Strip Wireless Sensor Networks(SWSNs)have drawn much attention in many applications such as monitoring rivers,highways and coal mines.Packet delivery in SWSN usually requires a large number of multi-hop transmissions ...Strip Wireless Sensor Networks(SWSNs)have drawn much attention in many applications such as monitoring rivers,highways and coal mines.Packet delivery in SWSN usually requires a large number of multi-hop transmissions which leads to long transmission latency in low-duty-cycle SWSNs.Several pipeline scheduling schemes have been proposed to reduce latency.However,when communication links are unreliable,pipeline scheduling is prone to failure.In this paper,we propose a pipeline scheduling transmission protocol based on constructive interference.The protocol first divides the whole network into multiple partitions and uses a pipelined mechanism to allocate active time slots for each partition.The nodes in the same partition wake up at the same time for concurrent transmission.Multiple identical signals interfere constructively at the receiver node,which enhances received signal strength and improves link quality.Simulations show that the proposed scheme can significantly reduce the transmission latency while maintaining low energy consumption compared with other schemes.展开更多
Deep learning has become the cornerstone of artificial intelligence,playing an increasingly important role in human production and lifestyle.However,as the complexity of problem-solving increases,deep learning models ...Deep learning has become the cornerstone of artificial intelligence,playing an increasingly important role in human production and lifestyle.However,as the complexity of problem-solving increases,deep learning models become increasingly intricate,resulting in a proliferation of large language models with an astonishing number of parameters.Pipeline model parallelism(PMP)has emerged as one of the mainstream approaches to addressing the significant challenge of training“big models”.This paper presents a comprehensive review of PMP.It covers the basic concepts and main challenges of PMP.It also comprehensively compares synchronous and asynchronous pipeline schedules for PMP approaches,and discusses the main techniques to achieve load balance for both intra-node and inter-node training.Furthermore,the main techniques to optimize computation,storage,and communication are presented,with potential research directions being discussed.展开更多
基金This work is supported in part by the National Natural Science Foundation of China(Grant No.61672282)the Basic Research Program of Jiangsu Province(Grant No.BK20161491).
文摘Strip Wireless Sensor Networks(SWSNs)have drawn much attention in many applications such as monitoring rivers,highways and coal mines.Packet delivery in SWSN usually requires a large number of multi-hop transmissions which leads to long transmission latency in low-duty-cycle SWSNs.Several pipeline scheduling schemes have been proposed to reduce latency.However,when communication links are unreliable,pipeline scheduling is prone to failure.In this paper,we propose a pipeline scheduling transmission protocol based on constructive interference.The protocol first divides the whole network into multiple partitions and uses a pipelined mechanism to allocate active time slots for each partition.The nodes in the same partition wake up at the same time for concurrent transmission.Multiple identical signals interfere constructively at the receiver node,which enhances received signal strength and improves link quality.Simulations show that the proposed scheme can significantly reduce the transmission latency while maintaining low energy consumption compared with other schemes.
基金supported in part by the National Natural Science Foundation of China under Grant Nos.62025208,U21A20473,U21A20513,62076154,and 62302512the State Administration of Science,Technology,and Industry for National Defense of China under Grant No.WDZC20235250118.
文摘Deep learning has become the cornerstone of artificial intelligence,playing an increasingly important role in human production and lifestyle.However,as the complexity of problem-solving increases,deep learning models become increasingly intricate,resulting in a proliferation of large language models with an astonishing number of parameters.Pipeline model parallelism(PMP)has emerged as one of the mainstream approaches to addressing the significant challenge of training“big models”.This paper presents a comprehensive review of PMP.It covers the basic concepts and main challenges of PMP.It also comprehensively compares synchronous and asynchronous pipeline schedules for PMP approaches,and discusses the main techniques to achieve load balance for both intra-node and inter-node training.Furthermore,the main techniques to optimize computation,storage,and communication are presented,with potential research directions being discussed.
