Ensuring Resilient Connectivity During the Holiday Rush

Black Friday, Cyber Monday, and the holiday season are always critical for retailers' bottom lines. Even so, they take on added importance this year as consumers adapt to uncertain economic conditions by keeping a sharper eye on potential deals.

Unfortunately, the holiday period poses a significant risk of outages that could result in lost revenue. Websites face heavy traffic far exceeding typical day-to-day visits, straining infrastructure to its breaking point. Around 80 million online shoppers made purchases on both Black Friday and Cyber Monday in 2021, and many more browsed retail websites. In addition to surges in legitimate traffic, distributed denial of service (DDoS) attacks are also growing in frequency each year and hindering some of the year’s biggest product launches and events.

Even as these challenges threaten to take retailers offline, customers’ standards for fast and responsive online experiences are higher than ever. In a world where over 50% of online shoppers expect a page to load in 3 seconds or fewer, customers will quickly abandon sluggish sites and apps in favor of competing retailers, taking their money with them.

As you prepare for the holiday rush, here are a few ways to improve your resilience, mitigate the impact of potential outages, and ensure customers have optimal e-retail experiences.

When uptime is a high priority, it’s essential to avoid putting all your eggs in one basket — or in this case, one data center. Your services are simply easier to incapacitate if a bad actor only has to knock out one target to take you offline. Even if you're not the target of a DDoS attack, your infrastructure might become collateral damage if a DDoS attack hits another client cohabiting in the same data center.

For this reason, many companies are now shifting to a distributed footprint, placing compute and network infrastructure at the edge to shrink the “blast radius” of a potential outage. Even if part of a distributed footprint does fail, it only impacts a small number of users, who can be quickly redirected elsewhere for minimal disruption.

Another way to reduce downtime risk is to implement traffic management strategies that rely on real-time data about the network, available resources, and user activity. This enables networking teams to deploy additional infrastructure to accommodate traffic spikes or route traffic around ongoing outages. Suppose there are strategic locations that typically have slow or unreliable internet connectivity. In that case, you can use traffic steering to spin up new cloud instances as needed to ensure users have unimpeded access to your site.

That said, it’s only possible to assess and respond to real-time conditions if you calibrate and utilize monitoring and observability tools. They allow you to identify and mitigate challenges before customers experience poor performance. Traditional observability approaches that rely on centrally compiling and analyzing data struggle to keep up with the rapid pace at which data is generated at the edge. However, emerging technology now facilitates rapid edge observability, even making it possible to generate analytics directly on the edge for faster insights and a shorter turnaround to resolve issues.

All traffic coming onto a website or app begins with the domain name system (DNS), so building it with resilience is essential. DNS resilience requires redundancy, meaning a secondary DNS that is always active but uses an entirely separate infrastructure from the primary one. If a DDoS attack takes the primary DNS offline, the secondary DNS can continue resolving queries. An anycast routing network will also help to direct DNS requests toward available servers in real time if the closest server goes down.

The traffic management strategies mentioned above support using real-time data for scaling infrastructure to meet demand but doing this manually is infeasible. Networks are becoming more complex, diverse, and distributed, and connected to more kinds of devices. The key to scaling with demand is network automation, which can modify infrastructure to accommodate fluctuations in demand.

Factors threatening network connectivity during the holiday rush are clear, but so are the steps you can take to address them. Shifting to a more distributed footprint, using traffic steering and real-time observability to identify and address concerns, ensuring resilience at the DNS layer, and implementing network automation can all prepare you to maximize revenue in the most important shopping season of the year.

Kris Beevers is co-founder and CEO of NS1.

Related articles: