Abstract | ||
---|---|---|
Graphic Processing Units (GPUs) have limited memory capacity. Training popular deep neural networks (DNNs) often requires a larger amount of memory than that a GPU may have. Consequently, training data needs to be swapped between CPUs and GPUs. Data swapping may become a bottleneck when its latency is longer than the latency of DNN computations. Tensor compression in GPUs can reduce the data swapp... |
Year | DOI | Venue |
---|---|---|
2021 | 10.1109/Cluster48925.2021.00019 | 2021 IEEE International Conference on Cluster Computing (CLUSTER) |
Keywords | DocType | ISSN |
Training,Deep learning,Tensors,Runtime,Memory management,Neural networks,Graphics processing units | Conference | 1552-5244 |
ISBN | Citations | PageRank |
978-1-7281-9666-4 | 0 | 0.34 |
References | Authors | |
0 | 8 |
Name | Order | Citations | PageRank |
---|---|---|---|
Ping Chen | 1 | 0 | 0.34 |
Shuibing He | 2 | 109 | 20.45 |
Xuechen Zhang | 3 | 292 | 21.94 |
Shuaiben Chen | 4 | 0 | 0.34 |
Peiyi Hong | 5 | 0 | 0.34 |
Yanlong Yin | 6 | 134 | 8.93 |
Xian-He Sun | 7 | 3 | 3.09 |
Gang Chen | 8 | 712 | 75.60 |