Edge computing is gaining traction due to its scalability benefits. However, deploying convolutional neural networks (CNNs) on resource-constrained IoT devices poses challenges for real-time performance. To address th...
详细信息
ISBN:
(纸本)9798400716751
Edge computing is gaining traction due to its scalability benefits. However, deploying convolutional neural networks (CNNs) on resource-constrained IoT devices poses challenges for real-time performance. To address these issues, we propose DeepDecompose, a distributed inference framework for CNNs on GPU-equipped IoT clusters. DeepDecompose optimizes parallelism and GPU utilization to reduce inference latency. Deployed on Nvidia Jetson Nano, a GPU-equipped IoT device, DeepDecompose divides CNN models into smaller sub-models for independent inference, resulting in up to 2x faster performance compared to single-GPU computation.
Withthe development of new technologies such as big data, cloud computing, internet of things, mobile internet, and artificial intelligence, the connotation of integrated energy services continues to expand. Integrat...
详细信息
this study explores the integration of Educational Robotics (ER) and the internet of things (IoT) in learning environments, highlighting their collective impact on educational practices. It assesses ER and IoT's a...
详细信息
ISBN:
(纸本)9798350369458;9798350369441
this study explores the integration of Educational Robotics (ER) and the internet of things (IoT) in learning environments, highlighting their collective impact on educational practices. It assesses ER and IoT's application, challenges, and opportunities, offering insights into their role in enhancing pedagogy and learning outcomes. Specifically, our exploration uncovers the transformative potential of ER and IoT in fostering critical thinking, problem-solving skills, and digital literacy among students and reveals the pivotal role these technologies play in preparing learners for the demands of the digital age and Industry 4.0, while also highlighting the need for strategic implementation and teacher support. Our findings elucidate the potential of ER and IoT to innovate teaching strategies and curriculum design, serving as a guide for educational stakeholders, such as curriculum developers, educators, researchers, and policymakers, in leveraging these technologies to revolutionize instructional practices.
Withthe evolution of networking technology, RDMA (Remote Direct Memory Access) has become an increasingly critical component in data centers. Its zero-copy and kernel-bypass capabilities significantly boost network t...
详细信息
ISBN:
(纸本)9798400716751
Withthe evolution of networking technology, RDMA (Remote Direct Memory Access) has become an increasingly critical component in data centers. Its zero-copy and kernel-bypass capabilities significantly boost network transmission speeds and reduce communication latency, leading to software layers such as drivers and libraries becoming the new bottlenecks. distributed storage also plays a vital role in data centers, and integrating RDMA into traditional distributed storage clusters presents a promising direction. However, distributed storage involves complex cluster designs, like hash index, making the deployment and effective utilization of features of RDMA a substantial challenge. Researchers must invest considerable effort to thoroughly understand these complexities. this paper organizes the principles of RDMA and its relation to distributed storage and challenges faced. We summarize recent research in the deployment of RDMA, covering aspects such as RDMA-based distributed file systems, distributed key-value storage systems, and compute offload using smart NICs. Finally, we look ahead at the future research of RDMA in distributed storage, offering direction and insight for ongoing and future research in this evolving field.
In the face of the problem of sensors placement under the complex geographical environment and traditions of the battlefield, the existing distributed detection station optimization methods have not considered the imp...
详细信息
ISBN:
(纸本)9798400716751
In the face of the problem of sensors placement under the complex geographical environment and traditions of the battlefield, the existing distributed detection station optimization methods have not considered the impact of time-varying electromagnetic environment on the positioning accuracy, and the optimization algorithm itself is also prone to fall into local convergence. A new method based on the identification of radiation sources and the optimization of network topology is proposed. Firstly, the electromagnetic environment dataset and regional constraints are established from the specific terrain conditions of the test area. Secondly, the system design indexes such as the Signal to Noise Ratio (SNR) of detection, the average detection probability and the effective location area are designed, and the target function of the multi-indicator detection station location optimization is established. Finally, the model is solved by stepwise convex optimization method. the proposed method improves the environmental adaptability of the optimization model. In addition, the system can obtain the optimal scattering characteristics under the time-varying electromagnetic environment.
the internet of Energy (IoE) is a distributed paradigm that leverages smart networks and distributed system technologies to enable decentralized energy systems. In contrast to the traditional centralized energy system...
详细信息
the optimization problem of resource allocation and task scheduling involves effectively utilizing limited computing resources and arranging user service requests or tasks in a distributedcomputing environment to ach...
详细信息
ISBN:
(纸本)9789819743896;9789819743902
the optimization problem of resource allocation and task scheduling involves effectively utilizing limited computing resources and arranging user service requests or tasks in a distributedcomputing environment to achieve improved performance, reduced costs, and energy savings. Withthe advancements in technologies such as the internet of things and 5G, edge computing has emerged as a new paradigm that deploys data processing and analysis capabilities closer to the data source, enabling faster, more secure, and reliable services. Consequently, the widespread adoption of edge computing presents new challenges and opportunities for resource allocation and task scheduling optimization, including the management of massive data streams, adaptation to dynamic network environments, and coordination between edge and cloud computing. this paper summarizes the issues, evaluation dimensions and methods of resource allocation and task scheduling optimization in edge computing and identifies future prospects and challenges in this area.
Cloud computing is the practice of offering services like servers, storage, analytics, and databases to networks or computer systems worldwide. With enhanced technical infrastructure, internet users can now use comput...
详细信息
As a distributed machine learning framework, federated learning has received considerable attention in recent years and has been researched and applied in various scenarios. However, the system heterogeneity due to th...
详细信息
ISBN:
(纸本)9798400716751
As a distributed machine learning framework, federated learning has received considerable attention in recent years and has been researched and applied in various scenarios. However, the system heterogeneity due to the physical characteristics of various terminal devices has led to the straggler effect, making the practical implementation of federated learning challenging. therefore, we propose a semi-asynchronous federated optimization method based on buffer pre-aggregation. this method allows every participant to engage in training through pre-aggregation and establishes a training time framework based on the pre-aggregation time. It updates the model adaptively using a semi-asynchronous communication method combined with lag factors, improving communication efficiency while maintaining stable accuracy. Experimental results on datasets demonstrate that our proposed method can effectively accelerate the training process of federated learning compared to existing federated optimization methods.
Nowadays, emerging technologies such as smart terminals, cloud computing, and the internet of things are developing rapidly. the"One-to-one" encrypted transmission method is difficult to meet the communicati...
详细信息
暂无评论