ElasticBeanStalk custom puma config for multiprocess x workers.

pumaconf.conf

container_commands:
  # directory '/var/app/current'
  # threads 8, 32
  # workers %x(grep -c processor /proc/cpuinfo)
  # bind 'unix:///var/run/puma/my_app.sock'
  # pidfile '/var/run/puma/puma.pid'
  # stdout_redirect '/var/log/puma/puma.log', '/var/log/puma/puma.log', true
  # daemonize false

  01backup_config:
    command: "cp -n /opt/elasticbeanstalk/support/conf/pumaconf.rb /opt/elasticbeanstalk/support/conf/pumaconf.rb.original"

  02_edit_comment_default_threads_config:
    # threads 8, 32 -> threads 4, 8
    command: "sed -i 's/threads 8, 32/threads 4, 8/' /opt/elasticbeanstalk/support/conf/pumaconf.rb"

  03_delete_default_worker_config:
    # workers %x(grep -c processor /proc/cpuinfo) -> workers 5 * %x( grep -c processor /proc/cpuinfo ).to_i
    command: "sed -i '/workers/d' /opt/elasticbeanstalk/support/conf/pumaconf.rb"
  04_insert_new_worker_after_threads_config:
    command: "sed -i '/threads/a workers 5 * %x( grep -c processor /proc/cpuinfo ).to_i' /opt/elasticbeanstalk/support/conf/pumaconf.rb"

Elasticbeanstalk puma default configuration which resides in /opt/elasticbeanstalk/support/conf/pumaconf.rb

directory '/var/app/current'
threads 8, 32
workers %x(grep -c processor /proc/cpuinfo)
bind 'unix:///var/run/puma/my_app.sock'
pidfile '/var/run/puma/puma.pid'
stdout_redirect '/var/log/puma/puma.log', '/var/log/puma/puma.log', true
daemonize false

The number of workers is set to equal to the number cpu cores. If the instance RAM is big this config does not seem good enough especially for Large instance type with MRI Ruby. As suggested from puma docs:

Rules of thumb In MRI:

• Use cluster mode and set the number of workers to 1.5x the number of CPU cores in machine, minimum 2.
• Set the number of threads to desired concurrent requests / number of workers. Puma defaults to 16 that is a decent number.

number_of_threads = wanted_number_of_concurrency/number_of_workers