In this blog, we will dive deeper into edge computing itself and examine the advantages and challenges before expecting success in Industrial Transformation (IX). Examining these aspects would help IX leaders ground their decisions for a robust edge management solution.
A quick refresher on the Edge
Since we're going to evaluate the edge computing paradigm, it's important to be reminded what it is. After a decade of IT datacenter consolidation and migration to public cloud infrastructure, architects have realized that centralized computing architectures have limitations around bandwidth and response times.
There are several business use cases that require the moving of data-based intelligence near the source data for faster response time, rather than the other way around. The edge computing model enables computing near the physical data collection and analysis location, rather than on a centralized server or in the cloud.
While the data is kept close to source to securely process it in real time on-site - other devices like industrial PCs are still connected to the network. This computing model is a key enabler of IT/OT convergence. It creates new and improved ways for industrial and enterprise-level businesses to maximize operational efficiency, improve performance and safety, automate all core business processes, and achieve always-on availability.
"By 2023, more than half of new enterprise IT infrastructure will be at the edge, according to IDC. By 2025, Gartner predicts 75% of enterprise-generated data will be created and processed outside a traditional datacenter or cloud." - TechTarget
Edge and cloud computing models are complimentary to each other and provide a flexible solution based on the unique needs of each organization. For real-time collection and analysis, the edge is ideal. At the same time, for centralized large-scale analytics, the cloud is ideal. Together they can provide real-time and longer-term insights into performance initiatives like machine learning and asset performance management.
Advantages of Edge Computing
Broadly speaking, edge computing can enhance OT solutions by addressing cloud computing issues like performance, latency, bandwidth, security, and proximity. Through these, it can power the next industrial revolution, transform manufacturing and promote an agile business ecosystem that's more efficient and easier to manage. Let's dive a little deeper into the key advantages:
The most common motivator for edge solutions is attributed to network latency, which represents a delay between an application request and the resulting response. For real-time applications, even milliseconds matter. In an OT environment, unpredictable latency could mean an increase in defect rates or safety issues.
Lower Bandwidth Limitations
With the proliferation of IoT devices and other connected equipment at the edge, data generation is growing exponentially. Depending on the location, high-speed network connections may not be available or could be prohibitively expensive. Thus, to address AI use cases requiring large data volumes at the edge, organizations are using edge computing to reduce data transmission requirements by filtering and processing data locally.
This is a direct advantage of achieving lower latency and bandwidth. By placing infrastructure and applications closer to data generation and consumption, the overall performance is optimized—leading to better user experience.
Edge computing also limits the movement of data—helping address corporate governance policies or other regulations that require sensitive information to remain onsite. In Europe, the General Data Protection Regulation (GDPR) dictates data sovereignty rules as to where data can be communicated and stored. Other jurisdictions are following suit with legislation such as the California Consumer Privacy Act (CCPA). Edge provides more control over such compliance.
Better Business Continuity
Lastly, edge computing improves business continuity. As more organizations automate operations, it is imperative that the underlying systems remain available even if the network or cloud is not. Every major cloud provider has had an unforeseen service outage over the past year—and will most probably continue to do so. However, Applications running on edge infrastructure can continue to function until network services are restored, limiting the impact on business operations.
Challenges around the Edge
It's time to examine the other side of the coin and assess the risks that need to be countered in edge computing initiatives. Remote edge locations have relatively limited resources unlike a corporation headquarters location, where there may be an on-premises datacenter, high-speed Internet connectivity for access to the cloud, and on-site IT staff. Let's dive a little deeper:
Single point of failure due to less IT equipment and space
Few organizations can afford to install redundant high-end servers, storage and networking equipment in all their remote locations. Instead, they may install critical applications on commodity servers (and in some cases, desktop PCs) that don’t have redundant components (such as power supplies and hard drives). Worse still, edge locations may not have a server room or other environmentally controlled space for IT equipment. The lack of redundancy, adequate power, cooling or ventilation can introduce a single point of failure.
Slower and more expensive Internet connectivity
Many remote locations today are connected to the corporate datacenter and the cloud via a broadband Internet connection from a local Internet service provider (ISP) with minimal service level agreements (SLAs). If broadband Internet isn’t available in certain remote locations, more expensive private circuits may need to be installed or the site may need to use 4G or 5G cellular connections with relatively limited and expensive data plans. For organizations that don’t deploy any infrastructure at their remote locations, an Internet outage can mean a complete work stoppage at a remote location.
Fewer IT staff
Specialized IT staff are generally located at the corporate headquarters. Remote locations may have little or no on-site IT staff, which means a full stop when systems are down. For sites that have no on-site IT staff, remote access from the corporate headquarters to commodity IT equipment in the edge location can be challenging, with limited remote management tools and no hands-on capabilities.
With heightened awareness of the beneifts and potential areas of caution for the edge computing paradigm, IX leaders need to strategicaly factor these aspects into their IX initiatives. As noted earlier, a successful IX strategy with edge comptuing paradigm may require investment in additional infrastructure, network connectivity and experienced staff. However, IX leaders must also educate themselves on additional considerations for choosing the right edge management solution for their needs. Stay tuned for the next installment of our edge exploration journey.
Article Source: Rockwell Automation