Choosing the right computing paradigm for industrial transformation

There has been a rapid increase in data volume collected by manufacturers. They are examining the advantages and limitations of cloud and edge computing and the potential benefits of a hybrid approach to optimize operational efficiency and decision-making.

By Manish Jain and Achim Thomsen June 7, 2024
Courtesy: Brett Sayles

 

Learning Objectives

  • Identify the significant increase in data volume collected by manufacturers and its impact on data processing needs.
  • Compare the advantages and limitations of cloud and edge computing in the context of industrial data management.
  • Explain the importance of a hybrid approach combining cloud and edge computing to optimize operational efficiency and decision-making in manufacturing.

 

Edge and cloud computing insights

  • The 2023 MLC Data Mastery and Analytics survey reveals a significant increase in data volume collected by manufacturers, necessitating efficient data processing strategies.
  • Cloud computing offers scalability, cost-effectiveness and accessibility, while edge computing provides reduced latency, enhanced security and improved reliability.
  • A hybrid approach that combines cloud and edge computing can help manufacturers leverage the strengths of both paradigms to optimize operations and drive innovation.

The 2023 MLC data mastery and analytics survey found that more than one-third of manufacturers say the volume of data they’re collecting has at least doubled in the last two years and nearly 20% say the amount of data has at least tripled. While this surge of data presents opportunities for manufacturers, it also presents a critical question: where should manufacturers process and analyze this ever-growing volume of information?

Enter cloud and edge computing: two distinct approaches to data processing in the industrial data management landscape. Cloud computing, with its centralized servers and vast storage capacity, offers scalability and accessibility. Edge computing brings processing power closer to the data source, enabling real-time decision-making and low latency.

Both cloud and edge computing come with their own set of advantages and limitations. Understanding these nuances is imperative for manufacturers to make the right decision that aligns with their unique IX goals.

Advantages of cloud versus edge computing

Cloud computing has revolutionized data management by shifting away from on-premises infrastructure. This computing paradigm allows manufacturers to transmit vast amounts of industrial data to information technology (IT) and operational technology (OT) applications through an internet connection, unlocking a range of benefits:

  • Scalability: Cloud computing offers unparalleled flexibility to adapt to dynamic business needs. The scalability of cloud services allows organizations to easily adjust computing resources based on fluctuations in demand. Whether experiencing a sudden surge in data processing requirements or scaling down during periods of reduced activity, cloud platforms provide the agility needed to optimize costs and maintain operational efficiency. This adaptability is particularly beneficial for businesses with varying workloads or evolving data processing demands.

  • Cost-effectiveness: One of the hallmark advantages of cloud computing is its cost-effectiveness. By outsourcing the management of software and back-end infrastructure to cloud providers organizations can sidestep the need for dedicated IT personnel and costly hardware investments. This translates to substantial cost savings and a reduction in operational complexities.

  • Accessibility: Cloud computing enhances accessibility by enabling seamless data access and collaboration from virtually any location with an internet connection. This flexibility is crucial in today’s globalized business landscape, allowing teams to work collaboratively and access data in real time, irrespective of geographic boundaries. Cloud-based applications facilitate remote work, empowering employees to be productive and make informed decisions regardless of their physical location.

In contrast to cloud computing’s centralized approach, edge computing embraces a decentralized architecture that processes and stores data closer to its source. Edge computing has gained significant traction in recent years, with Gartner predicting that over 75% of enterprise data will be created and processed outside the data center or cloud in just the next five years. This shift toward the edge offers significant benefits to manufacturers, including:

  • Reduced latency: One of the key advantages of edge computing is its ability to reduce latency by processing data at its source. This translates into a significant reduction in network travel time, making it a game-changer for applications where speed is of the essence.

  • Enhanced security: The decentralized nature of edge architecture means that sensitive data is processed and stored locally, minimizing the need for extensive data transfers over networks. Edge computing’s localized data processing and storage approach means that sensitive information remains within the confines of the organization, drastically reducing the surface area for potential cyberattacks.

  • Improved reliability: Edge computing plays a pivotal role in ensuring the reliability of critical systems. In traditional cloud computing models, disruptions at the central data center can impact the entire system. With edge computing, the decentralized architecture distributes computing power across multiple edge devices, reducing the risk of a single point of failure. This distributed approach not only enhances system reliability but also ensures continuous operation even in the face of network outages or disruptions.

Cloud and edge limitations

While cloud and edge computing offer immense potential for businesses, they are not without their challenges.

Cloud computing’s dependence on connectivity can be a significant hurdle for industrial settings. Constant internet access isn’t always guaranteed in remote locations, factories or situations with fluctuating network strength. This can lead to disruptions in operations, delays in decision-making and hinder overall efficiency. The sheer volume of data generated by industrial processes can also make cloud-based processing expensive. The constant data flow not only incurs high costs but also creates bottlenecks and potential lags in processing.

Perhaps the most critical limitation of cloud computing for certain applications is its inherent latency. The time it takes for data to travel to the cloud, be processed and return with instructions can range from seconds to minutes. This delay is unacceptable for applications demanding real-time responses, such as automated industrial processes. In time-sensitive scenarios, even a minor delay can have significant consequences, impacting operational efficiency and potentially compromising safety.

While edge computing offers solutions to cloud’s limitations, it comes with its own set of considerations. Managing a vast network of edge devices and sensors distributed across various locations adds significant complexity to the IT infrastructure. This requires specialized expertise and can strain existing IT resources, especially for smaller organizations.

The additional hardware and software needed to implement edge computing can also translate to higher costs compared to traditional cloud-based architectures. This can be a barrier for organizations with limited budgets or those hesitant to invest in a new infrastructure.

Harnessing the best of both worlds for IX

Both edge and cloud computing offer unique advantages, however the “best” choice hinges on the specific needs and priorities of manufacturing organizations, which involves careful consideration of factors such as cost, security, latency and the reliability of internet connections.

Industrial leaders may find cloud solutions more advantageous when dealing with applications that demand extensive computational power and storage capacity. Large-scale data analytics, machine learning and centralized data processing are instances where the scalability and flexibility of cloud infrastructure can shine. Conversely, edge computing is the more fitting choice for industrial leaders when real-time processing and reduced latency are crucial. Edge computing is ideal for manufacturing use cases where split-second decision-making is critical, such as in autonomous robotics, quality control on the production line and equipment monitoring.

In many cases, manufacturers opt for a hybrid approach that combines cloud and edge computing to take advantage of the strengths of both paradigms. For example, a manufacturer can use edge computing for real-time sensor data processing and anomaly detection to trigger immediate maintenance actions and send non-critical data to the cloud for long-term storage, analysis and optimization insights.

This hybrid approach empowers manufacturers to seamlessly navigate the demands of their industrial landscape, leveraging the strengths of both paradigms to optimize operational efficiency, enhance decision-making processes and ultimately drive innovation.

Cloud computing and edge computing both have important futures in supporting manufacturing. To make the best choice for each application, it is crucial to make informed architectural decisions early in the process. This includes planning for hybrid deployments, considering the total cost of ownership and ensuring alignment with the organization’s overall security posture. By taking these factors into account from the start, manufacturers can more easily determine whether cloud computing, edge computing or a combination of both is the most suitable option for their specific needs.

Manish Jain is the product leader for edge analytics and AI applications at Rockwell Automation. Achim Thomsen is the director of common connected applications at Rockwell Automation. Edited by Tyler Wall, associate editor, Control Engineering, WTWH Media, twall@wtwhmedia.com.


Author Bio: Manish Jain is the product leader for edge analytics and AI applications at Rockwell Automation. Achim Thomsen is the director of common connected applications at Rockwell Automation.