Intelligent IT solutions for the Internet of Things

By Patrick Steiner, Lead Architect at Red Hat.

The Internet of Things is changing business IT and holds great potential for companies. By analyzing data from networked devices, they can automate their business processes, increase productivity and lower costs. This requires a highly scalable, reliable and secure IT infrastructure, which should be based on standardized components and protocols and operate on three different layers: the device layer, the gateway layer and the data center layer.


The Internet of Things (IoT) networks intelligent devices of all kinds, such as sensors, mobile devices, machines or vehicles with each other and with the cloud. The analysis of IoT data offers great opportunities to companies – they can make decisions faster, optimize business processes or develop new applications or even business models. In this way, the Internet of Things impacts nearly every field, from energy, health, and transportation, to retail, hospitality, manufacturing and financial services.


This opens up a broad spectrum of new potential applications, ranging from intelligent building technology, automated lighting or energy management, intelligent manufacturing systems and optimized solutions for inventories, logistics and supply chain management to remote monitoring of patients’ vital data.


The sheer size and public nature of the Internet of Things, however, also involves huge technological challenges. Network and system architects have to optimize the IT infrastructure in order to meet the demanding requirements of the IoT in terms of scalability, reliability and security.


The Internet of Things poses entirely new challenges for scalability, for instance. In the study ‘Worldwide Internet of Things (IoT) 2013–2020 Forecast: Billions of Things, Trillions of Dollars,’ the market researcher IDC forecasts that more than 220 billion devices will be connected via the Internet of Things by 2020. A single intelligent system could then collect and analyze billions of data objects from millions of different endpoints. This will place unprecedented demand on processor performance, storage and networks.


IoT-based applications and automated business processes also place higher demands on the availability of the system. Many intelligent systems are used for mission-critical applications, and system failures can lead to lowered productivity, dissatisfied customers or a drop in sales. The same is true for emergency services, medical applications and monitoring solutions. In these cases a system failure can endanger property, the environment, people’s health or even lives.


Distributed IoT solutions create large-scale security challenges, since the systems are networked over the Internet and use processing capacity and storage resources from the cloud. That’s why companies need to expand their security infrastructure to efficiently protect themselves from data loss, theft and ever more sophisticated denial-of-service attacks. This infrastructure includes comprehensive authentication, authorizing and auditing features. These build confidence, regulate access to resources and ensure compliance with the legal guidelines and company policies. Companies should use powerful encryption methods in order to protect their intellectual property and customer data from theft.


Layer model satisfies demands
Intelligent IT solutions meet the requirements of IoT systems in terms of scalability, reliability and security. The solutions are based on a hierarchical model with a device layer, control layer and data center or cloud layer. The solutions also use standardized protocols and components.


The device layer includes a high number of intelligent devices, including mobile devices, wearables, sensors, control devices and autonomous machines and appliances. Communication between the devices and the control points is based on standard network protocols – either cable or wireless. Open messaging standards are also used for routing raw data and exchanging control system information. The device layer also includes gateways that allow for interoperability with older and proprietary devices.


The gateway layer serves as a link between the devices and data center or the cloud. It collects and saves data from the devices and sends them to the data center. Conversely, it also sends control information to the devices – all based on open messaging standards. The gateway layer also serves as a cache for data required for tactical analysis or regulatory standards.


In addition, the gateway layer plays a central role for the Business Rules Management System (BRMS), since it saves real-time data monitored by BRMS in order to speed up processes. This layer recognizes patterns in the data and functions according to predefined business rules. In addition, it distributes workloads and automates routine services for applications such as routing, transforming, distributing and aggregating data.


The data center and cloud layer is responsible for operating applications and services, analyzing data and adapting business rules abased on historical trends. It stores data for long-term analysis, contains the most user and management interfaces and offers a virtual environment for complex tasks, distributed computing and business analytics functions. Furthermore, aggregated data from the control layer is collected and examined in this layer and business rules are distributed downstream.


One alternative to the three-layer model outlined here is the two-layer model, in which devices are connected directly to the data centers or the cloud. This model is ideally suited to consumer applications that require less bandwidth and that do not need a gateway layer for the distribution of workloads.


Flexibility and lower costs
The multi-layer model and open standards allow the systems to utilize all the advantages of the Internet and the cloud – scalability, reliability, security and flexibility.
Scalability: The distributed model allows for maximum scalability and lowers costs. The intermediate control layer reduces the load on data centers and the cloud by processing raw data, carrying out business logic and communicating with devices. Since it also reduces devices’ demands on processor performance and storage, more simple, cost-effective devices can be used. The elements of the control layer can be geographically distributed in order to maximize performance (uniform distribution of computing, network and storage resources) and cost-effectiveness (lower costs for network bandwidth).


Reliability: The hierarchal topology is very robust on every layer and between layers. Architects can implement redundant elements and domains in order to eliminate single points of failure and deliver high quality service. A system with reliable network, computing and storage components ensures the continual availability of mission-critical applications.


Security: The multi-layer model offers multiple levels of security. Architects can implement different security measures at the individual levels (e.g. encryption, secure authentication, authorization, and auditing, as well as patch and configuration management) in order to protect the system from attacks and threats.


Flexibility: By using standard interfaces, companies gain a variety of protocols and products and avoid becoming dependent on any single manufacturer. They can utilize different components from different providers and freely combine these into one system that perfectly meets their requirements.


This type of flexible, high-performance system can master all the challenges that the Internet of Things poses with its millions of networked devices and huge data volume. As a result, companies can transform raw data into valuable information and create the conditions for long-term success in their business.