Network Computing is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

The Network Upgrades You Should Consider in 2021

network
(Source: Pixabay)

With 2020 receding into dismal history, the new year's arrival provides a perfect moment to look forward to a hopefully happier and calmer future. This is also a good time to examine existing network resources and to determine where improvements can be made to enhance service, performance, security, innovation, and other essential functions and attributes.

Quality and collaboration

In 2021, end-users will continue to demand more bandwidth, better Wi-Fi, and improved Quality of Experience (QoE), said Ed Fox, CTO at telecom service provider MetTel. "IT teams will need to accommodate [these needs] or risk alienating the user base," he warned, noting that employees in 2020 became very efficient and productive working at home. "Now, IT has to figure out how to make the office work more like home," Fox stated. Collaboration tools and VoIP-from-the-desktop/laptop will create a unique environment back in the office, he predicted. "Those [leaders] that make [the necessary] shifts early in 2021 will be the winners."

Double-down on security

As threats continue growing and evolving, many of the most critical network investments in 2021 will be security-related. "From DDOS mitigation to multi-factor authentication to IPsec, security investments take on countless forms," said Ali Marashi, CTO and senior vice president of engineering at co-location service provider vXchnge. "Think of network security as a constant journey forward, not a single event or annual checklist," he recommended. A security evaluation should coincide with every new development or change to the network to mitigate the risk of exposure, Marashi added.

Improve team coding skills

Every network engineer needs to start developing coding skills, whether it's Python, Ansible, or some other coding language, in order to stay competitive and decrease time to market, advised Josh Hoium, director of engineering for global network communications at Liberty Mutual Insurance and former director of engineering at Target. Having network engineering teams learn how to code is a significant commitment, he noted. "It's a heavy investment in your team and [is] often met with some pushback initially," Hoium explained. "Don't expect changes overnight." It's important to remember that in today's highly competitive marketplace, enterprises win or lose by their digital experiences. "The best software wins," he noted.

Embrace AI and ML

If they haven't already begun, network leaders should start harnessing artificial intelligence (AI) and machine learning (ML) technologies. These tools are essential to improve network efficiency and reliability as well as to prepare for the increasing connectivity demands that accompany emerging services and applications, advised Chris Bastian, CTO and senior vice president of engineering at SCTE-ISBE, a cable telecom industry association.

Network devices generate an avalanche of data, much of which now falls to the floor without being analyzed or leveraged. Meanwhile, networks continue to grow more complex and essential to daily operations. "AI and ML algorithms can automate the process of digging through this massive amount of data to detect and address correctable incidents, often before users even realize there's an issue," Bastian explained.

AI and ML tools can also have a substantial positive impact on network reliability, which directly impacts end-user QoE. "Organizations that don't prioritize improvements to their networks put customer satisfaction and even employee retention at risk," Bastian warned.

It's important to shop carefully for AI and ML tools and services, Bastien advised, to ensure that the technology selected matches the enterprise's specific needs. The technology also needs to be properly deployed. "Data sets should be cleaned and validated before they're analyzed and, even then, it's important to be cognizant of false correlations," Bastien cautioned. "IT leaders should also become comfortable with working closely with other business units to help determine what can be gleaned from automated reports," he added.

Consider deploying a hybrid cloud

A hybrid cloud infrastructure allows enterprises to address problems associated with exponential data growth while also prioritizing privacy, security, and compliance, observed Byron Chance, a senior network engineer with IT support and service company Guardian Computer. A hybrid cloud infrastructure combines a public and private cloud, making it easier to connect multiple devices, integrate IT resources, create tools to manage multiple clouds, and share information between different environments, he noted.

For optimal performance, a hybrid cloud network should incorporate tools and strategies that view the network as a whole rather than as individual pieces. "Keep in mind that you have less control over your information's security in a public cloud than a private cloud, so it's important to maintain proper separation of information between your public and private clouds," Chance recommended.