Corporate IT infrastructure has changed a lot in the past decade. From a relatively simple bounded space, with a defined “inside” and “outside,” IT networks have evolved to incorporate a wider range of devices, such as smartphones and tablets, and a growing amount of traffic from additional diverse networks, including the public Internet. However, nothing has the potential to disrupt traditional infrastructure topologies more than the Internet of Things (IoT). This has implications for infrastructure and operations (I&O) teams, as well as developers who are responsible for IoT solutions. A recent Forrester report titled “Edge Computing: IoT Will Spawn A New Infrastructure Market” highlights many of the changes and challenges that must be faced in this rapid evolution. Let’s take a look at a few of the highlights.
- Consider the full breadth of devices: The “things” that are connected in IoT require new approaches to development and management, but these endpoints are not the only new hardware you have to consider. Diverse components, including field-located IoT gateways and micro-datacenters, will become part of the networked environments. The need for edge infrastructure will depend on how much latency can be tolerated in the system and the complexity of the operations that need to be performed on data. Developers should understand not only IoT endpoints and hubs, but also the architecture in between and how to best take advantage of it to maximize performance and value. For scenarios in which advanced processing needs to happen on devices, Azure IoT Edge provides a fully managed service that delivers cloud intelligence locally.
- Adopt a hybrid mindset: As in many other areas of IT, solutions in the IoT space often use a mix of private and public cloud. IT professionals need to understand cloud development and operations principles and best practices, while being able to incorporate owned infrastructure and proprietary line-of-business software into the mix.
- Be open to new architectures: IoT systems can be deployed with a wide range of topologies. The control plane may be centralized in the cloud, hosted near the edge, or distributed across multiple locations. Components may be connected to more than one system, requiring a strict hierarchy to prevent conflicts. The flexibility to join, move, and remove components across different systems may also be required. The right approach to these issues is unique to your IoT implementation.
- Use software-defined approaches: With a wide range of devices, network protocols, security requirements, and data processing demands, the abstraction provided by the software layer is critical to connecting IoT components into functional, manageable, and secure systems. Using a platform-as-a-service (PaaS) approach can abstract away many of the complexities of managing IoT systems at lower levels and enable you to focus on optimizing the system for your business needs.
- Don’t go at it alone: The capacity to handle real-time data and scale to large numbers of devices is made much easier with the use of a cloud platform. For example, Azure IoT Hub and other Azure IoT platform services can give you complete, managed solutions for key challenges, including device provisioning and management, secure authentication, messaging, data processing, and more. Additionally, using such a platform enables you to connect easily to other services that may be critical to achieving your goals, such as machine learning and event handling.
There are many other worthwhile insights to be gained from the Forrester study, so make sure to download the full report.
Source: Azure Blog Feed