Migration to AWS Cloud Boosts Site Performance by 30%

The Solution

The Difference Between Blade Servers and AWS Cloud Hosting

Fine Linens’ Magento store was originally hosted on blade servers, which presented a series of challenges. In peak shopping seasons, when traffic surges occurred and site performance was most critical, the fixed resources of those servers struggled to accommodate the increased demand. This often led to slow load times and even complete site outages and downtime, causing unhappy customers to leave and Fine Linens to lose out on revenue. In addition, maintaining such a robust infrastructure to handle these spikes year-round was skyrocketing in cost. After performing a thorough analysis, our Chief Architect recommended that Fine Linens switch to AWS Cloud.

Managing Cloud Resources with Automated Accuracy

We also implemented Terraform for Infrastructure as Code (IaC), which helps make managing cloud resources more efficient and reliable. A key advantage is that it helps reduce the risk of human error, with the ability to define and provision infrastructure through declarative configuration files. On Fine Linens, Terraform helps promote consistency, streamline the deployment process, improve team collaboration—all while enhancing the scalability of infrastructure management.

Optimizing for Speed and Efficiency

To better optimize Fine Linens’ new infrastructure for users, we focused on speed and scale. Varnish and Nginx make a powerful combination that not only improves overall site speed and user experience, but also allows for better resource utilization. This makes it easier to scale applications during peak traffic periods (aka critical revenue-earning periods).

Varnish acts as an effective caching layer that accelerates content delivery by storing and serving cached copies of web pages, significantly reducing server load and response times. Meanwhile, Nginx serves as an efficient web server and reverse proxy, capable of handling a large number of concurrent connections with low resource consumption. With load time and resource drain down, sales can go up.

Prepared for Traffic Surges & Spikes

Elastic Load Balancing (ELB) in conjunction with Auto Scaling Groups is crucial for effectively managing spikes in traffic, particularly for dynamic applications. ELB automatically distributes incoming application traffic across multiple servers, ensuring no single server becomes overwhelmed during peak periods.

Meanwhile, Auto Scaling Groups dynamically adjust the number of running instances based on real-time traffic demands, automatically launching new instances to accommodate increased load and terminating them when demand decreases.

Armed Against DDoS Attacks

Another key piece of Fine Linens’ new infrastructure is the security. They needed a solution that would filter out harmful traffic, so legitimate customers can shop easily without slowdowns or interruption. We selected Cloudflare for DDoS protection, which helps keep websites safe from malicious attacks that can overwhelm servers and cause downtime.

Cloudflare not only improves the website’s reliability, but also enhances overall security, allowing Fine Linens to focus on operations without worrying about potential disruptions from DDoS attacks.

Reducing Errors with Proactive, Automated Testing

Tools like Selenium (New Relic Synthetics) can significantly reduce errors and downtime. Selenium can simulate user interactions and continuously monitor applications, quickly identifying issues or regressions that may arise after updates or changes. These automated tests provide real-time insights into site performance without pulling Fine Linens’ staff away from sales and operations.

Coupled with log analysis tools like Sentry, Fine Linens can now track errors, analyze patterns, and diagnose problems more effectively. This proactive approach enhances the overall quality of the website and infrastructure and helps minimize downtime with rapid detection and resolution of issues (ideally, before they even impact end users).