Janos Puskas (Accenture) – Azure IoT Reference Architecture for enterprise IoT use cases

Jun 8, 20210 likes95 views

This document provides an overview of an Azure IoT reference architecture for enterprise IoT use cases. It discusses the need for a reference architecture to implement best practices from previous work in a way that can address different scenarios. The reference architecture targets mature organizations using IoT for a collection of applications built around connected device data. It outlines key capabilities the platform must deliver, including sources to receive and retrieve data, processing for data preparation, analysis and compliance, storage for raw and generated data, and consumption through APIs and applications. Diagrams show examples of how these capabilities can be implemented in an Azure-based IoT platform.

INDUSTRY X
Azure IoT Reference
Architecture for Enterprise
IoT use cases
Codiax 2021
Puskás János
2021. 05. 14

Copyright © 2020 Accenture. All rights reserved. 2
INDUSTRY X
The Triple Zero for the future
ONE
FUTURE
ZERO HARM ZERO LOSS ZERO WASTE
“Potential reduction of
22,500 injuries and
498 lives saved
through autonomous
operations, robotics
and remote operating
centers.”
World Economic Forum & Accenture, 2017
Digital Transformation Initiative: Mining &
Metals Industry
“Typical cost of
downtime averages
$180,000 per
incident. Lost profit
based on lost
production time, up
to $130,000 per
hour for every
equipment failure
incident.”
Boltstress Ultrasonics, 2018
Cost of Downtime When Machinery
Breaks Down in the Mining Industry
“Comminution
(grinding)
consumes
50% of mine site
energy and up to
4% of electrical
energy globally.”
Jeswiet & Szekeres, 2016: Science
Direct
Energy Consumption in Mining
Comminution, 23rd CIRP
Conference on Life Cycle
Engineering

More Related Content

What's hot (20)

Using FIWARE and Microsoft Azure for the development of IoT solutionsDunavNET

GETTING STARTED WITH IOT DATA MANAGEMENTBarnaba Accardi

Industrial IOT Projects Research AssistanceMatlab Simulation

Introduction to Microsoft Azure IoTAltoros

IoT Solutions Made Simple with Everyware IoTEurotech

Building the Future with Technology: The Next Five Years Cisco Canada

The truth about IoT field gateways (Sam Vanhoutte @IoT Convention Europe 2017) Codit

IoT Panel- Cisco and Intel Bessie Wang

AIoT and edge computing solutions湯米吳 Tommy Wu

A Reference Architecture for IoT WSO2

Watson IoT - IBMDiego Alberto Tamayo

Windows for Raspberry Pi 2Makers (and more!)Guy Barrette

Eclipse IoT for Industry 4.0Ian Skerrett

Building Apps with Azure IoT EdgeRahul Rai

IoT Introduction Architecture and ApplicationsThe IOT Academy

IoT Platforms and ArchitectureLee House

The State of Edge Computing for IoTIan Skerrett

IoT-Use-Case-eBookNicolas Delorme

Microsoft's view of the Internet of Things (IoT) by Imran ShafqatAllied Consultants

Global Azure boot camp 2015 - Microsoft IoT Solutions with AzureVinoth Rajagopalan

Using FIWARE and Microsoft Azure for the development of IoT solutionsDunavNET

GETTING STARTED WITH IOT DATA MANAGEMENTBarnaba Accardi

Industrial IOT Projects Research AssistanceMatlab Simulation

Introduction to Microsoft Azure IoTAltoros

IoT Solutions Made Simple with Everyware IoTEurotech

Building the Future with Technology: The Next Five Years Cisco Canada

The truth about IoT field gateways (Sam Vanhoutte @IoT Convention Europe 2017) Codit

IoT Panel- Cisco and Intel Bessie Wang

AIoT and edge computing solutions湯米吳 Tommy Wu

A Reference Architecture for IoT WSO2

Watson IoT - IBMDiego Alberto Tamayo

Windows for Raspberry Pi 2Makers (and more!)Guy Barrette

Eclipse IoT for Industry 4.0Ian Skerrett

Building Apps with Azure IoT EdgeRahul Rai

IoT Introduction Architecture and ApplicationsThe IOT Academy

IoT Platforms and ArchitectureLee House

The State of Edge Computing for IoTIan Skerrett

IoT-Use-Case-eBookNicolas Delorme

Microsoft's view of the Internet of Things (IoT) by Imran ShafqatAllied Consultants

Global Azure boot camp 2015 - Microsoft IoT Solutions with AzureVinoth Rajagopalan

Similar to Janos Puskas (Accenture) – Azure IoT Reference Architecture for enterprise IoT use cases (20)

The Platform for the Industrial Internet of Things (IIoT)Gerardo Pardo-Castellote

Introduction to cloud securityIAEME Publication

Industrial IoT and OT/IT ConvergenceMichelle Holley

Mike McBride will provide a look at the Industrial IoT (IIoT) landscape and the OT/IT convergence. He will cover several use cases including healthcare, entertainment and smart buildings. He will cover the challenges IIoT networking faces with emerging technologies and how edge computing will provide increased performance, security and reliability. Mike will discuss the various Edge Computing standards & opensource forums along with proposed architectures. And Mike will present new solutions being proposed (ICN, slicing, Blockchain) to support the bandwidth, latency and security requirements within Industrial verticals. About the speaker: As Sr. Director of Innovation & Strategy, within Huawei's IP Network BU, Mike leads Industrial IoT, Edge Computing and IP/SDN architecture, standardization, and strategy across product lines and industry forums. He leads architecture and standardization activities within the IIc and BBF and has served as an IETF Working Group chair for 15 years. Mike has led emerging technology projects within opensource communities and played a key role in the formation of OPEN-O (Now ONAP). He is an Ericsson alum where he developed and directed SDN/NFV network architectures. And for many years with Cisco, Mike supported customers, worked in development teams and managed mobility, wireless and video projects across BUs. Mike began his career supporting customers at Apple Computer. He resides in Orange County, CA

A Review: The Internet of Things Using Fog ComputingIRJET Journal

Fog computing is a new computing paradigm that processes data and analytics at the edge of the network, rather than sending all data to a centralized cloud. This helps address issues with the cloud-based Internet of Things (IoT) model, such as high latency, bandwidth constraints, location awareness, and mobility. Fog computing brings computing resources closer to IoT devices and end users by using edge devices like routers, switches, and access points as "fog nodes" that can perform analytics and decision making. This allows time-sensitive IoT applications to function more efficiently. Fog computing also helps optimize resource usage by balancing processing between the edge and cloud.

2 pc enterprise summit cronin newfinal aug 18IntelAPAC

FIWARE Tech Summit - Accelerating Materialization of the IDS ArchitectureFIWARE

Internet of Industrial Things Presentation - Sophie Peachey - IoT Midlands Me...WMG, University of Warwick

10 IoT technology trends to watch in 2022 - IoT Analytics.pdfAtaulAzizIkram

Supermicro AI Pod that’s Super Simple, Super Scalable, and Super AffordableRebekah Rodriguez

The worlds of HPC and AI are evolving at a tremendous rate. The demands of modern-day applications put immense pressure on local IT teams and resources. More often than not, this pressure can come from requiring an AI strategy to speed up mission-critical applications - but this can come at a cost which can hinder adoption. In this webinar, Supermicro, together with International Computer Concepts (ICC) and Define Tech, will demonstrate their AI Super Pod that delivers on AI strategy needs without breaking the bank.

1. How will the IoT help your business - ciscoMITEF México

Open Source for Industry 4.0 – Open IoT Summit NA 2018Benjamin Cabé

Industry 4.0 is set to revolutionize the manufacturing industry. The potential for more flexible manufacturing, more efficient processes and lower costs are the driving factors behind the investment in Industry 4.0 solutions. A key part of creating successful Industry 4.0 solutions will be software on the factory floor and in the cloud. In this talk, we will introduce how open source software has become a trusted source of technology for the enterprise IT software industry and how the Eclipse IoT open source community and other open source communities are now ready to provide production ready technology for the manufacturing industry and Industry 4.0. Open source software will provide the key building blocks that will promote the interoperability and flexibility required by Industry 4.0 solutions.

The Cisco Connected FactoryMainstay

IoT challenges for Smart ManufacturingHeiko Koziolek

1. The document discusses several challenges for connecting industrial IoT devices like a laser level transmitter to the cloud, including dealing with resource constraints, designing appropriate information models, standardization, security, and integrating autonomy. 2. It provides examples of industrial IoT applications and discusses using OPC UA as an information modeling and connectivity technology for the IIoT. 3. The presentation addresses how software architecture research can help with these challenges through reference architectures, patterns, modeling languages, performance modeling, and security methods.

Ibm iot overviewEric Cattoir

The document provides an overview of the Internet of Things (IoT) and IBM's solutions in this area. It discusses the scalability issues with today's IoT solutions and the evolution needs to address billions of connected devices. The document outlines IBM's IoT viewpoint which covers the full lifecycle from connecting devices to analyzing data. It also summarizes some of IBM's IoT solutions including analytics offerings, IoT foundations, gateways, and DevOps solutions. The document stresses that both business models and technology need evolution to fully realize the potential of the IoT.

Internet of things by Mr.Pradeep_Kumarpradeep kumar

The document discusses recent trends and research areas related to the Internet of Things (IoT). It begins with an introduction to IoT, describing it as enabling connections between devices anytime, anywhere through any media. It then covers IoT architecture and applications, challenges including interoperability and security, and potential research directions such as developing energy efficient and secure algorithms for edge devices. The document also discusses IoT user growth ratios, platforms for IoT development, and potential future innovations. It concludes that IoT will play a major role in monitoring the environment and that further research is still needed in areas like applications, edge/fog/cloud systems, and addressing security and privacy concerns.

Arc view convergence of ai and io t reportMohammad Shamsuzzoha (Shams)

Artificial Intelligence of things (AIoT) is built for industrial companies looking for better ways to connect their evolving workforce to data-driven decision tools and digitally augment work and business processes. Aspen Technology announced its Aspen AIoT Hub, which provides integrated data management, edge, and cloud infrastructure to build, deploy and host Industrial AI applications at enterprise speed and scale. The convergence of IT and OT groups driven largely by digital transformation has created organizational confusion and a significant "gray space" of common technologies. Leveraging AI requires combining data science with domain expertise through a hybrid approach known as Industrial AI.

2019 06-19 EIT Digital industry event MIDIH_EU

The MIDIH project aims to provide opportunities for SMEs in manufacturing through digital innovation hubs and reference architectures. It has a budget of over 8 million Euros from the EU and involves partners across Europe. The project focuses on developing edge computing, interoperability, data solutions, and experiments in areas like smart factories, products, and supply chains. It recently opened a call for proposals worth 960,000 Euros for SMEs to conduct technological or experimental projects using the MIDIH reference architecture.

IRJET- A Smart Medical Monitoring Systems using Cloud Computing and Internet ...IRJET Journal

1. The document proposes a smart medical monitoring system using cloud computing and the Internet of Things. It presents an architecture called RMCPHI that uses body sensors, networks, communication modules, and cloud services to remotely monitor patient health data. 2. The RMCPHI architecture transfers sensor data through gateways to a medical information analysis platform where data is processed and statistics are generated. This allows quick decision making for remote health monitoring and management. 3. The system aims to improve remote patient monitoring by leveraging the flexible resources of cloud computing to handle large volumes of medical data generated by IoT sensors.

Lange - Industrial Data Space – Digital Sovereignty over DataVienna Data Science Group

The document provides an overview of the Industrial Data Space (IDS) project. The IDS aims to enable secure data exchange between companies while maintaining data sovereignty. It discusses the motivation around digitization of industry and outlines the IDS technical architecture and components. The IDS research project is developing reference use cases and a consortium of companies are involved to help define requirements and standards for an open ecosystem.

FIWARE Global Summit - Implementing Data Sovereignty: a FIWARE-MINDSPHERE Int...FIWARE

The Platform for the Industrial Internet of Things (IIoT)Gerardo Pardo-Castellote

Introduction to cloud securityIAEME Publication

Industrial IoT and OT/IT ConvergenceMichelle Holley

A Review: The Internet of Things Using Fog ComputingIRJET Journal

2 pc enterprise summit cronin newfinal aug 18IntelAPAC

FIWARE Tech Summit - Accelerating Materialization of the IDS ArchitectureFIWARE

Internet of Industrial Things Presentation - Sophie Peachey - IoT Midlands Me...WMG, University of Warwick

10 IoT technology trends to watch in 2022 - IoT Analytics.pdfAtaulAzizIkram

Supermicro AI Pod that’s Super Simple, Super Scalable, and Super AffordableRebekah Rodriguez

1. How will the IoT help your business - ciscoMITEF México

Open Source for Industry 4.0 – Open IoT Summit NA 2018Benjamin Cabé

The Cisco Connected FactoryMainstay

IoT challenges for Smart ManufacturingHeiko Koziolek

Ibm iot overviewEric Cattoir

Internet of things by Mr.Pradeep_Kumarpradeep kumar

Arc view convergence of ai and io t reportMohammad Shamsuzzoha (Shams)

2019 06-19 EIT Digital industry event MIDIH_EU

IRJET- A Smart Medical Monitoring Systems using Cloud Computing and Internet ...IRJET Journal

Lange - Industrial Data Space – Digital Sovereignty over DataVienna Data Science Group

FIWARE Global Summit - Implementing Data Sovereignty: a FIWARE-MINDSPHERE Int...FIWARE

More from Codiax (20)

Dr. Laura Kerber (NASA’s Jet Propulsion Laboratory) – Exploring Caves on the ...Codiax

The document discusses a proposed mission called Moon Diver that would send a robot to rappel down into a 100m wide and deep pit on the moon called the Tranquillitatis Pit to examine lava layers in the pit walls. Some key challenges of such a mission would be the extreme temperature swings between day and night, limited time on the surface due to the moon's 14 day day-night cycle, and hazardous lunar dust. While an exciting prospect, Moon Diver remains a conceptual mission that has not yet secured funding to become a reality.

Costas Voliotis (CodeWeTrust) – An AI-driven approach to source code evaluationCodiax

This document discusses an AI-driven approach to source code quality evaluation called source code inspection. It aims to reduce technical debt and the costs of maintenance, integration, and license compliance by continuously evaluating source code quality as part of the software development lifecycle. The proposed approach uses machine learning and analytics to classify issues, detect vulnerabilities, analyze dependencies, and provide recommendations to optimize the development process.

Dr. Lobna Karoui (Fortune 500) – Disruption, empathy & Trust for sustainable ...Codiax

This document outlines the agenda for a presentation on business growth in times of exponential technologies. The presentation will discuss the state of AI in 2020, how AI adoption can increase company revenues, and the future of employment and skills. It will also introduce the AI Exponential Thinker organization, which aims to leverage emerging technologies for a sustainable future for all. The presentation will explore uncertainties in predicting the future, exponential technology trends, and how organizations are using AI to generate value. It will also address skills needed for 2025 like problem solving, critical thinking, and creativity.

Luka Postružin (Superbet) – ‘From zero to hero’ in early life customer segmen...Codiax

This document outlines the steps to implement early life customer segmentation at Superbet, a betting company. The key steps include: 1. Collecting and storing customer behavioral data from various sources in a centralized data lake. 2. Cleaning and organizing the data for analysis using tools like Airflow and Vertica. 3. Analyzing the data to find patterns and create hypotheses about customer segments. 4. Building a predictive model to segment customers into "best", "worst", and VIP categories based on their betting behavior and risk to the business. 5. Optimizing the model and pushing it into production to serve recommendations in near real-time and improve customer segmentation.

Gema Parreno Piqueras (Apium Hub) – Videogames and Interactive Narrative Cont...Codiax

This document discusses interactive narrative control and the assurance, specification, and robustness of language agents. It mentions Apium Hub, which designs data-driven solutions using software. It discusses using reinforcement learning, imitation learning, and PPLM with GPT to define an attribute model using a bag of words approach. It also discusses fine-tuning GPT-2 models, running experiments with different prefixes, and searching parameters to create companionship for language games.

Adria Recasens, DeepMind – Multi-modal self-supervised learning from videosCodiax

The document summarizes a talk on multi-modal self-supervised learning from videos. It discusses using multiple modalities like vision, audio and language from videos for self-supervised learning. It presents two models: 1) A Multi-Modal Versatile network that can take any modality as input and respects the specificity of each while enabling comparison. 2) BraVe which learns representations by regressing a broad representation of the whole video from a narrow view to leverage different augmentations and modalities. Both models achieve state-of-the-art results on downstream tasks, showing videos provide rich self-supervision and using additional context improves representation learning.

Roelof Pieters (Overstory) – Tackling Forest Fires and Deforestation with Sat...Codiax

This document provides an overview of Overstory, a company that uses satellite data and AI to monitor forests and tackle issues like deforestation and wildfires. It discusses how Overstory uses machine learning on high-resolution satellite imagery to create segmentation maps and monitor changes in forests over time. It also describes Overstory's infrastructure including its use of JupyterHub, Dask, and Papermill to enable large-scale distributed processing of satellite data and training of deep learning models.

Javier Fuentes Alonso (Uizard) – Using machine learning to turn you into a de...Codiax

The document discusses using machine learning to assist with and accelerate the design process. It outlines the typical workflow for creating an app or website, from ideation through delivery. The author proposes leveraging computer vision, natural language processing, and other ML techniques to turn sketches into functional prototypes, define what constitutes "good looking" design, and generate design recommendations. The goal is to democratize design and lower the learning curve, so ML tools can be used by domain experts, founders, and others beyond just designers and engineers. Challenges discussed include accurately interpreting visual representations and semantics as well as defining objective metrics for aesthetics.

Emeli Dral (Evidently AI) – Analyze it: production monitoring for machine lea...Codiax

Machine learning models in production are susceptible to failures from data and concept drift issues that can degrade model performance over time. Effective monitoring of model health and data quality is needed to detect such issues. Key aspects to monitor include model accuracy, data schema changes, data distribution shifts, broken data pipelines, and the impact of concept drift. Starting simply by adding basic machine learning metrics to existing service monitoring for memory, latency, and uptime is pragmatic. The appropriate monitoring approach depends on factors like available resources, use case importance, and system complexity.

Matthias Feys (ML6) – Bias in ML: A Technical IntroCodiax

This document discusses bias in machine learning. It defines different types of bias, including sample bias, prejudicial bias, exclusion bias, and measurement bias. It also discusses various definitions of fairness, including unawareness, group fairness, and individual fairness. The document outlines group fairness metrics like demographic parity, equal opportunity, and equalized odds. It describes causes of bias and algorithms to mitigate bias, such as reducing bias in data, classifiers, and predictions. Overall, the document provides an overview of bias in ML, definitions of fairness, sources of bias, and methods for debiasing models.

Christophe Tallec, Hello Tomorrow – Solving our next decade challenges throug...Codiax

This document outlines major challenges facing the world over the next decade related to the environment, health, and food and calls for the development of scalable, affordable solutions through deep technology innovation. It notes that the climate crisis will intensify risks to food/water/health security and that cumulative emissions already in the atmosphere will drive temperatures increases the next two decades. Various startups are highlighted that are working on technologies to tackle root causes like batteries using CO2, renewable hydrogen, atmospheric monitoring satellites, parametric insurance, air purification, AI disease detection, cell-based meats, and indoor farming with LED lighting to transform industries sustainably.

Sean Holden (University of Cambridge) - Proving Theorems_ Still A Major Test ...Codiax

This document discusses applying machine learning techniques to automated theorem proving and formal proof checking. It begins by providing background on logic-based theorem proving and efforts to formally prove mathematical theorems. It then discusses using machine learning to help guide automated theorem provers by selecting optimal heuristics and recommending useful lemmas. The document concludes by noting the challenges of developing mathematical languages that are both natural for humans and amenable to formal verification.

Olga Afanasjeva (GoodAI) - Towards general artificial intelligence for common...Codiax

This document discusses GoodAI's mission to develop general artificial intelligence as quickly as possible to help humanity and understand the universe. GoodAI sees AI as a tool to augment human abilities, and discusses both narrow AI applications like using AI to help prevent pressure ulcers in hospitals through monitoring patient positioning, as well as long term goals for developing general AI through gradual learning techniques like curriculum learning and meta-learning to help AI systems understand human values.

Maciej Marek (Philip Morris International) - The Tools of The TradeCodiax

This document discusses best practices for developing and deploying data science products at PMI. It recommends using Docker containers to ensure reproducibility, project templates for organization, continuous integration/delivery (CI/CD) pipelines for testing and deployment, and monitoring tools like Kibana and Airflow. Adopting these practices can help data scientists, data engineers, and devops teams work more collaboratively and rapidly deliver machine learning models to production.

Joanna Bryson (University of Bath) - Intelligence by Design_ Systems engineer...Codiax

1. The document discusses issues around accountability and transparency in artificial intelligence. It notes that intelligence is a form of computation and AI extends and reuses human computation through techniques like machine learning. 2. Ensuring accountability and transparency in AI is important. Accountability refers to responsibility being assigned, while transparency allows demonstrating due diligence. Regulations aim to motivate transparency and proof of due diligence in AI systems. 3. Responsible AI design is discussed, such as behavior-oriented and modular design techniques that can improve transparency. Monitoring of AI systems and their development processes can also support accountability. However, attributing responsibility ultimately lies with human organizations and individuals.

Jakub Langr (University of Oxford) - Overview of Generative Adversarial Netwo...Codiax

This document provides an overview of Generative Adversarial Networks (GANs) in 3 sections. It begins by briefly discussing supervised and unsupervised machine learning. It then explains that GANs use two neural networks, a generator and discriminator, that compete against each other in a game theoretic setup. The generator learns to produce more realistic samples while the discriminator learns to better distinguish real and fake samples. Popular GAN architectures like CycleGAN and BigGAN are also summarized.

Jakub Bartoszek (Samsung Electronics) - Hardware Security in Connected WorldCodiax

The document discusses hardware security in the connected world. It notes that with 25+ billion endpoints expected by 2020 and the rise of 5G, IoT, and AI, security is more important than ever as devices have increased attack surfaces. It advocates for a comprehensive security strategy including validated hardware and firmware, secure networks, and security management infrastructure. The importance of hardware-based security is discussed, noting that truly secure computing requires addressing security at the hardware, firmware/OS, and application layers. Samsung's Knox platform is highlighted as an example of building and maintaining trust from manufacturing through runtime with features like hardware root keys, verified boot, encryption, and real-time kernel protection.

Jair Ribeiro - Defining a Successful Artificial Intelligence Strategy for you...Codiax

To develop a successful AI strategy, it must be part of an overarching business plan and tied to business priorities and strategic goals. An AI strategy also requires a data strategy to ensure sufficient and relevant data is available for training models. Additionally, ethical and legal issues around privacy, bias, and consent must be considered. Developing AI skills and managing cultural changes are also important to successfully implement an AI strategy.

Cindy Spelt (Zoom In Zoom Out) - How to beat the face recognition challenges?Codiax

This document discusses the challenges of face recognition technology. It addresses deep learning challenges, business challenges, and ethical management challenges. It discusses techniques like Angular Margin Loss to enhance discriminative power in face recognition algorithms. It also addresses issues like algorithmic bias, data bias, and lawsuits related to large face datasets. The document emphasizes the need to build face recognition products in an ethical manner while also dealing with unrealistic requests and attempts to fool the technology.

Alexey Borisenko (Cisco) - Creating IoT solution using LoRaWAN Network ServerCodiax