Edge data centre are gaining popularity rapidly for one simple reason: they provide faster services with low latency.
Edge data centre facilities, which are located close to their customers, are designed to deliver cloud computing resources and cached content to end users efficiently. Typically, the facilities are connected to a larger central data centre or to a network of data centres. Edge computing enables organisations to reduce latency and improve overall performance by processing data and services as close as possible to end users.
Edge computing, as provided by edge data centres, has the potential to be a transformative technology for IT leaders seeking to deliver both resilience and performance. According to IDC, the global edge computing market will reach $250 billion by 2024, growing at a compounded annual rate of 12.5 percent. According to Gartner, approximately 75% of enterprise-generated data will be created and processed outside of the traditional data centre or cloud by 2025.
Edge data centres are highly regarded for their ability to collect and process data locally while maintaining a high level of availability. Organizations that design applications and business processes to run independently can increase the survivability of critical business functions, according to Carl Fugate, lead of Capgemini Americas’ cloud and edge network centre of excellence. “When properly designed, this can mean business as usual or slightly reduced functionality in the event of simple failures, such as the loss of WAN connectivity,” he explained. “There are also significant benefits for IoT, as data loss or an inability to process and respond to data in real time can render systems ineffective or unavailable.”
Edge data centres can benefit offices, teams, retail locations, and other widely dispersed locations in terms of performance and efficiency. “Edge data centres eliminate much of the complexity associated with requiring all offices, locations, and workers to pass through a non-edge centralised data centre,” said David Linthicum, Deloitte Consulting’s chief cloud strategy officer.
Additionally, edge data centres are attractive to organisations with business or technology functions that cannot be supported by traditional wide area network (WAN) connections, as well as to entities that require real-time processing and storage of locally generated data. “We see this in manufacturing, distribution, energy, hospitality, and retail, where services cannot be completely centralised,” Fugate explained. “We also see this frequently in environments with operational technology (OT) networks, where sites have interdependent machines and processes.”
IT organisations considering an edge computing migration should begin by inventorying their applications and infrastructure. Additionally, it is prudent to assess current and future user requirements, with a particular emphasis on the sources of data and the actions that must be performed on that data. “In general, the more susceptible data is to latency, bandwidth, or security issues, the more likely a business will benefit from edge computing capabilities,” said Vipin Jain, CTO of edge computing startup Pensando. “Concentrate on a small number of pilot projects and collaborate with integrators/ISVs who have experience with similar deployments.”
Fugate recommended that businesses examine their functions and processes and connect them to the application and infrastructure services on which they rely. “This will eliminate the possibility of a single critical centralised service disrupting critical business functions,” he explained. “The objective is to ascertain which functions must continue to operate in the event of an infrastructure or connectivity failure.”
Fugate also recommended determining how to manage and secure distributed edge platforms effectively.
“The consolidation of services to the cloud has simplified management and security by leveraging cloud platform tools that are not available with some edge deployments,” he observed. “This is critical to consider, as tasks such as patching, backups, and security can be significantly more difficult to implement and manage at remote sites.”
Automation, security, and resilience are critical considerations when designing a highly federated data centre model, according to Simon Pincus, vice president of engineering at network monitoring company Opengear. “Recent high-profile failures of content delivery networks (CDNs) have demonstrated the extent to which a business can be harmed by an unreliable service,” he noted. “From the first planning session forward, organisations should consider failure scenarios and the management and restoration of networks and services.”
Pincus also suggested that designers of edge networks consider decoupling the management plane from the primary network in order to ensure that operations can continue even if primary connectivity is lost across a distributed network. “Ideally, the management plane will enable network automation, enabling reliable deployment, reconfiguration, and monitoring,” he explained.
The breadth and depth of available platforms for edge deployments provide a great deal of flexibility. “To maximise your organization’s potential, focus on developing teams and partnerships capable of developing cloud-independent solutions,” Jain advised. “There are a number of open-source architectures and technologies available that can help you accelerate development and avoid vendor lock-in.”