2023 IEEE International Conference on Edge Computing and Communications (EDGE)最新文献_第3页

Edge Computing Tasks Orchestration: An Energy-Aware Approach 边缘计算任务编排:能量感知方法

2023 IEEE International Conference on Edge Computing and Communications (EDGE) Pub Date : 2023-07-01 DOI: 10.1109/EDGE60047.2023.00027

Johan Løhde Thomsen, Kristian Dragsbæk Schmidt Thomsen, R. B. Schmidt, Søren D. Jakobsgaard, Thor Beregaard, M. Albano, Sergio Moreschini, D. Taibi

引用次数: 0

Intelligent Multi-Domain Edge Orchestration for Highly Distributed Immersive Services: An Immersive Virtual Touring Use Case 高度分布式沉浸式服务的智能多域边缘编排:沉浸式虚拟漫游用例

2023 IEEE International Conference on Edge Computing and Communications (EDGE) Pub Date : 2023-07-01 DOI: 10.1109/EDGE60047.2023.00061

Tarik Zakaria Benmerar, T. Theodoropoulos, Diogo Fevereiro, Luis Rosa, João Rodrigues, T. Taleb, Paolo Barone, K. Tserpes, Luís Cordeiro

{"title":"Intelligent Multi-Domain Edge Orchestration for Highly Distributed Immersive Services: An Immersive Virtual Touring Use Case","authors":"Tarik Zakaria Benmerar, T. Theodoropoulos, Diogo Fevereiro, Luis Rosa, João Rodrigues, T. Taleb, Paolo Barone, K. Tserpes, Luís Cordeiro","doi":"10.1109/EDGE60047.2023.00061","DOIUrl":"https://doi.org/10.1109/EDGE60047.2023.00061","url":null,"abstract":"Edge cloud technologies in tandem with AI-enabled solutions can contribute to overcoming the challenges that pertain the distributed execution of immersive services and contribute towards providing a positive experience for the end-users. Intelligent resource management, orchestration, and prediction systems can optimize the deployment of services, adapt to changing demands, and ensure that the services are running smoothly. This paper introduces a novel architectural paradigm capable of facilitating multi-domain edge orchestration for highly distributed immersive services by incorporating a plethora of AI solutions and technological enablers that can support multi-domain edge deployments. The proposed architecture is designed to operate on the basis of multi-level specification blueprints, which decouple the simple high-level user-intent infrastructure definition from the AI-driven orchestration and the final execution plan. The Application Management Framework (AMF) offers a visual language and tool that can be used as an alternative to a formal method for creating the intent blueprint. In the frame of this work, the latter is validated by an immersive virtual touring use-case scenario.","PeriodicalId":369407,"journal":{"name":"2023 IEEE International Conference on Edge Computing and Communications (EDGE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129143626","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Towards Intelligent Data Protocols for the Edge 面向边缘的智能数据协议

2023 IEEE International Conference on Edge Computing and Communications (EDGE) Pub Date : 2023-07-01 DOI: 10.1109/EDGE60047.2023.00060

Praveen Kumar Donta, S. Dustdar

{"title":"Towards Intelligent Data Protocols for the Edge","authors":"Praveen Kumar Donta, S. Dustdar","doi":"10.1109/EDGE60047.2023.00060","DOIUrl":"https://doi.org/10.1109/EDGE60047.2023.00060","url":null,"abstract":"The computing continuum is growing because multiple devices are added daily. Edge devices play a key role in this because computation is decentralized or distributed. Edge computing is advanced by using AI/ML algorithms to become more intelligent. Besides, Edge data protocols are useful for transmitting or receiving data between devices. Since, computation efficiency is possible when the data is received at the Edge timely, and it is possible only when the data protocols are efficient, reliable and fast. Most edge data protocols are defined with static set of rules and their primary purpose is to provide standardized and reliable data communications. Edge devices need autonomous or dynamic protocols that enable interoperability, autonomous decision making, scalability, and adaptability. This paper examines the limitations of popular data protocols used in edge networks, the need for intelligent data protocols, and their implications. We also explore possible ways to simplify learning for edge devices and discuss how intelligent data protocols can mitigate challenges such as congestion, message filtering, message expiration, prioritization, and resource handling.","PeriodicalId":369407,"journal":{"name":"2023 IEEE International Conference on Edge Computing and Communications (EDGE)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127648611","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 2

ConPrEF: A Context-based Privacy Enforcement Framework for Edge Computing 基于上下文的边缘计算隐私执行框架

2023 IEEE International Conference on Edge Computing and Communications (EDGE) Pub Date : 2023-07-01 DOI: 10.1109/EDGE60047.2023.00022

Giorgia Sirigu, B. Carminati, E. Ferrari

引用次数: 0

Spica: Exploring FPGA Optimizations to Enable an Efficient SpMV Implementation for Computations at Edge Spica:探索FPGA优化以实现有效的边缘计算SpMV实现

2023 IEEE International Conference on Edge Computing and Communications (EDGE) Pub Date : 2023-07-01 DOI: 10.1109/EDGE60047.2023.00018

Dheeraj Ramchandani, Bahar Asgari, Hyesoon Kim

{"title":"Spica: Exploring FPGA Optimizations to Enable an Efficient SpMV Implementation for Computations at Edge","authors":"Dheeraj Ramchandani, Bahar Asgari, Hyesoon Kim","doi":"10.1109/EDGE60047.2023.00018","DOIUrl":"https://doi.org/10.1109/EDGE60047.2023.00018","url":null,"abstract":"With the emergence of FPGA boards equipped with high bandwidth memory (HBM), these boards have become more attractive for implementing memory-intensive computational kernels such as sparse matrix-vector multiplication (SpMV), with a wide range of applications in edge computations from deep learning to robotics. Specialized implementation of SpMV on FPGAs enables efficient utilization of the limited resources in edge systems. High-level synthesis (HLS) compilers, on the other hand, have eased the programming of FPGAs, leading to a faster development cycle. Even though the programming of FPGAs has become easier, obtaining maximum throughput even for the straightforward kernel of SpMV still requires careful optimizations. Therefore, this paper explores the impact of deploying various optimization techniques such as temporal parallelism, spatial parallelism, and memory alignment to help SpMV fully utilize the available memory bandwidth of HBM on a Xilinx FPGA board to achieve close-to-peak throughput without wasting the resources. We conclude the optimizations by suggesting Spica, a high-throughput tree-based SpMV implementation.","PeriodicalId":369407,"journal":{"name":"2023 IEEE International Conference on Edge Computing and Communications (EDGE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128744357","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Creating Robust Deep Neural Networks with Coded Distributed Computing for IoT 为物联网创建具有编码分布式计算的鲁棒深度神经网络

2023 IEEE International Conference on Edge Computing and Communications (EDGE) Pub Date : 2023-07-01 DOI: 10.1109/EDGE60047.2023.00029

Ramyad Hadidi, Jiashen Cao, Bahar Asgari, Hyesoon Kim

{"title":"Creating Robust Deep Neural Networks with Coded Distributed Computing for IoT","authors":"Ramyad Hadidi, Jiashen Cao, Bahar Asgari, Hyesoon Kim","doi":"10.1109/EDGE60047.2023.00029","DOIUrl":"https://doi.org/10.1109/EDGE60047.2023.00029","url":null,"abstract":"The increasing interest in serverless computation and ubiquitous wireless networks has led to numerous connected devices in our surroundings. Such IoT devices have access to an abundance of raw data, but their inadequate resources in computing limit their capabilities. With the emergence of deep neural networks (DNNs), the demand for the computing power of IoT devices is increasing. To overcome inadequate resources, several studies have proposed distribution methods for IoT devices that harvest the aggregated computing power of idle IoT devices in an environment. However, since such a distributed system strongly relies on each device, unstable latency, and intermittent failures, the common characteristics of IoT devices and wireless networks, cause high recovery overheads. To reduce this overhead, we propose a novel robustness method with a close-to-zero recovery latency for DNN computations. Our solution never loses a request or spends time recovering from a failure. To do so, first, we analyze how matrix computations in DNNs are affected by distribution. Then, we introduce a novel coded distributed computing (CDC) method, the cost of which, unlike that of modular redundancies, is constant when the number of devices increases. Our method is applied at the library level, without requiring extensive changes to the program, while still ensuring a balanced work assignment during distribution.","PeriodicalId":369407,"journal":{"name":"2023 IEEE International Conference on Edge Computing and Communications (EDGE)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129147468","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

SQuBA: Social Quorum Based Access Control for Open IoT Environments SQuBA:基于社会仲裁的开放物联网环境访问控制

2023 IEEE International Conference on Edge Computing and Communications (EDGE) Pub Date : 2023-07-01 DOI: 10.1109/EDGE60047.2023.00020

Yixuan Wang, A. Chandra, J. Weissman

{"title":"SQuBA: Social Quorum Based Access Control for Open IoT Environments","authors":"Yixuan Wang, A. Chandra, J. Weissman","doi":"10.1109/EDGE60047.2023.00020","DOIUrl":"https://doi.org/10.1109/EDGE60047.2023.00020","url":null,"abstract":"Internet of things (IoT) devices have been ubiquitous in recent years. An emerging model for IoT deployment is an open edge-based infrastructure. Edge resources are commonly used to coordinate capabilities and manage access due to IoT device resource limitations and IoT vendor heterogeneity. The open IoT environment often exists in a multi-user setting, where multiple users interact with a single IoT device. In this setting, we assume that none of the users or the edges are fully trusted, thus IoT data privacy may be compromised. Limited attention has been paid to authorization and auditing in this environment. However, exploiting inter-user relationships gives us leverage. In this work, we propose a social quorum based architecture, SQuBA, as an access control mechanism for IoT which provides relationship-driven authorization and auditing. We present a tiered approach to support access control rules and relationship-based trustworthiness. We implemented a prototype and carried out experiments using a real-world dataset under various scenarios and configurations. The results demonstrate both SQuBA’s promising near real-time response latency that is in the order of milliseconds, and good resilience to different edge faulty models. We also compare with various baselines and SQuBA is able to improve end-to-end latency by up to 10X and tolerate the number of faulty edges by up to 2X.","PeriodicalId":369407,"journal":{"name":"2023 IEEE International Conference on Edge Computing and Communications (EDGE)","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130474412","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Human-Centered Explainable AI at the Edge for eHealth 以人为本的可解释人工智能在电子健康的边缘

2023 IEEE International Conference on Edge Computing and Communications (EDGE) Pub Date : 2023-07-01 DOI: 10.1109/EDGE60047.2023.00044

Joydeb Dutta, Deepak Puthal

引用次数: 0

FedCime: An Efficient Federated Learning Approach For Clients in Mobile Edge Computing FedCime:移动边缘计算客户端有效的联邦学习方法

2023 IEEE International Conference on Edge Computing and Communications (EDGE) Pub Date : 2023-07-01 DOI: 10.1109/EDGE60047.2023.00042

Paul Agbaje, A. Anjum, Zahidur Talukder, Mohammad Islam, E. Nwafor, Habeeb Olufowobi

引用次数: 0

DOSA: Organic Compilation for Neural Network Inference on Distributed FPGAs 分布式fpga上神经网络推理的有机编译

2023 IEEE International Conference on Edge Computing and Communications (EDGE) Pub Date : 2023-07-01 DOI: 10.1109/EDGE60047.2023.00019

Burkhard Ringlein, F. Abel, D. Diamantopoulos, B. Weiss, C. Hagleitner, D. Fey

{"title":"DOSA: Organic Compilation for Neural Network Inference on Distributed FPGAs","authors":"Burkhard Ringlein, F. Abel, D. Diamantopoulos, B. Weiss, C. Hagleitner, D. Fey","doi":"10.1109/EDGE60047.2023.00019","DOIUrl":"https://doi.org/10.1109/EDGE60047.2023.00019","url":null,"abstract":"The computational requirements of artificial intelligence workloads are growing exponentially. In addition, more and more compute is moved towards the edge due to latency or localization constraints. At the same time, Dennard scaling has ended and Moore’s law is winding down. These trends created an opportunity for specialized accelerators including field-programmable gate arrays (FPGAs), but the poor support and usability of today’s tools prevents FPGAs from being deployed at scale for deep neural network (DNN) inference applications. In this work, we propose an organic compiler — DOSA — that drastically lowers the barrier for deploying FPGAs. DOSA builds on the operation set architecture concept and integrates the DNN accelerator components generated by existing DNN-to-FPGA frameworks to produce an overall efficient solution. DOSA starts from DNNs represented in the community standard ONNX and automatically implements model- and data-parallelism, based on the performance targets and resource footprints provided by the user. Deploying a DNN using DOSA on 9 FPGAs exhibits a speedup of up to 52 times compared to a CPU and 18 times compared to a GPU.","PeriodicalId":369407,"journal":{"name":"2023 IEEE International Conference on Edge Computing and Communications (EDGE)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125868121","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0