Abstract | ||
---|---|---|
While current deep learning frameworks are mainly optimized for dense-accessed models, they show low throughput and poor scalability in training sparse-accessed recommendation models. Our investigation shows that the poor performance is due to the parameter synchronization bottleneck. We therefore propose BiPS, a bi-tier parameter synchronization system that alleviates the parameter update and the... |
Year | DOI | Venue |
---|---|---|
2021 | 10.1109/IPDPS49936.2021.00069 | 2021 IEEE International Parallel and Distributed Processing Symposium (IPDPS) |
Keywords | DocType | ISSN |
Training,Deep learning,Distributed processing,Scalability,Graphics processing units,Computer architecture,Throughput | Conference | 1530-2075 |
ISBN | Citations | PageRank |
978-1-6654-4066-0 | 0 | 0.34 |
References | Authors | |
0 | 7 |
Name | Order | Citations | PageRank |
---|---|---|---|
Qiming Zheng | 1 | 5 | 0.73 |
Quan Chen | 2 | 175 | 21.86 |
Kaihao Bai | 3 | 0 | 0.34 |
Guo Huifeng | 4 | 134 | 15.44 |
Yong Gao | 5 | 0 | 0.34 |
Xiuqiang He | 6 | 312 | 39.21 |
Minyi Guo | 7 | 3969 | 332.25 |