当前位置:首页 > 报告详情

Edgenuity:边缘与人工智能的交汇点.pdf

上传人: a****d 编号:185009 2024-10-07 8页 632.39KB

1、1|2024 SNIA.All Rights Reserved.Edgenuity:The Intersection of Edge and AIPresented by Jeff White Dell Technologies2|2024 SNIA.All Rights Reserved.Modern Edge Computation PresentFutureBoilerPumpjackCompressorIoT GatewayIoT GatewayIoT GatewayIndustrial PCIndustrial PCIndustrial PCDatacenterDatacenterD

2、atacenterBoilerPumpjackCompressorIoT GatewayIoT GatewayIoT GatewaySensorSensorSensorControllerControllerControllerSoftControllerSensorSensorSensorEdge NativeComputation1)Common System Platform.2)Higher Computational Power.3)Shared Data/Networking.4)Inherent MultiCloud Native.5)Open System.6)Remote S

3、upport.7)Unified Security Policy.8)Software Defined Concurrency.9)Power Efficient.10)Dynamic Twin/AI/ML Enabled.Modern Edges must secure,simplify and automate complex operations and technology from Edge Through Cloud1)Bespoke Hardware Control.2)Fragmented Systems.3)Limited Data Sharing.4)Inflexible

4、To Change.5)Closed Systems.6)Onsite Expert Support Required.7)Individual Security Policy.3|2024 SNIA.All Rights Reserved.Enterprise Verticals and Telco/Emerging Use CasesEnterprise Edge Manufacturing Retail Energy Transport/Logistics Digital Cities Autonomous Systems Connected and Autonomous Vehicle

5、s Robotics UAV/UUV Soft Robots/RPATelecommunications NFVi Open RAN CDN Virtualized CPESimulation Extended Reality Consumer XR Remote Services Digital Twins/Simulation 2D to 3D Image construction Public Sector/Defense Applications Require Aspects of All AboveEmerging4|2024 SNIA.All Rights Reserved.AI

6、 Technology Frameworks Artificial Intelligence is an area of Computer Science that utilizes functions that can be trained to perform a variety of statistical/probabilistic tasks classification,ranking,regression,clustering graph/geometry discovery,density estimation,etc.AI is an universal function a

word格式文档无特别注明外均可编辑修改,预览文件经过压缩,下载原文更清晰!
三个皮匠报告文库所有资源均是客户上传分享,仅供网友学习交流,未经上传用户书面授权,请勿作商用。
本文介绍了Dell Technologies的Edgenuity解决方案,强调了现代边缘计算的重要性,并展望了边缘计算的未来。文章提到,边缘计算需要具备1)通用系统平台,2)高计算能力,3)共享数据/网络,4)内建多云原生特性,5)开放系统,6)远程支持,7)统一安全策略,8)软件定义的并行计算,9)节能,10)支持动态孪生/AI/ML等特性。与传统边缘系统相比,现代边缘计算能更好地保障、简化并自动化复杂操作和技术,包括自定义硬件控制、整合系统、灵活的数据共享、适应性变化、开放系统、远程支持、统一安全政策等。 在AI技术方面,文章概述了生成式、判别式和强化学习等框架,并指出这些技术都适用于边缘计算。此外,还提到了机器学习、符号学习和多模态自动编码器等AI技术在边缘计算中的应用。文章最后讨论了边缘AI的未来趋势,包括多模态生成式和判别式技术、基于通信代理的具有机器推理能力的系统、交互式分布式训练/微调/模型增强,以及AI安全等。
"边缘计算如何助力现代企业?" "AI技术在边缘计算中的应用有哪些?" "未来边缘AI的发展趋势是什么?"
客服
商务合作
小程序
服务号
折叠