Superneurons: dynamic GPU memory management for training deep neural networks
Name
1801.04380.pdf
Description
Accepted version
Size
1.34 MB
Format
Unknown
Checksum (MD5)
81d5c4e21fc7f8b6db4c3be5ea6ac647
Author(s)
Wang, Linnan
Ye, Jinmian
Zhao, Yiyang
Wu, Wei
Li, Ang
Song, Shuaiwen Leon
Xu, Zenglin
Kraska, Tim
Date Issued
2018
Journal
ACM SIGPLAN Notices
Publisher
Association for Computing Machinery (ACM)
Version
Original manuscript
Abstract
© 2018 ACM. Going deeper and wider in neural architectures improves their accuracy, while the limited GPU DRAM places an undesired restriction on the network design domain. Deep Learning (DL) practitioners either need to change to less desired network architectures, or nontrivially dissect a network across multiGPUs. These distract DL practitioners from concentrating on their original machine learning tasks. We present SuperNeurons: a dynamic GPU memory scheduling runtime to enable the network training far beyond the GPU DRAM capacity. SuperNeurons features 3 memory optimizations, Liveness Analysis, Unified Tensor Pool, and Cost-Aware Recomputation; together they effectively reduce the network-wide peak memory usage down to the maximal memory usage among layers. We also address the performance issues in these memory-saving techniques. Given the limited GPU DRAM, SuperNeurons not only provisions the necessary memory for the training, but also dynamically allocates the memory for convolution workspaces to achieve the high performance. Evaluations against Caffe, Torch, MXNet and TensorFlow have demonstrated that SuperNeurons trains at least 3.2432 deeper network than current ones with the leading performance. Particularly, SuperNeurons can train ResNet2500 that has 104 basic network layers on a 12GB K40c.
MIT Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Terms of Use
Creative Commons Attribution-Noncommercial-Share Alike
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1145/3178487.3178491
