Can I read Vortex: Efficient Sample-Free Dynamic Tensor Program Optimization via Hardware-aware Strategy Space Hierarchization on EtoBox?

Vortex: Efficient Sample-Free Dynamic Tensor Program Optimization via Hardware-aware Strategy Space Hierarchization by Zhou, Yangjie; Zhu, Honglin; Qiu, Qian; Cui, Weihao; Liu, Zihan; Guo, Cong; Feng, Siyuan; Meng, Jintao; Lan, Haidong; Leng, Jingwen; Zhu, Wenxi; Deng, Minwen is a scholarly article available to read on EtoBox.

What is Vortex: Efficient Sample-Free Dynamic Tensor Program Optimization via Hardware-aware Strategy Space Hierarchization about?

Dynamic-shape deep neural networks (DNNs) are rapidly evolving, attracting attention for their ability to handle variable input sizes in real-time applications. However, existing compilation optimization methods for such networks often rely heavily on predefined samples to guide the compilation process, which restricts their adaptability and efficiency. These sample-driven methods struggle to efficiently manage the diverse and unpredictable shapes encountered in real-world scenarios, often resulting in suboptimal performance. To tackle these issues, we introduce Vortex, a hardware-driven and sample-free compiler tailored for dynamic-shape tensor programs. Vortex capitalizes on detailed hardware information and hierarchizes the strategy space to facilitate high-performance code generation without relying on runtime shape samples. It features a unique bidirectional compilation workflow, combining top-down abstraction for aligning tensor program execution with hardware hierarchies and bottom-up kernel construction to narrow the search space, enabling Vortex to achieve remarkable efficiency. Comprehensive evaluations confirm that Vortex reduces compilation time by $176\times$ compared

Author

Zhou, Yangjie; Zhu, Honglin; Qiu, Qian; Cui, Weihao; Liu, Zihan; Guo, Cong; Feng, Siyuan; Meng, Jintao; Lan, Haidong; Leng, Jingwen; Zhu, Wenxi; Deng, Minwen

Published

2024

Language

EN