.png){kind=icon}
返回
Federated Clouds for Efficient Multitasking in Distributed Artificial Intelligence Applications
.png){kind=icon}
PRE
AI
DOI：10.1109/TCC.2022.3184157.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Distributed cloud/edge resources are needed to execute pervasive artificial intelligence tasks, collectively. The AI workload and data sets have variable multitasking granularity, privacy constraints, and communication latency concerns. This article presents a novel federated cloud/edge (FCE) framework, illustrated by distributed medical image processing across multiple hospital sites. This federated cloud system appeals to train many machine learning models efficiently with workload balancing and reduced communication overheads. We tested the FCE model on a multi-cloud platform recently built at the Chinese University of Hong Kong in Shenzhen. We claim three distinct advantages in using the FCE system. First, our federated cloud system results in 41.3% reduction in total AI processing time in large-scale ML/DL experiments. Second, high machine model accuracy was achieved at 87% level in telemedicine experiments. The virtual graph helps reduce internode traffic latencies to avoid ML inference slowdowns. Third, the system can tolerate multiple cloud failures to enter a graceful degradation mode in case of node failures. The scalable performance gains in AI processing speed, model accuracy, and fault tolerance make our federated clouds a truly viable approach to solving massive AI multitasking problems in pervasive AI applications.
Keyword:
Cloud computing
Hospitals
Computational modeling
Servers
Atmospheric modeling
Training
Software defined networking
Cloud computing
federated machine learning
graphs and networks
latency reduction
and fault tolerance
期刊
I
IF:
5
论文数:
1.9K
被引数:
4.3K
机构
引用论文
Adult hippocampal neurogenesis poststroke: More new granule cells but aberrant morphology and impaired spatial memory.png)
PLOS ONE
OAAI
PLOS ONEIF0
OAAI分享收藏An unusual selectivity in Pd catalyzed cross-coupling of terminal alkynes with “Unactivated” alkynesOAAI
OAAI分享收藏分享收藏分享收藏Efficient Embedding of Virtual Networks to Distributed Clouds via Exploring Periodic Resource Demands通过探索周期性资源需求将虚拟网络高效嵌入到分布式云
PREAI
分享收藏Toward Resource-Efficient Federated Learning in Mobile Edge Computing在移动边缘计算中实现资源高效的联合学习IEEE NETWORK
IEEE NETWORKIF6.3
PREAI
分享收藏
查看更多内容