paper-with-me

Papers

Leveraging Neural Graph Compilers in Machine Learning Research for Edge-Cloud Systems

2025-04-28 · Alireza Furutanpey, Carmen Walser, Philipp Raith, Pantelis A. Frangoudis, Schahram Dustdar

This work presents a comprehensive evaluation of neural network graph compilers across heterogeneous hardware platforms, addressing the critical gap between theoretical optimization techniques and practical deployment scenarios. We demonstrate how vendor-specific optimizations can invalidate relative performance comparisons between architectural archetypes, with performance advantages sometimes completely reversing after compilation. Our systematic analysis reveals that graph compilers exhibit performance patterns highly dependent on both neural architecture and batch sizes. Through fine-grained block-level experimentation, we establish that vendor-specific compilers can leverage repeated patterns in simple architectures, yielding disproportionate throughput gains as model depth increases. We introduce novel metrics to quantify a compiler's ability to mitigate performance friction as batch size increases. Our methodology bridges the gap between academic research and practical deployment by incorporating compiler effects throughout the research process, providing actionable insights for practitioners navigating complex optimization landscapes across heterogeneous hardware environments.

📄 PDF Abstract BibTeX arXiv:2504.20198

Code (0)

등록된 구현이 없습니다.

Tasks

Friction

Similar Papers 제목 키워드 기반

Operator Fusion in XLA: Analysis and Evaluation

2023-01-30 · Daniel Snider, Ruofan Liang

Machine learning (ML) compilers are an active area of research because they offer the potential to automatically speedup tensor programs. Kernel fusion is often cited as an important optimization performed by ML compiler…

Transferable Graph Optimizers for ML Compilers

2020-10-21 · NeurIPS 2020 12 · Yanqi Zhou, Sudip Roy, Amirali Abdolrashidi, Daniel Wong 외

Most compilers for machine learning (ML) frameworks need to solve many correlated optimization problems to generate efficient machine code. Current ML compilers rely on heuristics based algorithms to solve these optimiza…

Deep Reinforcement LearningGraph Neural Network

ML-driven Hardware Cost Model for MLIR

2023-02-14 · Dibyendu Das, Sandya Mannarswamy

During early optimization passes, compilers must make predictions for machine-dependent characteristics such as execution unit utilization, number of register spills, latency, throughput etc. to generate better code. Oft…

CPUGPUmodelScheduling

GeoT: Tensor Centric Library for Graph Neural Network via Efficient Segment Reduction on GPU

2024-04-03 · Zhongming Yu, Genghan Zhang, Hanxian Huang, Xin Chen 외

In recent years, Graph Neural Networks (GNNs) have ignited a surge of innovation, significantly enhancing the processing of geometric data structures such as graphs, point clouds, and meshes. As the domain continues to e…

GPUGraph Neural Network

Compiler Auto-Vectorization with Imitation Learning

2019-12-01 · NeurIPS 2019 12 · Charith Mendis, Cambridge Yang, Yewen Pu, Dr.Saman Amarasinghe 외

Modern microprocessors are equipped with single instruction multiple data (SIMD) or vector instruction sets which allow compilers to exploit fine-grained data level parallelism. To exploit this parallelism, compilers emp…

Graph Neural NetworkImitation Learning