Good Starting Points for Optimizing Database Calls in Ruby on Rails?
- by viatropos
I have a menu in Rails which grabs a nested tree of Post models, each which have a Slug model associated via a polymorphic association (using the friendly_id gem for slugs and awesome_nested_set for the tree).
The database output in development looks like this (here's the full gist):
SQL (0.4ms) SELECT COUNT(*) AS count_id FROM "posts" WHERE ("posts".parent_id = 39)
CACHE (0.0ms) SELECT "posts".* FROM "posts" WHERE ("posts"."id" = 13) LIMIT 1
CACHE (0.0ms) SELECT "slugs".* FROM "slugs" WHERE ("slugs".sluggable_id = 13 AND "slugs".sluggable_type = 'Post') ORDER BY id DESC LIMIT 1
Slug Load (0.4ms) SELECT "slugs".* FROM "slugs" WHERE ("slugs".sluggable_id = 40 AND "slugs".sluggable_type = 'Post') ORDER BY id DESC LIMIT 1
SQL (0.3ms) SELECT COUNT(*) AS count_id FROM "posts" WHERE ("posts".parent_id = 40)
CACHE (0.0ms) SELECT "posts".* FROM "posts" WHERE ("posts"."id" = 13) LIMIT 1
CACHE (0.0ms) SELECT "slugs".* FROM "slugs" WHERE ("slugs".sluggable_id = 13 AND "slugs".sluggable_type = 'Post') ORDER BY id DESC LIMIT 1
Slug Load (0.4ms) SELECT "slugs".* FROM "slugs" WHERE ("slugs".sluggable_id = 41 AND "slugs".sluggable_type = 'Post') ORDER BY id DESC LIMIT 1
...
Rendered shared/_menu.html.haml (907.6ms)
What are some quick things I should always do to optimize this from the start (easy things)?
Some things I'm thinking now are:
Can Rails 3 eager load the whole Post tree + associated Slugs in one DB call? Can I do that easily with named scopes or custom SQL?
What is best practice in this situation?
Not really thinking about memcached in this situation as that can be applied to much more than just this.
