jhawthorn / vernier

📏 next generation CRuby profiler
https://vernier.prof/
MIT License
835 stars 19 forks source link
cpu-profiler flamegraph memory-profiler profiler ruby

Vernier

Next-generation Ruby 3.2.1+ sampling profiler. Tracks multiple threads, GVL activity, GC pauses, idle time, and more.

Screenshot 2024-02-29 at 22 47 43

Examples

Livestreamed demo: Pairin' with Aaron (YouTube)

Sidekiq jobs from Mastodon (time, threaded) : https://share.firefox.dev/44jZRf3

Puma web requests from Mastodon (time, threaded) : https://share.firefox.dev/48FOTnF

Rails benchmark - lobste.rs (time) : https://share.firefox.dev/3Ld89id

require "irb" (retained memory) : https://share.firefox.dev/3DhLsFa

Installation

Vernier requires Ruby version 3.2.1 or greater.

gem "vernier", "~> 1.0"

Usage

The output can be viewed in the web app at https://vernier.prof, locally using the profile-viewer gem (both lightly customized versions of the firefox profiler frontend, which profiles are also compatible with) or by using the verrnier view command in the CLI.

Time and Allocations

Command line

The easiest way to record a program or script is via the CLI:

$ vernier run -- ruby -e 'sleep 1'
starting profiler with interval 500 and allocation interval 0
#<Vernier::Result 1.001589 seconds, 1 threads, 1 samples, 1 unique>
written to /tmp/profile20240328-82441-gkzffc.vernier.json.gz
$ vernier run --interval 100 --allocation-interval 10 -- ruby -e '10.times { Object.new }'
starting profiler with interval 100 and allocation interval 10
#<Vernier::Result 0.00067 seconds, 1 threads, 1 samples, 1 unique>
written to /tmp/profile20241029-26525-dalmym.vernier.json.gz

Block of code

Vernier.profile(out: "time_profile.json") do
  some_slow_method
end
Vernier.profile(out: "time_profile.json", interval: 100, allocation_interval: 10) do
  some_slow_method
end

Alternatively you can use the aliases Vernier.run and Vernier.trace.

Start and stop

Vernier.start_profile(out: "time_profile.json", interval: 10_000, allocation_interval: 100_000)

some_slow_method

# some other file

some_other_slow_method

Vernier.stop_profile

Retained memory

Block of code

Record a flamegraph of all retained allocations from loading irb:

ruby -r vernier -e 'Vernier.trace_retained(out: "irb_profile.json") { require "irb" }'

[!NOTE] Retained-memory flamegraphs must be interpreted a little differently than a typical profiling flamegraph. In a retained-memory flamegraph, the x-axis represents a proportion of memory in bytes, not time or samples The topmost boxes on the y-axis represent the retained objects, with their stacktrace below; their width represents the percentage of overall retained memory each object occupies.

Options

Option Description
mode The sampling mode to use. One of :wall, :retained or :custom. Default is :wall.
out The file to write the profile to.
interval The sampling interval in microseconds. Default is 500. Only available in :wall mode.
allocation_interval The allocation sampling interval in number of allocations. Default is 0 (disabled). Only available in :wall mode.
gc Initiate a full and immediate garbage collection cycle before profiling. Default is true. Only available in :retained mode.

Development

After checking out the repo, run bin/setup to install dependencies. Then, run rake test to run the tests. You can also run bin/console for an interactive prompt that will allow you to experiment.

To install this gem onto your local machine, run bundle exec rake install. To release a new version, update the version number in version.rb, and then run bundle exec rake release, which will create a git tag for the version, push git commits and the created tag, and push the .gem file to rubygems.org.

Contributing

Bug reports and pull requests are welcome on GitHub at https://github.com/jhawthorn/vernier. This project is intended to be a safe, welcoming space for collaboration, and contributors are expected to adhere to the code of conduct.

Resources

License

The gem is available as open source under the terms of the MIT License.

Code of Conduct

Everyone interacting in the Vernier project's codebases, issue trackers, chat rooms and mailing lists is expected to follow the code of conduct.