How Tao Testing Achieved Cost Savings and Reliability with Kedify’s HTTP-Based Autoscaling Solution

The Challenge

• Tao Testing is a leading provider of assessment software for public and private institutions, enabling educators to create, manage, and deliver tests to students. The core challenge for Tao Testing lies in managing unpredictable traffic surges when thousands of students access the platform simultaneously. With test campaigns often coinciding nationwide, the platform must support fast, secure response times, from a stable baseline to significant, often sudden, demand peaks.
• To handle these unpredictable surges, Tao Testing previously over-provisioned its infrastructure, incurring significant costs from maintaining extra resources that were often underutilized. Attempts to use Prometheus for scaling solutions introduced delays that left the system vulnerable to downtime and user disruptions during high-demand periods.
• Seeking a solution that could adapt dynamically, Tao Testing turned to Kedify’s KEDA-based platform to ensure reliable, event-driven scalability tailored to their specific usage patterns.

The Solution

To address Tao Testing’s needs, Kedify developed an autoscaling solution centered around KEDA and enhanced by an Envoy-based HTTP scaler, purpose-built for Tao Testing’s traffic patterns.

Tao Testing was already using Envoy, which made Kedify’s integration both seamless and highly effective. The solution enabled rapid response scaling with no lag, allowing the platform to adjust resource allocation in near real-time.

This approach gave Tao Testing confidence that scaling would occur exactly when needed, without impacting service quality. Kedify’s solution effectively transformed Tao Testing’s infrastructure management, enabling rapid scaling with reduced complexity and greater operational efficiency.

With this setup, Kedify translated each incoming request as a demand for resources, automatically adapting the infrastructure to match the exact usage level and allowing Tao Testing to remain responsive even during critical peak times.

The Impact

This solution brought immediate, measurable benefits to Tao Testing, including significant cost savings, improved DevOps efficiency, and reliable scalability. Automating the scaling process removed the need for constant manual oversight, freeing up valuable resources and allowing the team to focus more on innovation. As a result, Tao Testing’s platform can now handle a staggering 200x increase in request volume, from 10 to over 2000 requests per second, without compromising performance or user experience.

“Before Kedify, scaling up was a constant challenge. Now, our platform adapts instantly to our users’ needs, and we’ve freed up our team to focus on new features rather than managing resource spikes,” says Tovar.

The shift to an event-driven scaling model has allowed Tao Testing to meet user expectations consistently, even during intensive campaigns. This approach also brought down operational costs by optimizing resource allocation precisely to the level of user demand, ensuring efficient use of infrastructure without sacrificing reliability.