Canix Upgrades to Rails 7.2, Boosts App Performance by 25%

At Canix, our mission is to continuously refine our platform and provide the best user experience possible. With that goal in mind, we’re excited to announce our recent upgrade to Rails 7.2. This update has brought significant improvements to performance—specifically, another 25% speed boost across the app—as well as more stable and predictable load times for all our users.

Below, we’ll take you through some of the key metrics we’ve been tracking and the steps we took to achieve these gains.

Why Rails 7.2?

Rails 7.2 brings a host of updates and improvements, including:

1. Enhanced Performance: The framework introduces optimizations that reduce overhead and speed up server responses.
2. Improved Caching and Database Interactions: Better support for caching strategies and updates to database adapters lead to lower response times.
3. Quality of Life Upgrades for Developers: New features and refined tools help our engineering team be more productive, ensuring we can implement updates and respond to feedback faster.

For us at Canix, the move to Rails 7.2 was primarily driven by the desire to keep everything up-to-date and improve speed at scale. The results have been incredibly positive.

Key Performance Metrics

1. Overall Speed Improvement

• Another ~25% speed improvement across the app
• After thorough load testing and monitoring in production, we saw a consistent 25% drop in app-wide response times. This means pages load faster, and tasks run more efficiently, directly impacting user satisfaction.

2. Average Response Time

• Average response time now < 100ms
• One of our big wins is bringing the average response time below 100ms. This is a crucial target because sub-100ms responses often feel instantaneous to users, making the app feel snappier.

3. Improved P99 Times

• P99 down to 400-600ms on average
• Moving the 99th percentile (P99) of requests to 400-600ms means that even in worst-case scenarios, our platform remains responsive. This dramatic improvement is a key driver in reducing frustration for users during peak loads.

4. Better Percentiles Across the Board

• P50, P95, and P99 load times all showing some improvement
• Performance improvements aren’t just about the median (P50) or the tail end (P99). We’ve seen gains across our entire range of response times. This consistency reflects better resource allocation, more efficient queries, and improved caching strategies.

‍