Advanced Performance Optimization of Rails Applications

Project Management and Scrum Software
Alexander Dymo
RailsConf 2009
Advanced Performance Optimization

of Rails Applications
http://en.oreilly.com/rails2009/public/schedule/detail/8615
www.acunote.com
What Am I Optimizing? Project Management and Scrum Software
Acunote www.acunote.com
Online project management and Scrum software

Ruby on Rails application since inception in 2006
● ~3700 customers
● Hosted on Engine Yard
● Hosted on Customers' Servers
● nginx + mongrel
● PostgreSQL
Alexander Dymo • Advanced Performance Optimization of Rails Applications • RailsConf 2009 2 / 84
Performance Degradation Over Time Project Management and Scrum Software
300
Request Time (on development box), %
250 Actually Happens: O(nc)
Best Case: O(log n)

200
150
100
April 2008 May 2008 June 2008 July 2008

Solutions? Project Management and Scrum Software
Throw Some Hardware at it!

Solutions? Project Management and Scrum Software
Performance Optimization!

What to optimize?

What To Optimize? Project Management and Scrum Software
Development?

What To Optimize? Project Management and Scrum Software
Development
AND Production

How to optimize?

How To Optimize? Project Management and Scrum Software
Three rules of
performance optimization

Three Rules Of Performance Optimization Project Management and Scrum Software
1. Measure!
...the universal experience of
programmers who have been using
measurement tools has been that
their intuitive guesses fail...
Knuth

2. Optimize only what's slow!

3. Optimize for the user!

Things To Optimize Project Management and Scrum Software
● Development
– Ruby code
– Rails code
– Database queries
– Alternative Ruby
● Production
– Shared filesystems and databases
– Live debugging
– Load balancing
● Frontend
– HTTP
– Javascript
– Internet Explorer
Optimizing Ruby: Date Class Project Management and Scrum Software
What's wrong with Date?

> puts Benchmark.realtime { 1000.times { Time.mktime(2009, 5, 6, 0, 0, 0) } }
0.005
> puts Benchmark.realtime { 1000.times { Date.civil(2009, 5, 6) } }
0.080
16x slower than Time! Why?

%self total self wait child calls name
7.23 0.66 0.18 0.00 0.48 18601 <Class::Rational>#reduce
6.83 0.27 0.17 0.00 0.10 5782 <Class::Date>#jd_to_civil
6.43 0.21 0.16 0.00 0.05 31528 Rational#initialize
5.62 0.23 0.14 0.00 0.09 18601 Integer#gcd

Fixing Date: Use C, Luke!
Date::Performance gem with Date partially rewritten in C

by Ryan Tomayko (with my patches in 0.4.7)
> puts Benchmark.realtime { 1000.times { Time.mktime(2009, 5, 6, 0, 0, 0) } }
0.005
> puts Benchmark.realtime { 1000.times { Date.civil(2009, 5, 6) } }
0.080
> require 'date/performance'

puts Benchmark.realtime { 1000.times { Date.civil(2009, 5, 6) } }
0.006
git clone git://github.com/rtomayko/date-performance.git

rake package:build
cd dist && gem install date-performance-0.4.7.gem
Real-world impact of Date::Performance:
Before: 0.95 sec

After: 0.65 sec
1.5x!

Optimizing Ruby: Misc Project Management and Scrum Software
Use String::<< instead of String::+=

> long_string = "foo" * 100000
> Benchmark.realtime { long_string += "foo" }
in theory: 75x
0.0003
in practice: up to 70x
> Benchmark.realtime { long_string << "foo" }
0.000004
Avoid BigDecimal comparisons with strings and integers

> n = BigDecimal("4.5")
> Benchmark.realtime { 10000.times { n <=> 4.5 } } in theory: 4.5x
0.063 in practice: 1.15x
> Benchmark.realtime { 10000.times { n <=> BigDecimal("4.5") } }
0.014

● Development
– Ruby code
– Rails code
● Production
– Live debugging
– Load balancing
● Frontend
– HTTP
– Javascript
Optimizing Rails: Preloading Associations Project Management and Scrum Software
Preloading associations is always a good idea...

except when Rails can't do the job:
class Foo
belongs_to :bar
end
foos = Foo.find_by_sql('select * from foos inner join bar')

foos.first.bar #extra SQL query!

Optimizing Rails: Preloading Associations Project Management and Scrum Software
Virtual Attributes plugin:

http://github.com/acunote/virtual_attributes/
class Bar
end
class Foo
belongs_to :bar
preloadable_association :bar
end
foos = Foo.find_by_sql('
select * from foos
left outer join
(select id as preloaded_bar_id,
name as preloaded_bar_name
from bars) as bars
on foos.bar_id = bars.preloaded_bar_id')
foos.first.bar #no extra SQL query!

Optimizing Rails: String Callbacks Project Management and Scrum Software
What can be wrong with this code?
class Task < ActiveRecord::Base

before_save "some_check()"
end
...
100.times {
Task.create attributes
}
Kernel#binding is called to eval() the string callback

That will duplicate your execution context in memory!
More memory taken => More time for GC

Optimizing Rails: Partial Rendering Project Management and Scrum Software
Not too uncommon, right?
<% for object in objects %> #1000 times

<%= render :partial => 'object',
:locals => { :object => object } %>
<% end %>
We create 1000 View instances for each object here!

Why?

Template inlining for the resque:
<% for object in objects %> #1000 times

<%= render :partial => 'object',
:locals => { :object => object },
:inline => true %>
<% end %>
list.rhtml
_object.rhtml _object.rhtml
list.rhtml

Template Inliner plugin:

http://github.com/acunote/template_inliner/
Real world effect from template inlining:
Rendering of 300 objects, 5 partials for each object

without inlining: 0.89 sec
with inlining: 0.75 sec
1.2x
● Development
– Ruby code
– Rails code
● Production
– Live debugging
– Load balancing
● Frontend
– HTTP
– Javascript
Optimizing Database Project Management and Scrum Software
How to optimize PostgreSQL:

explain analyze
explain analyze
explain analyze
...

Optimizing Database: PostgreSQL Tips Project Management and Scrum Software
EXPLAIN ANALYZE explains everything, but...

... run it also for the "cold" database state!
Example: complex query which works on 230 000 rows and

does 9 subselects / joins:
cold state: 28 sec, hot state: 2.42 sec
Database server restart doesn't help

Need to clear disk cache:
sudo echo 3 | sudo tee /proc/sys/vm/drop_caches
(Linux)

Optimizing Database: PostgreSQL Tips Project Management and Scrum Software
Use any(array ()) instead of in()

to force subselect and avoid join
explain analyze select * from issues where id in (select issue_id from tags_issues);
QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------
Merge IN Join (actual time=0.096..576.704 rows=55363 loops=1)
Merge Cond: (issues.id = tags_issues.issue_id)
-> Index Scan using issues_pkey on issues (actual time=0.027..270.557 rows=229991 loops=1)
-> Index Scan using tags_issues_issue_id_key on tags_issues (actual time=0.051..73.903 rows=70052loops=1)
Total runtime: 605.274 ms
explain analyze select * from issues where id = any( array( (select issue_id from tags_issues) ) );
QUERY PLAN
------------------------------------------------------------------------------------------------------------------------------
Bitmap Heap Scan on issues (actual time=247.358..297.932 rows=55363 loops=1)
Recheck Cond: (id = ANY ($0))
InitPlan
-> Seq Scan on tags_issues (actual time=0.017..51.291 rows=70052 loops=1)
-> Bitmap Index Scan on issues_pkey (actual time=246.589..246.589 rows=70052 loops=1)
Index Cond: (id = ANY ($0))
Total runtime: 325.205 ms
2x!
Database Optimization: PostgreSQL Tips Project Management and Scrum Software
Push down conditions into subselects and joins

PostgreSQL often won't do that for you
select *,
Issues (select notes.author from notes
id serial where notes.issue_id = issues.id
name varchar ) as note_authors
org_id integer from issues
where org_id = 1
Notes
id serial select *,
name varchar (select notes.author from notes
issue_id integer where notes.issue_id = issues.id and org_id = 1
org_id integer ) as note_authors
from issues
where org_id = 1

What To Do? Project Management and Scrum Software
● Optimize For Development Box

– Ruby code
– Rails code
● Optimize For Production
– Live debugging
– Load balancing
● Optimize For The User
– HTTP
– Javascript
Alternative Ruby Project Management and Scrum Software
Everybody says
"JRuby and Ruby 1.9 are faster"
Is that true in production?

In short, YES!
= Acunote Benchmarks = MRI JRuby 1.9.1

Date/Time Intensive Ops 1.23 0.58 0.53
Rendering Intensive Ops 0.61 0.44 0.30
Calculations Intensive Ops 2.57 1.79 1.33
Database Intensive Ops 5.58 4.63 3.29

In short, YES!
= Acunote Benchmarks = MRI JRuby 1.9.1

Date/Time Intensive Ops 1x 2.1x 2.3x
Rendering Intensive Ops 1x 1.4x 2.0x
Calculations Intensive Ops 1x 1.4x 1.9x
Database Intensive Ops 1x 1.2x 1.7x
JRuby: 1.5x faster

Ruby 1.9: 2.0x faster

What is faster?

This is where all the improvement went:
Acunote Copy Tasks Benchmark MRI JRuby 1.9.1

Request Time 5.52 4.45 3.24
Template Rendering Time 0.53 0.21 0.21
Database Time 0.70 1.32 0.69
GC Time 1.07 N/A 0.62
Much faster template rendering!

Less GC!
JDBC database driver performance issue with JRuby?

Why faster?

Things I usually see in the profiler after optimizing:

%self self calls name
2.73 0.05 351 Range#each-1
2.73 0.05 33822 Hash#[]=
2.19 0.04 4 Acts::AdvancedTree::Tree#walk_tree
2.19 0.04 44076 Hash#[]
1.64 0.03 1966 Array#each-1
1.64 0.03 378 Org#pricing_plan
1.64 0.03 1743 Array#each
1.09 0.02 1688 ActiveRecord::AttributeMethods#respond_to?
1.09 0.02 1311 Hash#each
1.09 0.02 6180 ActiveRecord::AttributeMethods#read_attribute_before_typecast
1.09 0.02 13725 Fixnum#==
1.09 0.02 46736 Array#[]
1.09 0.02 15631 String#to_s
1.09 0.02 24330 String#concat
1.09 0.02 916 ActiveRecord::Associations#association_instance_get
1.09 0.02 242 ActionView::Helpers::NumberHelper#number_with_precision
1.09 0.02 7417 Fixnum#to_s

# of method calls during one request:

50 000 - Array
35 000 - Hash
25 000 - String
Slow classes written in Ruby:

Date
Rational

Alternative Rubys optimize mostly:

● the cost of function call
● complex computations in pure Ruby
● memory by not keeping source code AST

Alternative Rubys optimize mostly:

● the cost of function call
● complex computations in pure Ruby
● memory by not keeping source code AST

So, shall I use alternative Ruby?

Definitely Yes!... but
JRuby: if your application works with it

(run requests hundreds of times to check)
Ruby 1.9: if all gems you need are ported

● Development
– Ruby code
– Rails code
● Production
– Live debugging
– Load balancing
● Frontend
– HTTP
– Javascript
Optimizing For Shared Environment Project Management and Scrum Software
Issues we experienced deploying on Engine Yard:
1) VPS is just too damn slow

2) VPS may have too little memory to run the request!
3) shared database server is a problem
4) network filesystem may cause harm as well

VPS may have too little memory to run the request
Think 512M should be enough?

Think again.
We saw requests that took 1G of memory!
Solutions:
● buy more memory
● optimize memory
● set memory limits for mongrels (with monit)

You're competing for memory cache on a shared server:

1. two databases with equal load share the cache

You're competing for memory cache on a shared server:

2. one of the databases gets more load and wins the cache

As a result, your database can always be in a "cold" state

and you read data from disk, not from memory!
complex query which works on 230 000 rows and

does 9 subselects / joins:
from disk: 28 sec, from memory: 2.42 sec
Solutions:
optimize for the cold state
sudo echo 3 | sudo tee /proc/sys/vm/drop_caches
push down SQL conditions

fstat() is slow on network filesystem (GFS)
Request to render list of tasks in Acunote:

on development box: 0.50 sec
on production box: 0.50 - 2.50 sec


Couldn't figure out why until we ran strace
We used
a) filesystem store for fragment caching
b) expire_fragment(regexp)
Later looked through all cache directories even though we knew

the cache is located in only one specific subdir


Solution:
memcached instead of filesystem
if filesystem is ok, here's a trick:

http://blog.pluron.com/2008/07/hell-is-paved-w.html

● Development
– Ruby code
– Rails code
● Production
– Live debugging
– Load balancing
● Frontend
– HTTP
– Javascript
Live Debugging Project Management and Scrum Software
To see what's wrong on "live" application:

For Linux: strace and oprofile
For Mac and Solaris: dtrace
For Windows: uhm... about time to switch ;)
To monitor for known problems:

monit
nagios
own scripts to analyze application logs

● Development
– Ruby code
– Rails code
● Production
– Live debugging
– Load balancing
● Frontend
– HTTP
– Javascript
Load Balancing Project Management and Scrum Software
The problem of round-robin and fair load balancing
per-process queues
Rails App 1 1 2
Rails App 2 1 2 3
Rails App 3 1 2 3

Solution: the global queue
Rails App 1
Rails App 2 1 2 3 4 5
Rails App 3

Think you need mod_rails / Passenger for this?

You're right, but...
you can emulate this with nginx and mongrels

Dedicated queues for long-running requests
queue for long-running requests
Rails App 1 1
regular per-process queues
Rails App 2 1 2
Rails App 3 1 2 3

nginx configuration for dedicated queues

upstream mongrel {
server 127.0.0.1:5000;
server 127.0.0.1:5001;
}
upstream rss_mongrel {
server 127.0.0.1:5002;
}
server {
location / {
location ~ ^/feeds/(rss|atom) {
if (!-f $request_filename) {
proxy_pass http://rss_mongrel;
break;
}
}
if (!-f $request_filename) {
proxy_pass http://mongrel;
}
}
}
● Development
– Ruby code
– Rails code
● Production
– Live debugging
– Load balancing
● Frontend
– HTTP
– Javascript
Optimize For The User: HTTP Project Management and Scrum Software
Backend Network and Frontend

5% 95%
Things to consider:
● Gzip HTML, CSS and JS
● Minify JS
● Collect JS and CSS
(javascript_include_tag :all, :cache => true)
● Far future expires headers for JS, CSS, images
● Sprites
● Cache-Control: public
● everything else YSlow tells you

● Development
– Ruby code
– Rails code
● Production
– Live debugging
– Load balancing
● Frontend
– HTTP
– Javascript
Optimize Frontend: Javascript Project Management and Scrum Software
The worst enemy of every

web developer is...


Things you don't want to hear from your users:
"...Your server is slow..."
said the user after clicking

on the link to show a form
with plain javascript (no AJAX)

Known hotspots in Javascript:

- eval()
- all DOM operations - avoid if possible, for example
- use element.className instead of element.readAttribute('class')
- use element.id instead of element.readAttirbute('id')
- $$() selectors, especially attribute selectors
- may be expensive, measure first
- $$('#some .listing td a.popup[accesslink]'
- use getElementsByTagName() and iterate results instead
- element.style.* changes
- change class instead
- $() and getElementById on large (~20000 elements) pages
● Development
– Ruby code
– Rails code
● Production
– Live debugging
– Load balancing
● Frontend
– HTTP
– Javascript
Optimize Frontend: IE Project Management and Scrum Software
Unsurprisingly...
IE is also slow!
Slow things that are especially slow in IE:

- $() and $$(), even on small pages
- getElementsByName()
- style switching

Optimize Frontend: IE Project Management and Scrum Software
Good things about IE:
profiler in IE8
fast in IE => fast everywhere else!

Keep It Fast! Project Management and Scrum Software
So, you've optimized your application.

How to keep it fast?

Keep It Fast! Project Management and Scrum Software
Measure, measure and measure...

Use profiler
Optimize CPU and Memory

Keep It Fast: Measure Project Management and Scrum Software
Keep a set of benchmarks for most frequent user requests.

For example:
Benchmark Burndown 120 0.70 ± 0.00
Benchmark Inc. Burndown 120 0.92 ± 0.01
Benchmark Sprint 20 x (1+5) (C) 0.45 ± 0.00
Benchmark Issues 100 (C) 0.34 ± 0.00
Benchmark Prediction 120 0.56 ± 0.00
Benchmark Progress 120 0.23 ± 0.00
Benchmark Sprint 20 x (1+5) 0.93 ± 0.00
Benchmark Timeline 5x100 0.11 ± 0.00
Benchmark Signup 0.77 ± 0.00
Benchmark Export 0.20 ± 0.00
Benchmark Move Here 20/120 0.89 ± 0.00
Benchmark Order By User 0.98 ± 0.00
Benchmark Set Field (EP) 0.21 ± 0.00
Benchmark Task Create + Tag 0.23 ± 0.00
... 30 more ...

Benchmarks as a special kind of tests:

class RenderingTest < ActionController::IntegrationTest
def test_sprint_rendering
login_with users(:user), "user"
benchmark :title => "Sprint 20 x (1+5) (C)",

:route => "projects/1/sprints/3/show",
:assert_template => "tasks/index"
end
end
Benchmark Sprint 20 x (1+5) (C) 0.45 ± 0.00

Benchmarks as a special kind of tests:

def benchmark(options = {})
(0..100).each do |i|
GC.start
pid = fork do
begin
out = File.open("values", "a")
ActiveRecord::Base.transaction do
elapsed_time = Benchmark::realtime do
request_method = options[:post] ? :post : :get
send(request_method, options[:route])
end
out.puts elapsed_time if i > 0
out.close
raise CustomTransactionError
end
rescue CustomTransactionError
exit
end
end
Process::waitpid pid
ActiveRecord::Base.connection.reconnect!
end
values = File.read("values")
print "#{mean(values).to_02f} ± #{sigma(values).to_02f}\n"
end
Keep It Fast: Query Testing Project Management and Scrum Software
Losing 10ms in benchmark might seem OK
Except that it's sometimes not

because you're running one more SQL query

def test_queries
queries = track_queries do
get :index
end
assert_equal queries, [
"Foo Load",
"Bar Load",
"Event Create"
]
end

module ActiveSupport
class BufferedLogger
attr_reader :tracked_queries
def tracking=(val)
@tracked_queries = []
@tracking = val
end
def debug_with_tracking(message)
@tracked_queries << $1 if @tracking && message =~ /3[56]\;1m(.* (Load|Create|
Update|Destroy)) \(/
debug_without_tracking(message)
end
alias_method_chain :debug, :tracking
end
end
class ActiveSupport::TestCase
def track_queries(&block)
RAILS_DEFAULT_LOGGER.tracking = true
yield
result = RAILS_DEFAULT_LOGGER.tracked_queries
RAILS_DEFAULT_LOGGER.tracking = false
result
end
end

Keep It Fast: Use Profiler Project Management and Scrum Software
Profiler will always tell you what's wrong:

%self total self child calls name
8.39 0.54 0.23 0.31 602 Array#each_index
7.30 0.41 0.20 0.21 1227 Integer#gcd
6.20 0.49 0.17 0.32 5760 Timecell#date
5.11 0.15 0.14 0.01 1 Magick::Image#to_blob
gem install ruby-prof

use "performance" tests for Rails 2.x
KCachegrind to visualize the results
http://kcachegrind.sourceforge.net

Keep It Fast: Use Profiler Project Management and Scrum Software

Keep It Fast: Optimize CPU and Memory Project Management and Scrum Software
Memory profiler will explain the missing details:
Example benchmark: 5.52 sec request time, 1.07 sec GC time

Consumed memory: 55M
Ruby runs GC after allocating

8M memory or doing 10000 allocations
Simple math: 55 / 8 = 6 GC calls

Each GC call takes 0.18 sec!

Keep It Fast: Optimize CPU and Memory Project Management and Scrum Software
How to use memory profiler:
Recompile Ruby with GC patch

http://www.acunote.com/system/ruby186-p287-gc.patch
Reinstall ruby-prof
Use RUBY_PROF_MEASURE_MODE=memory
when running ruby-prof
http://blog.pluron.com/2008/02/memory-profilin.html
Remember! Project Management and Scrum Software
Measure, measure, measure... (with ruby-prof)

Optimize only what's slow
Optimize not only CPU, but memory
Optimize for user experience
Keep a set of performance regression tests
Monitor performance

Help Us Make Ruby and Rails Faster! Project Management and Scrum Software
We need your help!
● we'd like to see GC patch applied to mainstream Ruby

(to profile memory without recompiling Ruby)
● we'd like to see people porting their gems to Ruby 1.9
● we'd like to see more real-world benchmarks

Thanks!
Slides for this talk:
Rails performance articles and more:

http://blog.pluron.com

Advanced Performance Optimization of Rails Applications

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Advanced Performance Optimization of Rails Applications

Uploaded by

Copyright:

Available Formats

Project Management and Scrum Software

Advanced Performance Optimization

Online project management and Scrum software

250 Actually Happens: O(nc)

Best Case: O(log n)

Alexander Dymo • Advanced Performance Optimization of Rails Applications • RailsConf 2009 3 / 84

Throw Some Hardware at it!

Alexander Dymo • Advanced Performance Optimization of Rails Applications • RailsConf 2009 4 / 84

Alexander Dymo • Advanced Performance Optimization of Rails Applications • RailsConf 2009 5 / 84

Alexander Dymo • Advanced Performance Optimization of Rails Applications • RailsConf 2009 6 / 84

Alexander Dymo • Advanced Performance Optimization of Rails Applications • RailsConf 2009 7 / 84

Alexander Dymo • Advanced Performance Optimization of Rails Applications • RailsConf 2009 8 / 84

Alexander Dymo • Advanced Performance Optimization of Rails Applications • RailsConf 2009 9 / 84

Alexander Dymo • Advanced Performance Optimization of Rails Applications • RailsConf 2009 10 / 84

Alexander Dymo • Advanced Performance Optimization of Rails Applications • RailsConf 2009 11 / 84

2. Optimize only what's slow!

Alexander Dymo • Advanced Performance Optimization of Rails Applications • RailsConf 2009 12 / 84

3. Optimize for the user!

Alexander Dymo • Advanced Performance Optimization of Rails Applications • RailsConf 2009 13 / 84

What's wrong with Date?

16x slower than Time! Why?

Alexander Dymo • Advanced Performance Optimization of Rails Applications • RailsConf 2009 15 / 84

Fixing Date: Use C, Luke!

Date::Performance gem with Date partially rewritten in C

> require 'date/performance'

git clone git://github.com/rtomayko/date-performance.git

Real-world impact of Date::Performance:

Before: 0.95 sec

Alexander Dymo • Advanced Performance Optimization of Rails Applications • RailsConf 2009 17 / 84

Use String::<< instead of String::+=

Avoid BigDecimal comparisons with strings and integers

Alexander Dymo • Advanced Performance Optimization of Rails Applications • RailsConf 2009 18 / 84

Preloading associations is always a good idea...

foos = Foo.find_by_sql('select * from foos inner join bar')

Alexander Dymo • Advanced Performance Optimization of Rails Applications • RailsConf 2009 20 / 84

Virtual Attributes plugin:

Alexander Dymo • Advanced Performance Optimization of Rails Applications • RailsConf 2009 21 / 84

What can be wrong with this code?

class Task < ActiveRecord::Base

Kernel#binding is called to eval() the string callback

Alexander Dymo • Advanced Performance Optimization of Rails Applications • RailsConf 2009 22 / 84

Not too uncommon, right?

<% for object in objects %> #1000 times

We create 1000 View instances for each object here!

Alexander Dymo • Advanced Performance Optimization of Rails Applications • RailsConf 2009 23 / 84

Template inlining for the resque:

<% for object in objects %> #1000 times

Alexander Dymo • Advanced Performance Optimization of Rails Applications • RailsConf 2009 24 / 84

Template Inliner plugin:

Real world effect from template inlining:

Rendering of 300 objects, 5 partials for each object

How to optimize PostgreSQL:

Alexander Dymo • Advanced Performance Optimization of Rails Applications • RailsConf 2009 27 / 84

EXPLAIN ANALYZE explains everything, but...

Example: complex query which works on 230 000 rows and

Database server restart doesn't help

Alexander Dymo • Advanced Performance Optimization of Rails Applications • RailsConf 2009 28 / 84

Use any(array ()) instead of in()

Push down conditions into subselects and joins

Alexander Dymo • Advanced Performance Optimization of Rails Applications • RailsConf 2009 30 / 84

● Optimize For Development Box

Is that true in production?

Alexander Dymo • Advanced Performance Optimization of Rails Applications • RailsConf 2009 32 / 84

= Acunote Benchmarks = MRI JRuby 1.9.1