A Deep Learning Compiler for Vector Processor
| العنوان: | A Deep Learning Compiler for Vector Processor |
|---|---|
| المؤلفون: | Jun Wu, Songyuan Zhao, Haoqi Ren, Zhifeng Zhang, Pingping Pan |
| المصدر: | Communications and Networking ISBN: 9783030677190 ChinaCom |
| بيانات النشر: | Springer International Publishing, 2021. |
| سنة النشر: | 2021 |
| مصطلحات موضوعية: | Intrinsic function, Computer science, business.industry, Deep learning, Process (computing), Parallel computing, computer.software_genre, Vector processor, Very long instruction word, Code generation, Compiler, Artificial intelligence, business, computer, Software architecture description |
| الوصف: | The technical route of machine learning compiler generally refers to the application of automatic or semi-automatic code generation in the optimization process instead of hand-optimization. This paper presents a deep learning compiler (DLCS) for target vector processor based on LLVM framework, which lowers deep learning (DL) models to an intermediate representation (IR) of two levels. The high-level IR realizes target-independent optimizations including kernel fusion, data replacement and data simplification, while the low-level IR allows the compiler to perform target-dependent optimizations, such as Eight-Slots VLIW and special intrinsic function. The proposed compiler customizes the architecture description of target vector processor to achieve a high-quality automatic code generation. We evaluate the performance comparison between DLCS and hand-optimization when deploying ResNet-18 model and MobileNet model to the target vector processor. Experimental results show that DLCS offers Multi-slot parallel performance for target vector processor and achieves speedups ranging from 1.5× to 3.0× over existing frameworks backed by hand-optimized libraries. |
| ردمك: | 978-3-030-67719-0 |
| DOI: | 10.1007/978-3-030-67720-6_46 |
| URL الوصول: | https://explore.openaire.eu/search/publication?articleId=doi_________::95e7391b6f95e33bb14cc1950936fc15 https://doi.org/10.1007/978-3-030-67720-6_46 |
| Rights: | CLOSED |
| رقم الانضمام: | edsair.doi...........95e7391b6f95e33bb14cc1950936fc15 |
| قاعدة البيانات: | OpenAIRE |
| ردمك: | 9783030677190 |
|---|---|
| DOI: | 10.1007/978-3-030-67720-6_46 |