You Need an Edge Strategy; Here’s How to Build the Right One for You

Having a plan to support emerging edge use cases is essential to ensuring business agility and limiting risk and complexity

Eric Olinger
You Need an Edge Strategy; Here’s How to Build the Right One for You

Digital transformation is happening at the edge. So many of the advanced applications that offer compelling innovations—including AI, machine learning and IoT—depend on having the right edge infrastructure in the right locations. When enterprises lack an edge strategy, they’ll find themselves managing the edge on an ad hoc basis.

Rather than designing their edge to drive greater business value, these businesses will be stuck in reactive mode. As new edge use cases continue to emerge, they’ll have to throw together custom solutions on the fly to support them. This will limit their business agility and make it difficult to future-proof their operations. The Gartner® report Building an Edge Computing Strategy puts it best:

“Edge computing without a strategy will eventually cause digital gridlock.”[1]

Enterprise IT leaders increasingly recognize the need for an edge compute strategy: The same Gartner report found that, “by year-end 2026, 70% of large enterprises will have a documented strategy for edge computing, compared to fewer than 10% in 2023.” If your company is among the many that will be adopting an edge strategy over the next several years, here are a few things you should consider as part of that strategy:

  • Defining digital edges
  • Identifying optimal locations
  • Ensuring proper connectivity
  • Reviewing edge security requirements

Defining digital edges and identifying optimal locations

There are many factors driving digital infrastructure to the edge. Among them are the need to ensure proximity for low-latency applications, keep the cost of data bandwidth low and meet data residency requirements. These drivers all fit under the umbrella of data gravity. This term refers to the tendency for large datasets to draw in workloads and infrastructure. This happens because data has inherent business value; therefore, the more data you have in a location, the more likely you’ll want to have infrastructure nearby to service that data.

One reason companies are struggling with their edge strategies is that the edge is not a specific location, nor even a particular architectural environment. The edge is simply anywhere your private digital infrastructure ends and public digital infrastructure begins. Most enterprises will have many diverse edges, and they’ll need to optimize each one for performance, cost-efficiency and security.

Most enterprises want to incorporate public cloud services into their edge strategy. They naturally assume this means they’ll have an edge at the public cloud, in addition to their various on-premises edges. However, this is a misconception; there are several reasons you wouldn’t want a public cloud edge. For instance, this would mean putting your workloads and data directly into the public cloud. This can drive higher costs, poor performance and loss of control over your data. This is particularly true if you use the public internet to move your data to and from the cloud.

A better option might be to host your edge at a digital infrastructure provider that also hosts low-latency cloud on-ramps. This allows you to establish a cloud adjacent data architecture, which brings together a core data storage environment with multicloud access on demand. With a cloud adjacent data architecture, you effectively place your public cloud services on the correct side of the edge, giving you greater control over costs and performance.

Ensuring proper connectivity

If an edge strategy is all about effectively managing connections between private and public infrastructure, this logically means you’ll need to incorporate both private and public networking. As mentioned previously, there are drawbacks to relying on the public internet in certain situations. For instance, leveraging the internet for public cloud access is notoriously the most expensive method. Also, since public internet connections are shared, your traffic is not prioritized over anyone else’s. This means that using the internet can exacerbate performance issues caused by latency. Finally, there are security and compliance concerns that go along with using the internet for certain workloads.

In contrast, using direct, private connections helps ensure your data gets from Point A to Point B quickly, all while staying in your control. Using a virtual networking solution such as Equinix Fabric®, you can create private connections between different edge environments on demand. When building out your cloud adjacent data architecture, you can connect your core data storage environment to multiple public clouds and to your various on-premises edge locations. Data can easily flow across the entire environment to go wherever it’s needed, from edge to core to cloud.

However, there are other edge use cases for which public internet connectivity will be essential. You need to get the best combination of availability and cost-efficiency across your edge locations, and you can’t get that by going all in with a single internet service provider (ISP). You must diversify your connections, and you can do this via blended internet access. With blended internet, you get connectivity from multiple ISPs via a single contract. This means you’ll be able to work with the ISPs that provide the best service and reliability in each of the edge locations you operate in.

Reviewing edge security requirements

The growth of edge infrastructure creates new challenges around cybersecurity. Today’s organizations no longer have centralized data centers to build security perimeters around. Instead, they need to establish and maintain a secure edge, which means embedding security infrastructure across a growing and dynamic perimeter.

In addition to providing performance and cost benefits, deploying edge security with a digital infrastructure provider can be significantly easier than securing and managing a public cloud edge. Imagine you need to deploy a virtual firewall device to protect your edge environment. If you treat the public cloud as your edge, that means you’re deploying the device directly with the cloud provider. You’ll end up deploying different instances for each cloud region and each process you need to support (development, testing, QA and so on), creating a complex and always-growing array of cloud connections. In addition to the sprawling number of security devices, you’ll likely need load balancers to achieve high availability.

In contrast, you could deploy a pair of highly available devices in a digital infrastructure environment and have the cloud access sit behind the devices. From that single pair of devices, you can create thousands of cloud connections, without having to manage each connection.

The strength of your digital ecosystem is essential for implementing a secure edge strategy. Working with leading security service providers can help you get the on-demand capabilities you need to implement a Secure Access Service Edge (SASE). A SASE architecture incorporates SD-WAN technology that provides the speed and flexibility you need from an edge networking solution, along with Security as a Service capabilities you can quickly deploy anywhere you need them.

Learn how Equinix can support your edge strategy

An effective edge strategy requires you to identify edge use cases that may arise in the future and create a plan to quickly and flexibly deploy the infrastructure to support those use cases. Whatever that infrastructure ends up looking like, it should offer high performance, cost efficiency, built-in security and easy access to ecosystem partners.

Equinix is the only digital infrastructure provider that can offer all these capabilities, along with a global footprint of colocation data centers to quickly scale your edge anywhere you need to. With the global digital infrastructure that Equinix operates, you can leverage the economic advantage of having an arm’s reach to thousands of network and internet service providers who can collectively lower your connectivity costs.

In addition, our data centers offer low-latency cloud on-ramps to all major cloud providers. This allows you to make multicloud access a part of your edge strategy without the cost and complexity of managing a public cloud edge.

Our partner ecosystem includes leading security service providers like Zscaler and Palo Alto Networks. Their services are available at Equinix via a variety of deployment options. For instance, you could deploy them as physical devices within your own colocation environment or as virtual network functions using Equinix Network Edge.

Finally, the most important benefit Equinix can offer to your edge strategy is our people. Every day, we work with customers to help them build the infrastructure they need to support a future-proof edge, and we can do the same for you. Learn more about how you can benefit from working with Equinix Global Solutions Architects, our team of experts specializing in distributed IT infrastructure.

Also, read the Gartner report Building an Edge Computing Strategy for more insights on how the right strategy can help you avoid edge sprawl, increase speed and flexibility and meet your biggest digital transformation priorities.

 

[1] Gartner, Building an Edge Computing Strategy, Thomas Bittman,  April 12, 2023.

GARTNER is a registered trademark and service mark of Gartner, Inc. and/or its affiliates in the U.S. and internationally and is used herein with permission. All rights reserved.

Subscribe to the Equinix Blog