Zobrazeno 1 - 10
of 69
pro vyhledávání: '"Guo, Chuanxiong"'
Autor:
Kong, Xinhao, Zhu, Yibo, Zhou, Huaping, Jiang, Zhuo, Ye, Jianxi, Guo, Chuanxiong, Zhuo, Danyang
High-speed RDMA networks are getting rapidly adopted in the industry for their low latency and reduced CPU overheads. To verify that RDMA can be used in production, system administrators need to understand the set of application workloads that can po
Externí odkaz:
http://arxiv.org/abs/2304.11467
Autor:
Hu, Hanpeng, Jiang, Chenyu, Zhong, Yuchen, Peng, Yanghua, Wu, Chuan, Zhu, Yibo, Lin, Haibin, Guo, Chuanxiong
Distributed training using multiple devices (e.g., GPUs) has been widely adopted for learning DNN models over large datasets. However, the performance of large-scale distributed training tends to be far from linear speed-up in practice. Given the com
Externí odkaz:
http://arxiv.org/abs/2205.02473
Companies build separate training and inference GPU clusters for deep learning, and use separate schedulers to manage them. This leads to problems for both training and inference: inference clusters have low GPU utilization when the traffic load is l
Externí odkaz:
http://arxiv.org/abs/2202.07896
Autor:
Liu, Heting, Li, Zhichao, Tan, Cheng, Yang, Rongqiu, Cao, Guohong, Liu, Zherui, Guo, Chuanxiong
Graphics processing units (GPUs) are the de facto standard for processing deep learning (DL) tasks. Meanwhile, GPU failures, which are inevitable, cause severe consequences in DL tasks: they disrupt distributed trainings, crash inference services, an
Externí odkaz:
http://arxiv.org/abs/2201.11853
Autor:
Liu, Tianfeng, Chen, Yangrui, Li, Dan, Wu, Chuan, Zhu, Yibo, He, Jun, Peng, Yanghua, Chen, Hongzheng, Chen, Hongzhi, Guo, Chuanxiong
Graph neural networks (GNNs) have extended the success of deep neural networks (DNNs) to non-Euclidean graph data, achieving ground-breaking performance on various tasks such as node classification and graph property prediction. Nonetheless, existing
Externí odkaz:
http://arxiv.org/abs/2112.08541
Serving DNN Models with Multi-Instance GPUs: A Case of the Reconfigurable Machine Scheduling Problem
Autor:
Tan, Cheng, Li, Zhichao, Zhang, Jian, Cao, Yu, Qi, Sikai, Liu, Zherui, Zhu, Yibo, Guo, Chuanxiong
Multi-Instance GPU (MIG) is a new feature introduced by NVIDIA A100 GPUs that partitions one physical GPU into multiple GPU instances. With MIG, A100 can be the most cost-efficient GPU ever for serving Deep Neural Networks (DNNs). However, discoverin
Externí odkaz:
http://arxiv.org/abs/2109.11067
Autor:
Jin, Yuchen, Zhou, Tianyi, Zhao, Liangyu, Zhu, Yibo, Guo, Chuanxiong, Canini, Marco, Krishnamurthy, Arvind
The learning rate (LR) schedule is one of the most important hyper-parameters needing careful tuning in training DNNs. However, it is also one of the least automated parts of machine learning systems and usually costs significant manual effort and co
Externí odkaz:
http://arxiv.org/abs/2105.10762
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.