Optimizing Performance in Ruby on Rails

After over 5 years working with Ruby on Rails in different contexts - from public administration systems to high-demand e-commerce - I learned that performance is not just about fast code, but about understanding where the real bottlenecks are.

The N+1 Query Problem

One of the most common (and often ignored) problems in Rails applications are N+1 queries. In my experience at inChurch, I managed to reduce response time from +60ms to ~14ms just by optimizing these queries.

Identifying the Problem

# Bad: N+1 queries
posts = Post.all
posts.each do |post|
  puts post.author.name  # One query for each post!
end

The Solution

# Good: Eager loading
posts = Post.includes(:author).all
posts.each do |post|
  puts post.author.name  # No additional queries
end

Strategic Caching

Not everything needs to be recalculated on every request. At Gocase, we implemented layered caching that significantly reduced database load.

Fragment Caching

<% cache product do %>
  <%= render product %>
<% end %>

Russian Doll Caching

<% cache [product, product.category] do %>
  <div class="product">
    <% cache product.category do %>
      <%= render product.category %>
    <% end %>
    <%= product.name %>
  </div>
<% end %>

Profiling is Essential

Before optimizing, measure. Tools like rack-mini-profiler and bullet are essential for identifying real problems.

# Gemfile
group :development do
  gem 'rack-mini-profiler'
  gem 'bullet'
end

Conclusion

Performance is an ongoing process. The techniques I shared here are just the beginning - each application has its own challenges. The important thing is:

1. Measure before optimizing
2. Focus on real bottlenecks
3. Test changes in production

If you have questions or want to share your own experiences, find me on LinkedIn!