The Insider’s Guide to Building an Intelligent Edge

Building an Intelligent Edge by Scott Millmaker, Director of Sales Engineering with Rahi Systems

Many organisations have adopted a cloud-first IT strategy to improve the way they do business. Public, private and hybrid cloud platforms maximize efficiency while allowing IT staff to manage resources, save costs, increase scalability, streamline and coordinate data management, and reduce the human error factor. But while the cloud has many benefits, it is not always the right fit for modern IT workloads. Performance issues can be experienced when providing services to remote locations.

With the growth of the Internet of Things (IoT) and developments in augmented reality and artificial intelligence, many organisations want to extend their cloud to edge data centres that are closer to users and IoT devices. Edge data centres provide faster access to data, enhanced regulatory compliance, and the connectivity needed to collect real-time data that can be pushed into the cloud for analysis.

Building an edge data centre can be challenging. In some cases, organisations are faced with implementing an edge data centre in a harsh environment with limited space and IT resources. Other times, it may be an ideal location that allows them to follow traditional data centre design practices. In both scenarios, careful planning and execution are required to create a scalable edge data centre that supports business needs without impacting existing services or introducing new risks.

If your company is working with an outside service provider it is imperative that they have experience in designing, building and supporting data centres that meet customer up-time and SLA requirements. They will also need to be capable of looking at the bigger picture when choosing technology, with features that support and integrate with existing systems and applications, manage costs, and enable growth of the business.

It is important to have the chosen engineering team meet with you and carry out a full discovery of requirements for each step. They must take the time to understand the edge location and existing technology and applications being used, and any preferences for technology vendors and features.

This is followed up with a site survey to gain a clear picture of any challenges and obstacles that will need to be considered.

Step 1 – Physical. The physical infrastructure is the foundation layer that supports the edge data centre. We will review power, cooling and space requirements to support the desired up-time SLA and mitigate risks that could cause outages and impact application or service availability.

Step 2 – Security. Security is crucial to the success and performance of the edge data centre. A combination of hardware and software security is needed to protect applications, data and the network from actions or events that could cause serious damage to the organisation. Mechanical, electronic or software-defined controls are also needed to prevent unauthorized access to the physical IT environment. Depending on the customer’s requirements, this can be as simple as locks and keys or more sophisticated multi-factor authentication with alerting when an event occurs. CCTV systems can be used to monitor the site remotely and provide record retention to support business, security and compliance requirements.

Step 3 – Connectivity. Robust network infrastructure is needed for high-performance, scalable, reliable and secure access to the edge data centre. Engineers should have extensive experience in ensuring that wired and wireless LAN and telecom services support mission-critical applications and edge services. Ideally, they will also provide cabling services and work with innovative technology partners to provide connectivity options to future-proof the network.

Step 4 – Management and Deployment. A secure remote management platform is needed to control the entire technology stack. In most cases, this will need to be a vendor-neutral solution that seamlessly integrates with existing and new systems and addresses the critical needs of networking, in-band and out-of-band IT infrastructure.

Because it’s no longer acceptable to dispatch preconfigured equipment containing sensitive information such as user credentials and network configurations, organisations need a management system that enables remote automated deployment of new equipment. Ideally, this will be a cloud-ready solution which allows for a hands-free approach to managing and deploying infrastructure which in turn will provide visibility into the entire global IT estate from a single interface.

Step 5 – Compute and Storage. With the huge variety of compute and storage solutions available today, it can be difficult to choose the right platform for various workloads. The edge environment creates an additional set of requirements that must be considered. It has been recommended that leveraging converged infrastructure and software-defined systems to design an edge data centre that is easy to deploy, highly resilient, and capable of operating in the environment dictated by the edge location.