Watching A Replay
To access the replay sessions(s) click “View Archive” next to the desired session below. For your convenience and ease of locating sessions of interest to you, we recommend using the event Filter by Date or Track which will allow you to sort sessions by conference/event track.
China is World’s 2nd largest economy, growing at high single digit over 4 decades. China is fast growing IoT market. A lot of IoT innovation, new usages are appearing in Smart Cities, Smart Manufacture, Smart Home, etc. In this session, speaker will introduce Intel IoT strategy in China, Intel’s portfolio for IoT and Intel’s new offering to developer to promote IoT innovation with AI technology.
This session introduces Intel® VTune™ Amplifier, a powerful performance profiler that’s part of Intel® System Studio.
Build your profiling expertise with these how-to demonstrations:
• Configuring to profile different target platforms and devices
• Key features for a range of analysis types
• How to interpret and understand collected data
• Performance tuning methodologies
• Real workload analysis case studies
• Using Intel® SocWatch for energy profiling and power optimizations
• GPU, memory, threading, and storage analysis
Traditional hypervisors are all for server, where it works with request-and-response model in big iron with focus on sharing CPU cores, network and storage devices. Unlike a server, the IoT system is quite different: it requires small, simple, high-performance and certifiable solutions. As computing power goes up, IoT systems start to consolidate multiple different subsystems together with higher performance and lower cost.
In this session, we present the ACRN* hypervisor, which is designed for dedicated IoT systems. It takes the advantage of IoT usages, and maximizes the benefits of multi-core and modern virtualization technologies. It minimizes the footprint and achieves high performance for both partitioning and sharing usage in IoT systems, making it widely applicable to different IoT market segments from small devices to large edge computing devices.
The IoT is evolving rapidly with cloud technology and smart hardware. It’s widely used in CAD: for building maintenance, construction management, intelligent manufacturing, smart city, etc. This session showcases one of the hottest businesses today: IoT workflow with a viewable real 3D model. Using the new UP Squared* Grove IoT Dev Kit UP2 and Autodesk Forge* cloud service, developers can get started easier than ever before using with an IoT + 3D model solution.
These two technologies broke the barriers to data connecting, increasing speed and accuracy across the enterprise through connected systems while reducing complexity due to isolated data islands. In this session we explore building applications using UP Squared* Grove IoT Dev Kit and Autodesk Forge services.
Voice, vision and text are the most natural means of human communication by which we interact. These communication mechanisms can be extended to machines to achieve a natural digital experience (NDE). Touch revolutionized the way humans interact with devices such as mobile phones. Now, with the advent of digital assistants like Amazon Alexa* and ambient intelligence, voice is already on its way to heralding the next revolution.
This session provides an overview of NDE, including:
• Real-world use cases
• An architecture deep dive for NDE proof-of-concept (POC) involving voice, intelligent vision, and gestures, as well as a POC demo
• A vision for how machines can collaborate with humans, and each other, to achieve an autonomous ecosystem
This paper introduces the development and industrial implementation of artificial intelligence in the security IOT industry, including products and solutions. How to take advantage of the Movidius chip on the production side in the smart front-end camera and how to provide more space for the extension of artificial intelligence.
The industrial world is undergoing a transformation known as Industry 4.0, or the Industrial Internet of Things. This transformation is leading to smart manufacturing systems with more compute capability at the edge. However, adding more compute capability increases OPEX, due to complexity of computer integration, as well as cabling, power and maintenance costs. The solution: “workload consolidation.”In this session, you’ll learn how advances in processing performance, real-time operating systems, and virtualization technology are enabling workload consolidation in the factory to drive global industries from “automatic to autonomous” for Industry 4.0.
Today, network and system security are of paramount importance in the digital communication environment. On par with the developments in technology, many threats to information security have emerged that can impact sensitive transactions: intruders can easily cause all kinds of breaches and crash networks, launch denial of service attacks, inject malware, and so on. To avoid such breaches, security administrators badly need a means to detect intruders and prevent them from entering the network. Nowadays, new threats and associated solutions are emerging together.
This session showcases a hybrid intrusion detection system that leverages the benefits of machine learning techniques to build a system that detects intrusion and alerts network administrators. This system can be extended from intrusion to breach detection as well.
The developed system analyzes and predicts user behavior, which in turn classifies as an anomaly or a normal behavior. Other systems use just one machine learning algorithm to solve the problem, while this hybrid intrusion detection system uses a combination of algorithms for classification, achieving greater accuracy.
This session shares different ways of performing machine vision analyses efficiently on the Intel® platforms, including how to deal with follow-up motion control.
Electricity is a major problem in Africa, where most people are not connected to their regional grid. This session presents a real-world use case for solar energy and a LoRa* network to provide and monitor electricity generated across a very long distance. Each system is equipped with a long range, low power LoRa* network to serve remote locations. Users can be monitored accordingly and given real-time information about their energy usage.
We’ve moved beyond the initial IoT hype. Technology and connectivity are here and sensors are collecting robust data. The foundation stage is complete, and now we can turn our attention to more exciting experimentation and development.
This session is for IoT developers with a desire to embed cellular communications into solutions. Prepare for a journey from proof-of-concept and tinkering with small-scale tests to larger scale deployments and lifecycle management.
Gemalto, a longtime Intel® IoT Solutions Alliance member, offers a range of simple and reliable, Intel® technology-based development kits that ship with embedded Java* allowing for easy and fast application development. Receive an expert introduction to Gemalto's Cinterion* Connect Shield, to help accelerate deployment through rapid experimentation and building proof-of-concept solutions. Learn how you can bring your IoT ideas to life, fast. The easy-to-use maker board features on-board sensors and combines the flexible and widely used Arduino open source platform with Gemalto’s reliable and highly efficient cellular 2G, 3G and 4G LTE connectivity.
In a pair of live demos, we’ll explore the Arduino* library through an MQTT example and demonstrate the IoT module’s embedded Java processing capabilities by reading sensor values and lifecycle management of hardware and devices.
For most people, the cloud represents the backbone of IoT. However, many are starting to understand that the cloud is only one element in the IoT universe. It makes good sense for some portion of computing to be done locally, in real time, and not be pushed to the cloud, whether for security, latency or cost concerns.
This session focuses on the sound reasons for deploying AI at the Edge and its competitive advantages for safety, security and data protection. You learn how edge computing is not a replacement for cloud, but rather a complementary strategy for key business areas where using a cloud-only approach simply won’t cut it.
The training also provides real-world use cases from our experience in developing real-time AI for edge devices.
Why shouldn’t building and managing a fleet of devices be as easy as building web services? This is the premise under which developers now start their IoT projects. Rather than drawn-out proof of concepts (PoCs) with undefined goals, building an IoT device should be a streamlined process from concept to production to distribution, with the capability for continuous software development in the field.
In this session, learn what makes this possible: a combination of certified hardware, open-source software and packaged applications to deploy devices with a vastly reduced time to market.
Artificial intelligence (AI), the next big wave in computing, is an increasingly important source for competitive advantage that is already transforming industries. Client AI is growing in importance as it becomes a big trend in terms of competitive strategies and requirements from ecosystem partners.
Attend this session to understand why now is the ideal time to start integrating AI into your products, services and business processes. Intel has a complete AI portfolio, world-class silicon, and decades of experience successfully driving major computing transformations.
Advancements in technology have steadily increased the processing capabilities of the edge devices. Earlier analytics were primarily running on powerful servers. Now, thanks to exponential increase in camera and server processing power, analytics can be used by all kinds of users and in a variety of environments. Analytics can run on the camera or on an edge or fog server running multiple video streams or multiple applications. Consequently, opportunities in these areas are expanding as investments steadily increase in Safe City projects, with a strong interest in IoT and automation from enterprises as well.
This deep-dive session guides you through enabling intelligent video analytics using OpenVINO toolkit on Intel® IoT Gateway edge/gateway platforms. These include Intel Atom® processor Apollo Lake, Intel® Movidius™ NCS, as well as Intel® Core™ processor and Intel® Xeon® processor-based platforms. We cover key optimization techniques using Media and OpenVINO™ toolkit, as well as performance profiling tools to improve TCO.
Artificial Intelligence, AI, is changing our lives from the past to the future. It enables machine intelligent by using a variety of training models to simulate and infer the status or appearance of an object. For example, the inference system with video analysis model can perform face and car plate analysis for safety and convenience purposes. IEI introducing TANK AIoT Dev. Kit features rich I/O and dual PCIe by 16 slots with by 8 signal for add-on card installation such as PoE (IPCEI-4POE) card & IEI AI acceleration card (Mustang-200, F100 & V100 series) to enhance function and performance for various AI applications.
This session and demo will cover the challenges to professional, secure, scalable IIoT operations. Many IIoT solutions focus on provisioning of IoT GWs and attached IoT devices; they utilize common IoT protocols and methods (MQTT*, LoRa*, etc.), and address the data exchange between users, IoT data brokers, IoT gateways and IoT devices in the field. But, what industrial clients demand but is frequently missed by “IoT-only-focused designs” are:
• Service Assurance and deployment of the IoT infrastructure, provider and network backbones, access to IoT devices (infrastructure availability), utilization, and performance.
• Quick and automated information on IoT services and infrastructure status for timely reaction, service restore, security, and vulnerability tracking.
• Automated and timely “inventarization” and change tracking of all involved solution parts and devices.
Our demonstration, from the Infosim StableNet* demo lab, lays out best practices, and how these challenges are resolved by a professional management solution. Learn how to:
• Use an automated management solution to speed up your build and project deployments
• Initiate infrastructure and IoT gateway and sensor discovery
• Automate service assurance for IIoT deployments
• Use OTA SW updates
• Build a management cockpit for multi-client operation and troubleshooting
• Consider technical and commercial handling within the lifecycle of large, multi-client Industrial IoT deployments
Computer Vision (CV) is this year's breakthrough technology, and Go is this year's hottest programming language. When you put them together, you can create CV applications that provide high performance with minimum code. Adding Deep Neural Network (DNN) capabilities to the mix allows you to create whole new kinds of software applications.
In this talk, I will take you from the basics of CV, all the way to flying a Intel Movidius powered drone. Get ready for a wild ride!
All aboard for a session about a real-time railway condition monitoring system that allows the implementation of predictive maintenance activities. Learn about its solutions, technical properties, and the results obtained from its application examples.
Instead of a non-real time maintenance strategy that only responds when (or after) a problem occurs, this railway condition monitoring system, called DS-xR, can determine real-time situations where operating conditions are abnormal based on vibration, temperature and speed data, as well as rail line track, traction motor and its bearing, wheel, gear box and axle box.
As a result, the system helps normalize operating conditions before problems occur or grow. Using this system can extend equipment lifetime, increase operation safety, optimize spare parts use, lower maintenance costs, and decrease maintenance time.
MindSphere* is the open, cloud-based IoT operating system from Siemens designed for the industrial Internet of Things. It offers an intuitive, straightforward environment that lets users build, scale and deploy new ideas quickly.
In this session, Siemens MindSphere technical experts share what MindSphere developer offerings are all about, as they discuss key features and functionality, and answer technical questions. The session is designed to help application developers, solution architects and other professionals make use of powerful technology resources.
Flexibility and interoperability in the field of IoT software is a paramount objective. The mix of platforms, “things,” edge analytics, etc. that are connected in an IoT deployment is extensive and ever growing. How do you create a software platform that connects to a variety of greenfield and brownfield sensors, incorporates various edge analytics, and communicates to a diverse set of cloud and enterprise platforms? The solution, as exhibited in EdgeX Foundry*, an open source, vendor neutral IoT platform, is to use a microservice architecture.
In this session, you’ll learn about EdgeX Foundry and a microservices architecture can help:
• Address the challenges of dealing with the IoT protocol soup (Modbus*, BACnet*, BLE, Zigbee*, MQTT, OPC-UA*, etc.)
• Help organizations to incorporate use case specific edge analytics or event processing
• Allow for continual improvements and upgrades of various parts of the IoT solution without requiring a redo of the entire platform
• Allow for value additions into your IoT solution in order to maximize return on investment
• Maximize the utilization of available resources which tend to be more constrained at the edge
No software architecture is a silver bullet. The presentation will also explore other challenges (and lessons learned) that a microservices IoT architecture introduces. Those challenges to be addressed include dealing with orchestration/deployment, security and additional communication latency.
Smart Fridge and Smart Shelf real-time monitoring for efficient inventory management can be used in different verticals such as convenience and grocery stores, laboratories, and healthcare facilities. This end-to-end IoT solution uses Bluetooth*, WiFi*, sensors, gateways and a cloud infrastructure to track, monitor, collect data, and alert end-users of inventory.
This session demonstrates both solutions and explains the technologies, tools and protocols used to create each:
• Smart Fridge uses temperature, humidity, weight, and other sensors to track refrigerator conditions, power usage, whether the doors are open or closed, and identify the number and types of products taken from the refrigerator.
• Smart Shelf uses load cells, proximity, temperature, humidity sensors and an e-paper display. This application can identify the weight on a shelf, temperature and humidity, and how many people have come within close proximity of the shelf. With the e-paper display, the color scheme, product, cost and other display attributes can be changed locally or remotely.
IoT gateways are an important component in the IoT infrastucture. With the emergence of intelligence at the edge, how do we ensure productivity, i.e., collecting actionable sensor data, connecting more devices, and processing large data streams from a variety of sources, from cameras to sensor nodes, and ensuring security of these data.
This session demonstrates how to use the power of Intel® processors, as well as technologies such as Intel® Movidius™ Compute Module, and Intel Atom® processors. We’ll share some firmware and Os tricks to promote safety and increased productivity of IoT Gateways. And, the session offers a detailed guide for quickly prototyping a gateway to add Bluetooth*, NFR, GSM and LTE, and build an application sending data to the cloud.
The use of IoT and Artificial intelligence in plants, warehouses and machines is drastically changing the way businesses make informed decisions that are critical to performance.
This session demonstrates how the IoT, Artificial Intelligence, ERP cloud, maintenance, Finance, WMS, robots and drones are all integrated at the business use case level. These technologies are enabling systemic, predictive and proactive decisions, for new best practices to improve performance, human safety and quality. IoT and AI are not standalone practices and have to be considered in a holistic way with business applications to create value.
Many U.S. communities recognize that Smart Cities only exist when Smart Applications are built on a Smart Network that is open to innovation. Cities today are filled with technology silos. The true power of the Smart City comes when the city has control to program the network at will. This is only possible when the city can control the network and connect to any node on the network.
This session introduces IoTbrix* from American Megatrends Inc., a simple, fast, coding-free, cloud-based firmware development environment for IoT devices. Designed for IoT developers, makers and enthusiasts, IoTbrix lets users:
• Create custom IoT firmware to power-on and bring-up IoT devices
• Work with intuitive drag-and-drop interface – no coding required
• Operate on a cloud-based platform – no need to license, install or set up an IDE or compiler
• Focus on project features and value-adds, instead of code
• Lower the barrier of entry to IoT development and speed TTM
Starting a new project with IoTbrix becomes as simple as dragging and dropping chips, sensors, actuators and other components into the web-based design workspace. It features support for over 200 sensors and commonly used software modules on a range of silicon platforms. Connect the leads by clicking on the pinouts and IoTbrix does the rest, generating the firmware image and the underlying code.
IoTbrix represents a bold step forward in IoT platform design by lowering the barrier of entry to IoT firmware development. It enables developers, makers and enthusiasts with little programming experience to quickly create prototypes and bring them to production.
Many of today´s applications across a variety of industry sectors require compatibility with more than one operating system. Often, at least one workload requires deterministic real-time behavior at minimum jitter and lowest possible interrupt latencies. Consolidating such high-performance, real-time workloads running, for example, in Microsoft Windows* or another general-purpose operating system (GPOS) in addition to an IoT gateway – all on a single multicore platform – can be challenging.
Guest operating systems executing various workloads in parallel must not influence each other. And, virtualization overhead for real-time operating systems must be kept to an absolute minimum.
This session details the impacts on real-time behavior, jitter and latencies due to simultaneous parallel access of shared resources. We’ll also cover the use of cache or advanced power management features. And, we’ll share best practices, including the use of Intel Cache Allocation Ttechnology (CAT) and other remedies to optimize real-time system performance.
Join Stack Overflow CEO Joel Spolsky talk with Aaron Tersteeg of Intel about how he built & grew Stack Overflow (and what he thinks of IOT & the software industry today
The Internet of Things is the next wave in the digital transformation. As enterprises get ready to fully adopt IoT in their businesses and systems, it is key to plan your infrastructure strategy wisely. According to Gartner, through 2018, 75 percent of IoT projects will take up to twice as long as planned due to challenges with integration, security and people.
This session underscores your need for an edge-to-cloud architecture that’s secure, agile and, most important, consistent. Your goal is to reduce wasted integration costs and to maximize your ROI in existing/ legacy investments. Attend and learn why VMware Pulse IoT Solutions is the right choice.
Securing your IoT devices is challenging. High-performance FPGAs such as the Cyclone* V and Arria* 10 can run computationally intensive security algorithms such as RSA and ECC, but the low-resource endpoints they connect to cannot. In addition to being too slow, these methods often do not fit in available RAM and ROM of constrained devices. As a result, many low-resource devices -- and therefore many systems -- are deployed with little or no protection. In addition, applications for automotive, industrial, medical, accounts payable, and others with multi-year lifecycles must address the coming threat from quantum computing that will break RSA and ECC.
In this session, experts from SecureRF and Intel will introduce security methods that are fast, flexible, and future-proof, running on low-resource devices. Topics will include:
• Current and future threats to connected low-resource devices
• How to use FPGAs for secure, flexible and upgradeable designs
• Strategies for delivering fast authentication and data protection for applications with resource constrained devices
• A real-world case study: The DE10-Nano development board and a low-resource device using SecureRF methods, including how-to tips for fast deployment.
IoT is at the center of exciting new capabilities in a variety of settings, including manufacturing, retail, health, transportation, and the home. It is also at the forefront of public, industry, and regulatory concerns about data, algorithms and automation. News reports of algorithmic bias, privacy, transparency, and the future of work abound, and a new crop of think tanks, research institutes, and standards development efforts have emerged in response.
Ethics issues are easy to spot in hindsight but much harder to catch and fix early on, before harm is done. What does it take to meaningfully examine social and ethical issues, given the realities of constant build/test/learn cycles? What questions need asking, and when? What creative solutions have others found to resolve these issues?
This session will help you move beyond the media hype cycles. We will clarify the areas where credible researchers believe there is cause for concern, and show you some approaches and frameworks to help you know the right questions to ask at the right time. We’ll also dive into real-world examples where things did go wrong, and pose promising interventions that could help.
The visual fog paradigm envisions tens of thousands of heterogeneous, camera-enabled edge devices distributed across the Internet, providing live sensing for a myriad of different visual processing applications. The scale, computational demands, and bandwidth needed for visual computing pipelines necessitates offloading intelligently to distributed computing infrastructure, including the cloud, Internet gateway devices, and the edge devices themselves.
This session deep-dives into two aspects of visual fog orchestration: offloading and scheduling. Offloading is a mechanism for realizing (live) workload migration, whereas scheduling attempts to resolve the problem of assigning the visual computing tasks to various devices to optimize network utilization. You’ll be introduced to our pioneered study, in which we demonstrate sub-minute computation time to optimally schedule 20,000 tasks across over 7,000 devices, and just 7-minute execution time to place 60,000 tasks across 20,000 devices. Witness an approach that’s ready to meet the scale challenges: visual fog is feasible and a viable paradigm to scale out video analytics systems.
An increasing amount of data is being generated at the edge, and we desire the Things generating it to connect, exchange, and transact autonomously on our behalves. But we also require privacy, provenance, and careful data stewardship.
This session examines how IoT demands are driving a blockchain evolution to support enterprise requirements. We’ll show how the results have already begun to transform supply chains from predictable linear processes to dynamic, distributed, near real-time organisms. Join us for a conversation about evolving enterprise blockchain features and the real-world supply chain usages they enable.
Want to load your OS securely off the device of your choice as quickly as possible, with a minimum of fuss? In this session, developers learn how to simplify their life with Insyde Software’s BlinkBoot* UEFI Bboot Lloader.
See how the IoT-focused architecture, streamlined UEFI secure boot flow and easy customization takes you from a readily available reference board to your family of Intel® IoT platforms.
There are already hundreds of thousands of cameras in many smart cities, but how intelligent are their video cameras and video management systems? This session digs into the use of artificial intelligence to control cameras, and explains how sensor data can be used to analyze video stored at the edge.
Only using current Open Source solutions, we show best practices for implementing Smart Cities, across Things' cameras and sensors to Gateway to Cloud, emphasizing open build scripts, security and DevOps lifecycle, a Reference Architecture with compliance for US, India, China and European Smart Cities. Solutions include (in demo build order):
1. Platform infrastructure with RASS (Reliable, Available, Safe and Secure): Kubernetes orchestration of trusted pods of Docker containers, with Kafka publish-subscribe containerized microservices (city data & AI, cloud lambda, 5G directory services). Cloud: Amazon S3 frontend, using Pithos compatible API, ensuring local gateway data governance, compliance and control.
2. Control plane with Dimension Remote AAA (Authorization, Authentication and Accounting): publish and subscribe ledger-assured city services, for compliance, billing and payment. Monitor: Helm and Collectd with cAdvisor plug-in for monitoring and log generation. ELK: Elasticsearch v6.2, with Logstash and Kibana plug-ins for scalable log handling and viewing.
3. Data plane with CIAP (Confidential, Integrity, Audit and Privacy): InfluxDB spatial and time series, for Digital Twin correlating who, what, when and where, including feeding AI. AI: Tensorflow neural net with Python-Skit, inference engine and OpenCV computer vision for facial and license plate recognition, hosting images, video and non-scalar data. Scale: Spark cluster, extremely scalable, fast, and local so no network bottlenecks
The Intel® Movidius™ NCS development and deployment solutions have grown in depth and breadth, enabling a diverse range of applications on multiple operating systems and hardware hosts. This session walks through the process of developing and deploying IoT edge applications using some of these developer tools.
Software containers are not a new technology, but their recent adaptation to the embedded industry offers new ways to dynamically deploy services on embedded devices that were once locked down by monolithic software stacks. Since embedded devices are typically low-power solutions with smaller footprints, new ways of managing embedded applications in containers becomes critical for creating extensible solutions with relatively limited system resources available.
The Internet of Things creates both challenges and opportunities as it brings together the physical and digital worlds. OT/IT convergence, for example, is implicit when connecting operational technologies and physical assets to IT-like computing resources and systems. The promise of this convergence is to yield IT scale, innovation and efficiencies to traditionally isolated or monolithic embedded systems. The challenge is that the underlying infrastructure within the operational domain typically has very different, and often more stringent, quality of service requirements, such as latency, safety or reliability characteristics.
In this session, Wind River outlines a vision for enabling an autonomous, software-defined world for critical infrastructure. Join us as we explore the challenges of enabling applications across a range of computing infrastructure and mixed criticality systems.