Overcome these common edge computing challenges –

Overcome these common edge computing challenges: edge computing In many ways is a natural extension of other distributed architectures, especially hybrid cloud environments.

There are immense potential benefits – it explains Growing interest and adoption – Combining cloud and edge can be even stronger.

“By utilizing edge computing, edge security, and the appropriate network, cloud computing services are improved and complemented.

This benefit must be paid for. Distributed IT environments are by their very nature complex. Avoiding trends is not justified by this. It’s merely a motivation to plan, then plan some more.

Let’s take a look at the four main challenges edge computing poses and insights on how to tackle them.

Challenge #1: Managing Highly Distributed Environments

Managing a large edge architecture is like managing hundreds or thousands of smaller IT environments. Nor can you dispatch a help desk expert every time an issue needs attention.

“There are probably a lot of devices at the edge and very little in the way of local IT staff,” says technology evangelist Gordon Haff. red hat recently told us.

Trying to manually handle things like configuration management, OS patches, other software updates, and a host of other common operational requirements can be a bit of a pain.

The same applies to standardization (or lack thereof) of operating systems and other software and tools.

“Deploying and operating a large distributed infrastructure is difficult enough without mixing randomness and silos,” Haff says.

Teams already managing hybrid cloud or multi-cloud systems (especially those with a mix of containerized applications) probably already “got it”, but it’s always a good memory. You need a practical plan for day two and beyond.

Overcome this challenge: Automation and consistency are the pillars of edge operations.

“Automation and management are essential for tasks such as large-scale configuration, taking action in response to events, and centrally updating applications,” says Haff. “The maximization of consistency from the data center to the edge follows from this necessity. Cluster orchestration and standardized operating system settings are your friends.”

Challenge 2: Finding the Right Problem Solution

To realize the benefits of edge computing, thoughtful strategy and Careful evaluation of your use case. This allows the advantage to reduce the natural complexity of the edge environment. (See issue #1.)

“Edge computing should not be adopted or imposed by CIOs merely because it is trendy. There are actual issues that need to be resolved, and not every scenario involves them.” A simile.

Part of the inherent challenge here is where one of the biggest edge computing problem solvers fits. latency – It has a fundamental appeal. Not many IT leaders aspire to Slower application. But that doesn’t mean it’s a good idea (or feasible) to move everything from the data center or cloud to the edge.

“For example, A self-driving car, for instance, might have some of its workloads in the cloud, but by its very nature, it must respond to events very fast [to avoid danger] and internet access is essential. We must do that even when we aren’t available, according to Linden. “In this situation, edge computing makes sense.”.”

Many IT leaders don’t want slower applications. But that doesn’t mean it’s a good idea (or feasible) to move everything from the data center or cloud to the edge.

In Linden’s own work (Asimily does IoT security for healthcare and medical devices), a detailed look at the workload is required to optimize the cost-effectiveness assessment.

“Data security issues would arise if all network traffic gathered by IoT devices were sent to the cloud, which would require an excessive amount of bandwidth. to the cloud, then send the gleaned insights back,” explains Linden. But a lot of our programs, including those that compute device risk scores, may easily be used in a standard cloud paradigm.”

Overcome this challenge: Honesty conducts internal cost-benefit assessments of potential edge use cases and focuses on when performance/latency (or other goals) really matter. Remember: As with the cloud, you don’t have to take an all-or-nothing approach.

Challenge #3: Building an ecosystem that is still mature

From a software perspective, the same or similar set of fundamental challenges apply everywhere, whether it’s an edge node, the cloud, or a traditional endpoint. Memfault:

• How do you deploy it and track its versions?
• How to monitor performance / What is an observability strategy?
• How do you track defects?
• How do you detect and mitigate security risks?

Cloud and cloud-native applications already have strong answers to these kinds of questions. Baldassari is Site reliability engineering As an area that inherently exists to address these and other challenges in modern software and infrastructure. DevOps, DevSecOps, GitOpsand other disciplines as well offer overlapping approaches to these types of challenges.

These practices can be extended to edge deployments, but the edge ecosystem itself is still in its infancy compared to the cloud.

“The ecosystem is still immature and there is a hodgepodge of more or less imperfect solutions on the market today,” says Baldassari.

Like cloud computing before it, the edge computing ecosystem will mature over time. In the meantime, IT leaders should exercise caution when choosing platforms, tools, and other requirements.

Overcome this challenge: “CIOs and other IT executives should collaborate with their suppliers to learn more about their dependability strategy, the SLAs they provide, and the resources available to IT teams for monitoring edge devices on the company network.,” says Baldassari. “When all else fails, it is necessary to rigorously manage and limit what edge devices can do on the network using the right network policies.”

Challenge #4: Approaching Edge Infrastructure and Application Security

Just as security has become a major subdomain of cloud computing, so will it be a major priority in edge computing. Edges, by definition, extend the risk surface.

“Edge computing brings great benefits to the user experience, but at the cost of fundamental security concerns,” he said. hidden layer.

Also, like cloud computing, the edge itself is no “less secure” than a centralized environment. Building a strong stance requires management-backed priorities and some modern strategies.

Security experts like Sestito say that in edge environments, for example, a granular approach at the user or node level is important. “Access management is even more important, and multi-factor authentication should be employed wherever possible,” he says.

security automation – Sestito cites XDR and EDR tools as examples – As well as careful consideration of how data is exposed or moved between network devices is important here.

Edge initiatives can be a reason to start if you’re not already using a Zero Trust model.

“The use of Zero Trust security design principles is rapidly becoming the standard of choice for well-segmented and well-protected corporate resources.” recently told us.

Overcome this challenge: Prioritized properly, edge security can actually strengthen an organization’s overall posture in a more distributed IT reality. Leaders and teams who have already taken the test A layered approach to cloud and container security You should feel at home.

“Industry leaders are realizing that not all IT and business network requirements can be solved using cloud-centric enterprise architectures alone,” Howell says. “Cloud computing services are enhanced and complemented by the use of edge computing, edge security, and the right network.”