[PERF] Regression in 4.2+ when creating/updating records

[tgxworld]

The following URLs link to the benchmark scripts and results that we've benchmarked for every Rails releases since 3.2.0:

• http://rubybench.org/rails/rails/releases?result_type=activerecord/sqlite3_create
• http://rubybench.org/rails/rails/releases?result_type=activerecord/sqlite3_save
• http://rubybench.org/rails/rails/releases?result_type=activerecord/sqlite3_update

Benchmark Results ran on my local machine for the sqlite3_update benchmark:

{
  "label":"activerecord/sqlite3_update",
  "version":"5.0.0.alpha",
  "iterations_per_second":6294.703952433717,
  "iterations_per_second_standard_deviation":2.0493418113829613,
  "total_allocated_objects_per_iteration":184
}

{
  "label":"activerecord/sqlite3_update",
  "version":"4.2.1",
  "iterations_per_second":6276.6898115635695,
  "iterations_per_second_standard_deviation":0.9877818063555915,
  "total_allocated_objects_per_iteration":188
}

{
  "label":"activerecord/sqlite3_update",
  "version":"4.1.10",
  "iterations_per_second":9944.386576623452,
  "iterations_per_second_standard_deviation":1.9106256432813897,
  "total_allocated_objects_per_iteration":133
}

Do let me know if I can provide more information and I'll be happy to help.

cc/ @sgrif @senny

[rafaelfranca]

@tgxworld thank you so much! I can't express with words how you work is important for us so I'll give you some hearts ❤️ 💚 💙 💛 💜

[sgrif]

Thanks for the additional info. There was a lot of focus on getting reads back up to speed so we could get 4.2 out, since writes are generally less of a bottleneck than reads. I have a couple of ideas of ways we can improve it, but I just haven't had time yet. This is very high up on the list, though.

[tgxworld]

@rafaelfranca 😄

@sgrif No worries! 😄 I just wanted to highlight those regressions first.

[sgrif]

❤️

[kernel-io]

❤️ ❤️

[rails-bot]

This issue has been automatically marked as stale because it has not been commented on for at least
three months.

The resources of the Rails team are limited, and so we are asking for your help.

If you can still reproduce this error on the 4-2-stable, 4-1-stable branches or on master,
please reply with all of the information you have about it in order to keep the issue open.

Thank you for all your contributions.

[sgrif]

Sorry for the delay in getting to this. Rails 5 now should be comparable to 4.1 in this regard. We're still slightly slower, but it's much less drastic, and I should be able to eke out several more minor improvements before we ship. Either way I think we're close enough to call this fixed.