We've been using find_in_batches to reduce memory usage, and recently
noticed one of our more intensive background processes had a huge
memory footprint (600mb+) and was getting killed by our memory
monitor. We were unable to reproduce this in development, and after
investigation, the culprit is query_cache. Wrapping the task in
ActiveRecord::Base#uncached kept the task stable (~200mb), and it's
not hard to imagine why. Looping over thousands of items while eager
loading many more likely grows the cache to huge amounts, which seems
counter to the use case for find_in_batches.
So first of all, this is an FYI. Beyond that, all the ways in which we
use find_in_batches would be aversely affected by the query_cache;
sure, it *might* make a query faster, but it *definitely* will grow in
memory, as you are expected to use it across thousands of records.
Given that find_in_batches' use case is to reduce memory when
searching across thousands of records, should it not be default
behavior to disable query cache for find_in_batches operations?
--
You received this message because you are subscribed to the Google Groups "Ruby on Rails: Talk" group.
To post to this group, send email to rubyonrails-talk@googlegroups.com.
To unsubscribe from this group, send email to rubyonrails-talk+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/rubyonrails-talk?hl=en.
No comments:
Post a Comment