In the age of the Industrial Internet of Things (IIoT), the transformation of manufacturing environments into smart factories has become a reality. This shift is driven by advancements in artificial intelligence (AI) and cloud computing technologies that enhance machine performance and optimize manufacturing operations.
To achieve this, specialized edge hardware must interact with operational technology (OT), which encompasses various industrial machines and assets that often utilize their unique fieldbus protocols. This edge hardware not only translates these protocols but also processes the data before relaying it to Amazon Web Services (AWS).
In this article, we will delve into how EXOR International’s systems-on-module (SOM) and edge gateways, powered by Intel’s Cyclone V FPGA, empower system integrators and application developers to create AWS-based IIoT solutions more rapidly, while minimizing total cost of ownership (TCO) and development efforts. As an AWS Partner Network (APN) Advanced Technology Partner with the AWS IoT Competency, Intel plays a significant role in this development.
Industrial Internet of Things
The convergence of cloud computing, machine intelligence, and ubiquitous internet access has ushered in the IIoT era, highlighted by the emergence of smart factories and enhanced manufacturing processes. Various IIoT applications cater to different use cases, ranging from efficiency improvements and cost reductions to production optimization and quality enhancement.
Implementing these applications necessitates a scalable and reliable cloud infrastructure for data aggregation, storage, analytics, and visualization. Edge devices are essential for data collection, machine control, and cloud connectivity. In some scenarios, data analytics and AI models are executed directly on these edge devices.
As illustrated below, edge devices can efficiently capture and process data from motors, programmable logic controllers (PLCs), and other operational technologies. The diagram showcases the typical division between OT and IT, encompassing on-premises network infrastructure and AWS.
AWS for Industrial IoT
Successful IIoT implementations require cloud infrastructure capable of meeting the following criteria:
- Functionality: OT generates vast amounts of unstructured data. IIoT analytics necessitate the aggregation, transformation, storage, and analysis of this data. AWS offers the building blocks to facilitate these functions, allowing for the seamless creation of data pipelines.
- Scalability: IIoT deployments often involve numerous endpoints across diverse locations, each consistently generating data. AWS is equipped to manage substantial volumes of data, whether at-rest or in-transit, without sacrificing performance or responsiveness.
- Reliability: Industrial processes relying on cloud-based functions demand robust and highly available infrastructure, as interruptions can lead to significant financial losses.
- Security: Since machine data frequently contains sensitive information, it is crucial to protect this data during transmission from edge to cloud and while stored on AWS. The cloud infrastructure must also feature secure access controls to safeguard the factory’s operations.
AWS’s infrastructure is not only robust and secure but also highly elastic. The extensive AWS service portfolio provides the necessary components for data pipelines tailored to various IIoT applications. Notable AWS services include AWS Lambda for data transformation, Amazon S3 for storage, Amazon Aurora for database support, Amazon EMR for big data analytics, and Amazon QuickSight for data visualization. For an in-depth exploration of IIoT, check out this blog post.
Industrial Edge Devices
Generally, IIoT edge devices can be categorized into two types: Field Gateways (FGW) and Industrial PCs (IPC). A field gateway collects data from operational technology endpoints and forwards it to AWS. Additionally, the FGW must accept command and control messages from AWS and relay them to the appropriate endpoint, which may include sensors, PLCs, and other machinery controllers.
Communication with these endpoints typically occurs over a fieldbus, necessitating that the FGW supports the specific protocol employed by the fieldbus. Numerous fieldbus types exist on factory floors, each with its unique protocol, including Ethercat, Profinet, Ethernet IP, Modbus TCP, and many more.
In contrast, an Industrial PC not only performs all the functions of a field gateway but also supports advanced edge workloads, such as data analytics and AI inference using deep learning algorithms. Both types of edge devices commonly require support for time-sensitive networking (TSN), which ensures deterministic latency in local area network communications.
With TSN, edge devices can synchronize with other network devices to execute real-time tasks. For instance, an IPC can coordinate multiple robotic arms over a TSN, ensuring timely delivery of control messages. Edge devices must also be capable of operating continuously in industrial settings, often necessitating a compact form factor to fit into constrained spaces around machinery.
For further insights into building industrial edge devices, refer to this authoritative source that provides comprehensive information on the topic. Additionally, this resource on Amazon’s fulfillment center safety and training is excellent for understanding operational standards.
Leave a Reply