enabling sub-second design-space exploration that identifies binding constraints and synthesizes ideal hardware specifications across the entire ML systems lifecycle.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 37 of its 74 references that are in the map. First posted on arXiv 28 Jun 2026.

Builds on

37 of 74 references.

References · 37

  1. 2024The Llama 3 Herd of Modelscited by 14,715
  2. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  3. 2022FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awarenesscited by 3,673
  4. 2017Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layercited by 3,509
  5. 2021Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsitycited by 3,073
  6. 2022Training Compute-Optimal Large Language Modelscited by 3,019
  7. 2016Deep Learning with Differential Privacycited by 2,728
  8. 2015Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Codingcited by 2,329
  9. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameterscited by 1,986
  10. 2022GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformerscited by 1,919
  11. 2019Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelismcited by 1,868
  12. 2022Fast Inference from Transformers via Speculative Decodingcited by 1,653
25 more
  1. 2020GShard: Scaling Giant Models with Conditional Computation and Automatic Shardingcited by 1,631
  2. 2019ZeRO: Memory Optimizations Toward Training Trillion Parameter Modelscited by 1,301
  3. 2023AWQ: Activation-aware Weight Quantization for LLM Compression and Accelerationcited by 1,298
  4. 2023SGLang: Efficient Execution of Structured Language Model Programscited by 1,141
  5. 2021A Survey of Quantization Methods for Efficient Neural Network Inferencecited by 815
  6. 2021Efficient Large-Scale Language Model Training on GPU Clusters Using Megatron-LMcited by 632
  7. 2024DistServe: Disaggregating Prefill and Decoding for Goodput-optimized Large Language Model Servingcited by 617
  8. 2021Carbon Emissions and Large Neural Network Trainingcited by 567
  9. 2023Splitwise: Efficient generative LLM inference using phase splittingcited by 558
  10. 2022Efficiently Scaling Transformer Inferencecited by 539
  11. 2024Taming Throughput-Latency Tradeoff in LLM Inference with Sarathi-Servecited by 526
  12. 2022Reducing Activation Recomputation in Large Transformer Modelscited by 302
  13. 2018Beyond Data and Model Parallelism for Deep Neural Networkscited by 179
  14. 2024LLM Inference Unveiled: Survey and Roofline Model Insightscited by 152
  15. 2024Vidur: A Large-Scale Simulation Framework For LLM Inferencecited by 117
  16. 2020Chasing Carbon: The Elusive Environmental Footprint of Computingcited by 72
  17. 2023LLMCarbon: Modeling the end-to-end Carbon Footprint of Large Language Modelscited by 72
  18. 2023ASTRA-sim2.0: Modeling Hierarchical Networks and Disaggregated Systems for Large-model Training at Scalecited by 47
  19. 2019Energy Usage Reports: Environmental awareness as part of algorithmic accountabilitycited by 44
  20. 2021tf.data: A Machine Learning Data Processing Frameworkcited by 41
  21. 2025Insights into DeepSeek-V3: Scaling Challenges and Reflections on Hardware for AI Architecturescited by 39
  22. 2020Analyzing and Mitigating Data Stalls in DNN Trainingcited by 36
  23. 2021A Runtime-Based Computational Performance Predictor for Deep Neural Network Trainingcited by 29
  24. 2025Lumos: Efficient Performance Modeling and Estimation for Large-scale LLM Trainingcited by 7
  25. 2025Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodputcited by 2

Built on by

Not cited yet.