In Chapter 5, Assessing Industrial Internet Applications, we described some considerations regarding applications and emerging frameworks. As this book was being published, industrial providers of applications and public Cloud Services Providers (CSPs) were continuing to develop frameworks of components useful in developing these applications, along with new IIoT applications. While some of the industrial providers initially focused on deploying their applications and frameworks to their own clouds, there appears to be movement by many of these companies to redeploy their offerings upon public clouds, leveraging the significant investments in infrastructure and aggressive pricing provided by leading CSPs.
The major CSPs (such as AWS, Google, IBM, Microsoft Azure, Oracle, and others) continue to extend functionality of components in their IIoT frameworks. Some have created applications development workspaces focused on broad areas with preconfigured components to speed applications development. A sample of the areas being addressed by packaged development environments includes the connected factory, remote monitoring, predictive maintenance, and connected field services. Systems integrators frequently team with the CSPs to offer services needed to build out these IIoT solutions.
Many industrial manufacturers have adopted one or more of these frameworks and offer IIoT applications that manage and monitor the equipment that they sell. For example, Rockwell Automation offers solutions for providing intelligence and manages manufacturing and production and energy utilization on the plant floor. Johnson Controls offers IIoT applications to control and manage building HVAC systems and enable enhanced troubleshooting.
As some of the extremely large and diversified industrial companies created IIoT applications for their line of business (LOB) and the devices that they manufacture, they saw an opportunity to promote their underlying frameworks for usage in further development. This appears to be a growing trend. GE's Predix platform is probably the most well-known example.
GE focused initially on building applications used in deploying and managing GE assets in industries that included transportation, aviation, healthcare, industrial manufacturing, mining, oil and gas, power distribution and generation (including wind), transportation, and water resources. GE Digital is creating a new architecture for applications – combining systems of record with systems of innovation – to power digital industrial companies. The Predix platform enables the building of solutions that include scheduling and logistics, connected products, intelligent environments, field force management, and predictive and prescriptive asset optimization and repair. The pre-built applications powered by Predix platform include Asset Performance Management (APM), Operations Performance Management (OPM) and Brilliant Manufacturing Suite (BMS). The APM suite gives industrial businesses a complete, integrated view of their assets and equipment at all levels, allowing for more intelligent decision-making and improved operations. The OPM provides advisory analytics from the boardroom to the asset level, combining real-time and historical data to power decision-making with a single dashboard view. Brilliant Manufacturing is the digital technology behind Brilliant Factories. It enables manufacturers to make precise, real-time decisions through data-driven insights. With the acquisition of ServiceMax, a leader in cloud-based field service management (FSM) solutions, GE Digital is poised to reinvent the way assets are managed, maintained and serviced, reducing downtime and removing costs.
The Predix foundation utilizes microservices, an increasingly popular approach that speeds code reuse, development, and deployment. Microservices are self-contained software services with well-defined APIs, and they communicate with each other using a standard communications protocol. Predix has defined microservices that address assets, analytics, data, security, and operations.
Microservices are sometimes deployed to manage single aspects of a physical asset. By breaking the management into microservices, upgrades for each management aspect become faster and simpler.
Predix applications are deployed in the multi-tier architecture that we described throughout this book and consist of edge devices, networks, and backend components. The Predix foundation can be looked at as a PaaS model and utilizes Pivotal's Cloud Foundry.
The Cloud Foundry is an industry standard open source cloud application platform for developing and deploying enterprise cloud applications. It helps automate, scale, and manage cloud-based applications throughout their life cycle. Microservices and applications can be written in multiple languages and are deployed in container images in many different cloud infrastructures.
Many other frameworks were being promoted by other industrial applications providers as this book was being published. Emerson, Honeywell Process Solutions, and Schneider Electric each promoted their own frameworks (known as PlantWeb, Uniformance Suite, and Exostructure, respectively). For example, Schneider Electric's Exostructure framework provides an energy-focused IIoT architecture and platform. It is used primarily for managing buildings, data centers, industrial facilities, and power grids. The architecture was created to solve problems in machine automation and process automation, especially in the oil and gas industry, food and beverage industry, mining and metal industry, minerals and cement industry, and at water and wastewater facilities.
These frameworks share much in common. They rely on popular communications protocols for networking, usually including support for emerging popular standards. They typically support certain devices and sensors that have been tested and that are certified to be part of the solution. Some examples of the devices offered in these frameworks include transmitters, flow meters, analyzers, actuators, and controllers.
The frameworks and applications are continuing to mature. Some industry analysts speculated a few years ago that every industrial company would need to become a software company in the future. However, we believe it is more likely that most industrial companies will simply purchase complete integrated suites that include specific devices, networking, and backend cloud-based infrastructure from a few of the major vendors and industrial companies making large investments in developing these solutions.
A growing trend in many industrial companies is the deployment of multi-cloud solutions. One reason for doing this is the belief that deploying on multiple CSPs will assure highly competitive pricing among the CSPs in the future. Another driver of multi-cloud strategies are the frequent mergers and acquisitions common for many industrial companies. For example, a company managing plants that have a mixture of Emerson, Honeywell, and Schneider Electric controls from a headquarters location might want to gather data that resides in multiple CSPs and multiple regions. Network exchange vendors are typically engaged to provide a network bridge among these sites to facilitate data exchange. It is important for the architect to consider the need for and cost of highly available network architectures that connect the devices to clouds or are used to transmit data among multiple clouds.