Wk05 Cloud Computing for AI-enabled IoT_student.pdf

Cloud Computing for AI- enabled IoT The Future of Smart Systems Introduction to Cloud Computing What is cloud computing? What does cloud computing mean to you? Why do we n...

Cloud Computing for AI- enabled IoT The Future of Smart Systems Introduction to Cloud Computing What is cloud computing? What does cloud computing mean to you? Why do we need to use cloud computing for AI-enabled IoT? Please type your thoughts on Padlet. https://padlet.com/enpauli/eie3127wk5 © 2019 Amazon Web Services, Inc. or its Affiliates. All rights reserved. 3 Cloud computing defined Cloud computing is the on-demand delivery of compute power, database, storage, applications, and other IT resources via the internet with pay-as- you-go pricing. © 2019 Amazon Web Services, Inc. or its Affiliates. All rights reserved. 4 Infrastructure as software Cloud computing enables you to stop thinking of your infrastructure as hardware, and instead think of (and use) it as software. © 2019 Amazon Web Services, Inc. or its Affiliates. All rights reserved. 5 Traditional computing model Infrastructure as hardware Hardware solutions: Require space, staff, physical security, planning, capital expenditure Have a long hardware procurement cycle Require you to provision capacity by guessing theoretical maximum peaks © 2019 Amazon Web Services, Inc. or its Affiliates. All rights reserved. 6 Cloud computing model Infrastructure as software Software solutions: Are flexible Can change more quickly, easily, and cost-effectively than hardware solutions Eliminate the undifferentiated heavy- lifting tasks © 2019 Amazon Web Services, Inc. or its Affiliates. All rights reserved. 7 Cloud service models IaaS PaaS SaaS (infrastructure as a (platform as a (software as a service) service) service) More control Less control over IT resources over IT resources © 2019 Amazon Web Services, Inc. or its Affiliates. All rights reserved. 8 Cloud computing deployment models Cloud Hybrid On-premises (private cloud) © 2019 Amazon Web Services, Inc. or its Affiliates. All rights reserved. 9  Cloud computing is the Section 1 key delivery of IT resources via the internet with pricing. takeaways Cloud computing enables you to think of (and use) your as software. There are three cloud service models:. There are three cloud deployment models:. Almost anything you can implement with traditional IT can also be implemented as an AWS cloud computing service. 10 © 2019 Amazon Web Services, Inc. or its Affiliates. All rights reserved. Why do we need to use cloud computing for AI- enabled IoT? Advantages of cloud computing for IoT and AI integration Benefits of Using Cloud Computing for AI-Enabled IoT Processing Power AI requires significant computing power to analyze large amounts of data and train complex models. Storage and Data Management Cloud computing provides flexible and scalable storage options for IoT applications that generate huge amounts of data. Data Analytics and AI Services Pre-built AI Services Big Data Analytics Benefits of Using Cloud Computing for AI-Enabled IoT Scalability scalable infrastructure allowing for easy expansion as the system grows in complexity. Cost-Effectiveness reducing the need for physical hardware and minimizing maintenance costs. pay only for the resources they use Flexibility allowing organizations to choose the right services and configurations that best meet their needs. Benefits of Using Cloud Computing for AI-Enabled IoT Reduced Latency Cloud computing can reduce latency by providing localized computing resources that are closer to the devices. Improved Efficiency offloading some of the processing and storage requirements to the cloud free up resources on the devices and allow them to focus on their primary functions Increased Security providing centralized control over data access, authentication, and encryption Challenges of Cloud Computing for IoT Latency Distance to Data Centers Impact on Performance Security Data Privacy Concerns Reliability Service Downtime Disaster Recovery Bandwidth and Connectivity Data Transfer Limitations Dependency on Internet Connectivity Balancing the Benefits and Challenges Strategies for Success Hybrid Approaches Enhanced Security Measures Optimizing Cloud Usage  Introduction to Cloud Computing Platforms for IoT An overview of AWS IoT and Microsoft Azure IoT AWS IoT Overview: AWS IoT is a managed cloud platform that enables devices to connect, interact, and exchange data securely with cloud applications and other devices. Key Features: Device Management Real-Time Analytics Secure Communication AWS IoT Examples: AWS IoT Core AWS Greengrass AWS IoT Analytics AWS IoT Greengrass Solution Demo for Process Manufacturing Optimization, Advantech (EN) https://youtu.be/rYGQsf7h3Es?si=u5sCKqmCrq8CVdey Microsoft Azure IoT Overview: Microsoft Azure IoT offers a comprehensive suite of IoT services that enable users to build, connect, monitor, and manage IoT applications securely and at scale. Key Features: IoT Hub Edge Computing Advanced Analytics Microsoft Azure IoT Examples: Azure IoT Hub Azure IoT Central Azure Sphere Azure IoT Services | IoT Hub, IoT Central, Azure Sphere https://youtu.be/RHkqFxJWhr8?si=zW8tve7XiOgSVMcJ Azure IoT Operations Demo https://youtu.be/cw9nhPFE_qE?si=34QRiR7AWhP3r2u1  Edge Computing for AI-enabled IoT Distributed Computing for Faster IoT and AI What is Edge Computing? Distributed Computing Paradigm Edge computing is a distributed computing paradigm that brings computation and data storage closer to the devices where it is being gathered, enabling faster processing and reduced latency. What is edge computing? (pizza example) https://youtu.be/WMvb9LLXzUs?si=Umq7Et38_c5zeyhC What is Edge Computing? https://youtu.be/3hScMLH7B4o?si=PpWn5PpPRCY-JjCi  Cloud Computing centralized computing model ideal for applications that require a high degree Edge of scalability and availability Edge Computing Computing vs distributed computing model ideal for applications that require real-time data Cloud processing and low latency Key Differences Computing related to data processing, network latency, and security Benefits of Edge Reduced Latency Edge computing reduces latency by Computing for IoT bringing computing power closer to the data source. Improved Security Edge computing improves security by keeping data closer to the source Decreased Bandwidth Usage Edge computing decreases bandwidth usage by processing data locally AI-enabled Edge Computing  Benefits of AI-enabled Edge Computing improved response times, reduced latency, increased security, and lower bandwidth requirements. more efficient and real-time decision- making for IoT devices Machine Learning enable predictive maintenance and anomaly detection Deep Learning enable more accurate and sophisticated image and speech AI at the Edge recognition Challenges of AI-enabled Edge Computing Limited Computational Resources running AI algorithms on devices with limited computational resources Data Security sensitive data may be processed on the edge devices and transmitted over insecure networks Data Privacy personal data may be processed on the edge devices and transmitted over insecure networks. Use Cases of Edge Computing in IoT Federated Learning for AI enabled IoT Revolutionizing Machine Learning with IoT devices What is Federated Learning? Federated Learning Advantages of Federated Learning increased privacy and security, reduced data transfer and storage requirements, and better performance on edge devices. Federated Learning in AI- Enabled IoT By training machine learning models on edge devices, Federated Learning enables more efficient and effective AI applications in IoT. How does Federated Learning work? Centralized Model Initialization Device Training Model Aggregation Further Training Workflow Summary Step 1: Devices collect and store data locally. Step 2: Local models are trained on-device using this data. Step 3: Trained model updates are sent to a central server. Step 4: The server aggregates updates to form an improved global model. Step 5: The global model is redistributed to devices, where the process repeats. What is Federated Learning? https://youtu.be/X8YYWunttOY?si=KDqSTz_kqrUa0VvZ Making every phone smarter with Federated Learning https://youtu.be/gbRJPa9d-VU?si=QBcjjIWoFjI5H-AS Federated Learning for AI-enabled IoT Federated Learning Architecture distributed machine learning approach that promotes data privacy while also enabling the development of more accurate machine learning models AI-Enabled IoT Devices great potential for building AI-enabled IoT devices that can operate effectively in real-world environments Potential Applications including predictive maintenance, anomaly detection, and real-time decision-making.  Personalized Smart Assistants Use Cases of Remote Health Monitoring Autonomous Vehicles and Federated Transportation Computing in Predictive Maintenance and Quality Control AI-Enabled IoT