data_fabric

DataFabric provides flexible database connection switching for ActiveRecord.

We needed two features to scale our mysql database: application-level sharding and master/slave replication. Sharding is the process of splitting a dataset across many independent databases. This often happens based on geographical region (e.g. craigslist) or category (e.g. ebay). Replication provides a near-real-time copy of a database which can be used for fault tolerance and to reduce load on the master node. Combined, you get a scalable database solution which does not require huge hardware to scale to huge volumes. Or: DPAYEIOB - don’t put all your eggs in one basket. :-)

Installation

gem install data_fabric

then in config/environment.rb, add a config.gem entry:

Rails::Initializer.run do |config|
  ...
  config.gem 'data_fabric'

or add data_fabric to your Gemfile:

gem 'data_fabric'

data_fabric has been tested with Rails 2.3.10 and 3.0.7.

How does it work?

You describe the topology for your database infrastructure in your model(s).

Different models can use different topologies.

class MyHugeVolumeOfDataModel < ActiveRecord::Base
  data_fabric :replicated => true, :shard_by => :city
end

There are four supported modes of operation, depending on the options given to the data_fabric method. The plugin will look for connections in your config/database.yml with the following convention:

No connection topology: #environment - this is the default, as with ActiveRecord, e.g. “production”

data_fabric :replicated => true

#environment_#role - no sharding, just replication, where role is “master” or “slave”, e.g. “production_master”

data_fabric :shard_by => :city

#group_#shard_#environment - sharding, no replication, e.g. “city_austin_production”

data_fabric :replicated => true, :shard_by => :city

#group_#shard_#environment_#role - sharding with replication, e.g. “city_austin_production_master”

When marked as replicated, all write and transactional operations for the model go to the master, whereas read operations go to the slave.

Since sharding is an application-level concern, your application must set the shard to use based on the current request or environment. The current shard is set on a thread local variable. For example, you can set the shard in an ActionController around_filter based on the user as follows:

class ApplicationController < ActionController::Base
  around_filter :select_shard

  private
  def select_shard(&block)
    DataFabric.activate_shard(:city => @current_user.city, &block)
  end
end

Warnings

  • Sharded models should never be placed in the session store or you will get “Shard not set” errors when the session is persisted.

  • DataFabric does not work in development mode with cache_classes = false. You will get an error in ActiveRecord’s clear_reloadable_connections! method.

Testing and Bug Reports

If you think you’ve found a problem with data_fabric, please use the example23 or example30 application to reproduce the bug and send me the diff.

Thanks to…

Rick Olsen of Rails Core

for the Masochism plugin, which showed me how to bend AR’s connection handling to my will

Bradley Taylor of RailsMachine

for the advice to shard while at acts_as_conference

Author

Mike Perham <[email protected]> mikeperham.com twitter.com/mperham

LICENSE:

Copyright © 2010 Mike Perham

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the ‘Software’), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED ‘AS IS’, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.