一句话：提出上下文感知、跨平台跨框架的 DeepContext 工具，关联深度学习全栈执行以定位性能瓶颈并给出优化建议。
深度学习模型在异构计算环境中的性能优化需要贯穿高层算子到底层执行的全面剖析，但现有工具难以关联 Python 代码、深度学习框架、C/C++ 库与 GPU 执行之间的语义鸿沟。DeepContext 提出一种上下文感知的剖析系统：通过框架无关的 shim 层把深度学习框架行为与硬件性能指标关联起来，并基于整个软件栈的全局视图自动分析性能并给出可操作的优化建议。它支持主流深度学习框架，运行于现代 CPU+GPU 架构且开销低。评估表明，DeepContext 能发现真实应用中此前隐藏的性能瓶颈；在其指导下修复多个性能问题，获得 1.06× 到 1.66× 的加速。
Effective performance optimization of deep learning models requires comprehensive profiling across heterogeneous computing environments, yet existing tools fail to bridge the semantic gap between high-level operations and low-level execution. This paper presents DeepContext, a novel profiling system that correlates program contexts across Python code, deep learning frameworks, C/C++ libraries, and GPU execution. DeepContext features a framework-agnostic shim layer that seamlessly correlates the behavior of the deep learning framework with hardware performance metrics. Furthermore, DeepContext provides an automated performance analyzer that offers actionable optimization guidance based on its holistic view of the entire software stack of deep learning applications. DeepContext works for mainstream deep learning frameworks and runs on modern CPU+GPU architectures with low overhead. Our evaluation demonstrates that DeepContext uncovers previously hidden performance bottlenecks in real-world deep-learning applications. Guided by DeepContext, we are able to fix multiple performance issues, achieving speed-ups between 1.06× and 1.66×.
DOI 原文 · arXiv · PDF（开放获取） · DBLP
@inproceedings{ZhaoWHY00025,
title = {DeepContext: A Context-aware, Cross-platform, and Cross-framework Tool for Performance Profiling and Analysis of Deep Learning Workloads},
author = {Qidong Zhao and Hao Wu and Yueming Hao and Zilingfeng Ye and Jiajia Li and Xu Liu and Keren Zhou},
booktitle = {ASPLOS},
year = {2025},
doi = {10.1145/3676642.3736127}
}