The world is increasingly adopting cloud-first strategy for a number of reasons such as increased scalability, flexibility, and agility at far lesser costs when compared to on-premise infrastructure and applications.
To help customers migrate IoT solutions from on-premise to Azure, Microsoft launched Azure IoT Suite so that businesses can leverage the power of cloud along with IoT. Azure IoT Suite comprises of IoT Hub, Machine Learning, Stream Analytics, Notification Hubs and PowerBI services.
While consumer apps such as fitness monitors and self-driving cars attract the most attention and can create significant value, it is estimated that B2B apps can generate nearly 70 percent of potential value enabled by IoT (Source: Mckinsey).
How Microsoft Azure IoT can create successful IoT solutions
1. Get started with pre-configured solutions
The Azure IoT Suite preconfigured solutions are applications of common IoT solution patterns such as remote monitoring and predictive maintenance. These solutions are complete, production-ready, end-to-end solutions that include simulated devices, Azure services, and solution-specific management consoles. In addition to deploying and running the solutions in Azure, you can download the source code and customize a solution to meet your specific IoT requirements.
2. Connect millions of devices
Azure IoT Hub easily integrates with the existing devices and supports a broad set of operating systems and protocols. With it, you can send thousands of bi-directional messages to heterogeneous devices through a mobile push-notification system. For a device to connect with the solution back end, it must identify itself to IoT Hub using valid credentials.
To get started, you can use Azure IoT Starter Kits that include development boards consisting of sensors and actuators. Or you can choose from hundreds of Certified for IoT devices that are platform-independent and tested to connect easily to IoT Hub.
Azure governs the rate at which new device connections can be established with IoT hub – to protect the service from DoS attacks, but the number of devices to connect simultaneously is independent of the number of units purchased.
For instance, if you buy one S1 unit, you get a throttle of 100 connections/sec. So, to connect 100,000 devices, it will take at least 1000 seconds. However, you can simultaneously connect any number of devices that are registered in your identity registry. If you want to connect more than 500,000 devices to a single IoT hub, you will need to contact Microsoft Support.
3. Analyze and visualize data
Currently, most of the IoT data generated is not used. For instance, only 1 % of data from an oil rig with 30,000 sensors is utilized for anomaly detection and control, not optimization and prediction, which provide the greatest value.
With Azure Event Hubs, it is possible to combine data from IoT Hub with other data sources and processing engines without losing the streaming nature of the computations. You can also create an Azure function app to get incoming messages from your IoT hub and write them to Azure Table storage.
Stream Analytics connects directly to Event Hubs and IoT Hubs for stream ingestion and ingests historical data into Azure Blob service. To get real-time insights of your business operations and performance of devices, use Stream analytics along with PowerBI, or with Azure Machine learning for predictive analytics. If companies have expertise in open source solutions, they can use HDInsight or Apache Spark for advanced analytics.
Integration of various line-of-business applications can provide valuable insights from your operational data which can further be used to make smarter decisions. Past telemetry data can be used to identify patterns that indicate maintenance is due, or say if a temperature is above normal, send a turn-off command automatically.
4. Enhance security
For both azure-to-device and device-to-azure messages, devices must first be registered and have valid credentials. Azure IoT has an identity registry which stores the authentication and security related data for each IoT device. It is stored separately from other device metadata to avoid any unintended access to identity information of IoT devices. You can retrieve the device credentials from the solution dashboard and include in your client application.
5. Scale with efficiency
You can start with Adafruit Raspberry Pi 3 kit which has spreadsheets, word-processing, and video games. It runs on Windows 10 and has all necessary sensors and components. Or you can use Seeed – Intel Edison Kit ideal for prototyping projects or commercial solutions where performance matters. Once PoC project is successful, scale to broader deployment.
Moreover, companies only pay when they are using the resources. Thus, starting with a PoC is a very cost-effective and practical approach to see how Azure IoT can help you create a successful IoT implementation.
With more than 100 Azure datacenters, Azure IoT Suite ensures continuous availability of your apps. Azure uses multiple data centers to reduce latency and store information closer to the device location to provide the best performance. You can first localize your solution and then scale your solution for multiple locations.
.png){kind=icon}