ruby-kafka-ec2 0.1.0

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA256:
+  metadata.gz: a216c0064e93662929aa08a27aee3caad6eb7d4a7eb02a658b6edef1ced9fa33
+  data.tar.gz: 5982cfb402ea097fbc45580b3bf17c88d1ef6fef4c9f3fc3c06b71c0ba9fd8f7
+SHA512:
+  metadata.gz: 47ef0c231763ba3b9c8ee95417eba748a205e8791bd973bdebdb7e4b30aa103f32c965fbbc47cc8aa157f6ae80bd8e4f141ac153264eb3514accb569a375f8d5
+  data.tar.gz: '009ad55259d086a25252b8fd0aa6f9f3eecdbf0c25b79ac00c07a920b28fe7d4d1758043e7ef804364e56c869bd57e51b51690b44c16745fba45dc12ec2bdb40'

data/.github/workflows/test.yml

@@ -0,0 +1,22 @@
+name: CI
+
+on:
+  push:
+  pull_request:
+
+jobs:
+  build:
+
+    runs-on: ubuntu-latest
+
+    steps:
+    - uses: actions/checkout@v2
+    - name: Set up Ruby 2.6
+      uses: actions/setup-ruby@v1
+      with:
+        ruby-version: 2.6.x
+    - name: Build and test with Rake
+      run: |
+        gem install bundler
+        bundle install --jobs 4 --retry 3
+        bundle exec rake

data/.gitignore

@@ -0,0 +1,11 @@
+/.bundle/
+/.yardoc
+/_yardoc/
+/coverage/
+/doc/
+/pkg/
+/spec/reports/
+/tmp/
+
+# rspec failure tracking
+.rspec_status

data/.rspec

@@ -0,0 +1,3 @@
+--format documentation
+--color
+--require spec_helper

data/Gemfile

@@ -0,0 +1,7 @@
+source "https://rubygems.org"
+
+# Specify your gem's dependencies in ruby-kafka-ec2.gemspec
+gemspec
+
+gem "rake", "~> 12.0"
+gem "rspec", "~> 3.0"

data/LICENSE.txt

@@ -0,0 +1,21 @@
+The MIT License (MIT)
+
+Copyright (c) 2020 abicky
+
+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.

data/README.md

@@ -0,0 +1,93 @@
+# ruby-kafka-ec2
+
+![](https://github.com/abicky/ruby-kafka-ec2/workflows/CI/badge.svg?branch=master)
+
+ruby-kafka-ec2 is an extension of ruby-kafka that provides useful features for EC2 like Kafka::EC2::MixedInstanceAssignmentStrategy.
+
+## Installation
+
+Add this line to your application's Gemfile:
+
+```ruby
+gem 'ruby-kafka-ec2'
+```
+
+And then execute:
+
+    $ bundle install
+
+Or install it yourself as:
+
+    $ gem install ruby-kafka-ec2
+
+## Usage
+
+### Kafka::EC2::MixedInstanceAssignmentStrategy
+
+`Kafka::EC2::MixedInstanceAssignmentStrategy` is an assignor for auto-scaling groups with mixed instance policies. The throughputs of consumers usually depend on instance families and availability zones. For example, if your application writes data to a database, the throughputs of consumers running on the same availability zone as the writer DB instance is higher.
+
+To assign more partitions to consumers with high throughputs, you have to define `Kafka::EC2::MixedInstanceAssignmentStrategyFactory` first like below:
+
+```ruby
+require "aws-sdk-rds"
+require "kafka"
+require "kafka/ec2"
+
+rds = Aws::RDS::Client.new(region: "ap-northeast-1")
+assignment_strategy_factory = Kafka::EC2::MixedInstanceAssignmentStrategyFactory.new(
+  instance_family_weights: {
+    "r4" => 1.00,
+    "r5" => 1.20,
+    "m5" => 1.35,
+    "c5" => 1.50,
+  },
+  availability_zone_weights: ->() {
+    db_cluster = rds.describe_db_clusters(filters: [
+      { name: "db-cluster-id", values: [ENV["RDS_CLUSTER"]] },
+    ]).db_clusters.first
+    db_instance_id = db_cluster.db_cluster_members.find { |m| m.is_cluster_writer }.db_instance_identifier
+    db_instance = rds.describe_db_instances(filters: [
+      { name: "db-cluster-id", values: [ENV["RDS_CLUSTER"]] },
+      { name: "db-instance-id", values: [db_instance_id] },
+    ]).db_instances.first
+
+    if db_instance.availability_zone == "ap-northeast-1a"
+      {
+        "ap-northeast-1a" => 1,
+        "ap-northeast-1c" => 0.25,
+      }
+    else
+      {
+        "ap-northeast-1a" => 0.25,
+        "ap-northeast-1c" => 1,
+      }
+    end
+  },
+)
+```
+
+In the preceding example, consumers running on c5 instances will have 1.5x as many partitions compared to consumers running on r4 instances. In a similar way, if the writer DB instance is in ap-northeast-1a, consumers in ap-northeast-1a will have 4x as many partitions compared to consumers in ap-northeast-1c.
+
+You can use `Kafka::EC2::MixedInstanceAssignmentStrategy` by specifying the factory to `Kafka::EC2.with_assignment_strategy_factory` and creating a consumer in the block:
+
+
+```ruby
+consumer = Kafka::EC2.with_assignment_strategy_factory(assignment_strategy_factory) do
+  kafka.consumer(group_id: ENV["KAFKA_CONSUMER_GROUP_ID"])
+end
+```
+
+## Development
+
+After checking out the repo, run `bin/setup` to install dependencies. Then, run `rake spec` 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 tags, and push the `.gem` file to [rubygems.org](https://rubygems.org).
+
+## Contributing
+
+Bug reports and pull requests are welcome on GitHub at https://github.com/abicky/ruby-kafka-ec2.
+
+
+## License
+
+The gem is available as open source under the terms of the [MIT License](https://opensource.org/licenses/MIT).

data/Rakefile

@@ -0,0 +1,6 @@
+require "bundler/gem_tasks"
+require "rspec/core/rake_task"
+
+RSpec::Core::RakeTask.new(:spec)
+
+task :default => :spec

data/benchmark/.gitignore

@@ -0,0 +1,4 @@
+/.terraform
+/terraform.tfstate
+/terraform.tfstate.backup
+/terraform.tfvars

data/benchmark/Dockerfile

@@ -0,0 +1,13 @@
+FROM ruby:2.7
+
+WORKDIR /work
+
+COPY Gemfile ruby-kafka-ec2.gemspec /tmp/ruby-kafka-ec2/
+COPY lib /tmp/ruby-kafka-ec2/lib/
+
+COPY benchmark/Gemfile /work/
+RUN bundle install -j4
+
+COPY benchmark/produce_messages.rb benchmark/consume_messages.rb /work/
+
+ENTRYPOINT ["bundle", "exec", "ruby"]

data/benchmark/Dockerfile.dockerignore

@@ -0,0 +1 @@
+.terraform

data/benchmark/Gemfile

@@ -0,0 +1,11 @@
+# frozen_string_literal: true
+
+source "https://rubygems.org"
+
+git_source(:github) {|repo_name| "https://github.com/#{repo_name}" }
+
+gem "ruby-kafka"
+gem "concurrent-ruby" # cf. https://github.com/zendesk/ruby-kafka/pull/835
+gem "mysql2"
+gem "ruby-kafka-ec2", path: "/tmp/ruby-kafka-ec2"
+gem "aws-sdk-rds"

data/benchmark/README.md

@@ -0,0 +1,99 @@
+# Benchmark
+
+## Procedures
+
+### 1. Create AWS resources
+
+```
+terraform init
+terraform apply
+```
+
+### 2. Register docker image
+
+```
+./register_docker_image.sh
+```
+
+### 3. Produce messages and create database records
+
+Set the environment variables `KAFKA_BROKERS`, `KAFKA_TOPIC`, `MYSQL_HOST`, `MYSQL_PASSWORD`, and `CLUSTER`, and execute the following command:
+
+```
+./produce_messages.sh
+```
+
+### 4. Consume messages
+
+Set the environment variables `KAFKA_BROKERS`, `KAFKA_TOPIC`, `MYSQL_HOST`, `MYSQL_PASSWORD`, and `CLUSTER`, and execute the following command:
+
+```
+USE_KAFKA_EC2=false ./consume_messages.sh
+```
+
+Stop all the tasks if all lags become 0. You can check the lags by executing the following command in the kafka client instance:
+
+```
+./kafka-consumer-groups.sh \
+  --bootstrap-server <bootstrap-server> \
+  --describe \
+  --group net.abicky.ruby-kafka-ec2.benchmark
+```
+
+Reset the offsets in the kafka client instance:
+
+```
+./kafka-consumer-groups.sh \
+  --bootstrap-server <bootstrap-server> \
+  --group net.abicky.ruby-kafka-ec2.benchmark \
+  --reset-offsets \
+  --to-earliest \
+  --topic <topic> \
+  --execute
+```
+
+Set the environment variables `KAFKA_BROKERS`, `KAFKA_TOPIC`, `MYSQL_HOST`, `MYSQL_PASSWORD`, and `CLUSTER`, and execute the following command:
+
+```
+USE_KAFKA_EC2=true ./consume_messages.sh
+```
+
+## Result
+
+### USE_KAFKA_EC2=false
+
+No. | instance type | availability zone | partition count | processed count | duration (sec)
+----|---------------|-------------------|-----------------|-----------------|-----------
+1 | m5.large | ap-northeast-1a | 16 | 80000 | 224.1
+2 | m5.large | ap-northeast-1a | 17 | 85000 | 237.7
+3 | m5.large | ap-northeast-1c | 17 | 85000 | 56.8
+4 | m5.large | ap-northeast-1c | 17 | 85000 | 57.3
+5 | r4.large | ap-northeast-1a | 16 | 80000 | 257.6
+6 | r4.large | ap-northeast-1a | 17 | 85000 | 258.9
+7 | r4.large | ap-northeast-1c | 16 | 80000 | 74.2
+8 | r4.large | ap-northeast-1c | 17 | 85000 | 76.6
+9 | r5.large | ap-northeast-1a | 17 | 80000 | 238.0
+10 | r5.large | ap-northeast-1a | 17 | 85000 | 238.1
+11 | r5.large | ap-northeast-1c | 16 | 80000 | 54.1
+12 | r5.large | ap-northeast-1c | 16 | 85000 | 55.1
+
+See ruby-kafka.log for more details.
+
+### USE_KAFKA_EC2=true
+
+No. | instance type | availability zone | partition count | processed count | duration (sec)
+----|---------------|-------------------|-----------------|-----------------|-----------
+1 | m5.large | ap-northeast-1a | 7 | 35000 | 101.4
+2 | m5.large | ap-northeast-1a | 8 | 40000 | 114.4
+3 | m5.large | ap-northeast-1c | 31 | 155000 | 101.6
+4 | m5.large | ap-northeast-1c | 30 | 150000 | 98.6
+5 | r4.large | ap-northeast-1a | 5 | 25000 | 79.0
+6 | r4.large | ap-northeast-1a | 6 | 30000 | 92.4
+7 | r4.large | ap-northeast-1c | 23 | 115000 | 105.9
+8 | r4.large | ap-northeast-1c | 22 | 110000 | 106.1
+9 | r5.large | ap-northeast-1a | 7 | 35000 | 102.1
+10 | r5.large | ap-northeast-1a | 7 | 35000 | 102.1
+11 | r5.large | ap-northeast-1c | 27 | 135000 | 94.2
+12 | r5.large | ap-northeast-1c | 27 | 135000 | 95.7
+
+See ruby-kafka-ec2.log for more details.

data/benchmark/aws.tf

@@ -0,0 +1,135 @@
+provider "aws" {
+  region = "ap-northeast-1"
+}
+
+variable "vpc_id" {}
+variable "ec2_instance_security_group_id" {}
+variable "ec2_key_name" {}
+variable "rds_master_password" {}
+variable "kafka_broker_subnet_ids" {}
+
+# RDS
+
+resource "aws_security_group" "rds" {
+  name        = "ruby-kafka-ec2-rds"
+  vpc_id      = var.vpc_id
+
+  egress {
+    from_port   = 0
+    to_port     = 0
+    protocol    = "-1"
+    cidr_blocks = ["0.0.0.0/0"]
+  }
+}
+
+resource "aws_security_group_rule" "rds_access_from_ec2" {
+  security_group_id        = aws_security_group.rds.id
+  type                     = "ingress"
+  from_port                = 3306
+  to_port                  = 3306
+  protocol                 = "tcp"
+  source_security_group_id = var.ec2_instance_security_group_id
+}
+
+
+resource "aws_rds_cluster" "benchmark" {
+  cluster_identifier              = "ruby-kafka-ec2-benchmark"
+  engine                          = "aurora"
+  engine_mode                     = "provisioned"
+  engine_version                  = "5.6.10a"
+  availability_zones              = ["ap-northeast-1a", "ap-northeast-1c"]
+  master_username                 = "admin"
+  master_password                 = var.rds_master_password
+  backup_retention_period         = 1
+  skip_final_snapshot             = true
+  vpc_security_group_ids          = [aws_security_group.rds.id]
+}
+
+resource "aws_rds_cluster_instance" "instances" {
+  count                   = 1
+  identifier              = "${aws_rds_cluster.benchmark.cluster_identifier}-${count.index}"
+  cluster_identifier      = aws_rds_cluster.benchmark.id
+  instance_class          = "db.r5.large"
+}
+
+# Kafka
+
+data "aws_kms_key" "kafka" {
+  key_id = "alias/aws/kafka"
+}
+
+resource "aws_security_group" "kafka_cluster" {
+  name   = "ruby-kafka-ec2-kafka-cluster"
+  vpc_id = var.vpc_id
+
+  egress {
+    from_port   = 0
+    to_port     = 0
+    protocol    = "-1"
+    cidr_blocks = ["0.0.0.0/0"]
+  }
+}
+
+resource "aws_security_group_rule" "kafka_cluster_from_ec2" {
+  security_group_id = aws_security_group.kafka_cluster.id
+  type              = "ingress"
+  from_port         = 0
+  to_port           = 0
+  protocol          = "-1"
+  source_security_group_id = var.ec2_instance_security_group_id
+}
+
+resource "aws_msk_cluster" "benchmark" {
+  cluster_name           = "ruby-kafka-ec2-benchmark"
+  kafka_version          = "2.2.1"
+  number_of_broker_nodes = 3
+
+  broker_node_group_info {
+    instance_type   = "kafka.m5.large"
+    ebs_volume_size = "100"
+    client_subnets = var.kafka_broker_subnet_ids
+    security_groups = [aws_security_group.kafka_cluster.id]
+  }
+
+  encryption_info {
+    encryption_at_rest_kms_key_arn = data.aws_kms_key.kafka.arn
+
+    encryption_in_transit {
+      client_broker = "TLS_PLAINTEXT"
+      in_cluster    = true
+    }
+  }
+}
+
+data "aws_ami" "most_recent_amazon_linux2" {
+  most_recent = true
+  owners      = ["amazon"]
+
+  filter {
+    name   = "name"
+    values = ["amzn2-ami-hvm-2.0.*-x86_64-gp2"]
+  }
+}
+
+resource "aws_instance" "kafka_client" {
+  ami           = data.aws_ami.most_recent_amazon_linux2.image_id
+  instance_type = "t3.nano"
+  key_name      = var.ec2_key_name
+
+  subnet_id              = var.kafka_broker_subnet_ids[0]
+  vpc_security_group_ids = [var.ec2_instance_security_group_id]
+
+  associate_public_ip_address = true
+
+  # cf. https://docs.aws.amazon.com/msk/latest/developerguide/create-client-machine.html
+  user_data = <<EOF
+#!/bin/bash
+yum install -y java-1.8.0
+wget https://archive.apache.org/dist/kafka/2.2.1/kafka_2.12-2.2.1.tgz
+tar -xzf kafka_2.12-2.2.1.tgz
+EOF
+
+  tags = {
+    Name = "kafka-client"
+  }
+}

data/benchmark/consume_messages.rb

@@ -0,0 +1,106 @@
+require "json"
+require "net/http"
+require "time"
+
+require "kafka"
+require "concurrent/map" # cf. https://github.com/zendesk/ruby-kafka/pull/835
+require "mysql2"
+
+KAFKA_BROKERS = ENV.fetch("KAFKA_BROKERS", "localhost:9092").split(/\p{Space}*,\p{Space}*/)
+KAFKA_CLIENT_ID = "net.abicky.ruby-kafka-ec2"
+KAFKA_CONSUMER_GROUP_ID = "net.abicky.ruby-kafka-ec2.benchmark"
+KAFKA_TOPIC = ENV.fetch("KAFKA_TOPIC") do
+  raise 'The environment variable "KAFKA_TOPIC" must be specified'
+end
+
+$stdout.sync = true
+logger = Logger.new($stdout)
+
+kafka = Kafka.new(KAFKA_BROKERS, client_id: KAFKA_CLIENT_ID)
+if ENV["USE_KAFKA_EC2"] == "true"
+  logger.info "Use ruby-kafka-ec2"
+  require "aws-sdk-rds"
+  require "kafka/ec2"
+
+  rds = Aws::RDS::Client.new(region: "ap-northeast-1")
+  assignment_strategy_factory = Kafka::EC2::MixedInstanceAssignmentStrategyFactory.new(
+    instance_family_weights: {
+      "r4" => 1.00,
+      "r5" => 1.20,
+      "m5" => 1.35,
+    },
+    availability_zone_weights: ->() {
+      db_cluster = rds.describe_db_clusters(filters: [
+        { name: "db-cluster-id", values: ["ruby-kafka-ec2-benchmark"] },
+      ]).db_clusters.first
+      db_instance_id = db_cluster.db_cluster_members.find { |m| m.is_cluster_writer }.db_instance_identifier
+      db_instance = rds.describe_db_instances(filters: [
+        { name: "db-cluster-id", values: ["ruby-kafka-ec2-benchmark"] },
+        { name: "db-instance-id", values: [db_instance_id] },
+      ]).db_instances.first
+
+      if db_instance.availability_zone == "ap-northeast-1a"
+        {
+          "ap-northeast-1a" => 1,
+          "ap-northeast-1c" => 0.25,
+        }
+      else
+        {
+          "ap-northeast-1a" => 0.25,
+          "ap-northeast-1c" => 1,
+        }
+      end
+    },
+  )
+  consumer = Kafka::EC2.with_assignment_strategy_factory(assignment_strategy_factory) do
+    kafka.consumer(group_id: KAFKA_CONSUMER_GROUP_ID)
+  end
+else
+  logger.info "Use ruby-kafka"
+  consumer = kafka.consumer(group_id: KAFKA_CONSUMER_GROUP_ID)
+end
+
+consumer.subscribe(KAFKA_TOPIC)
+
+trap(:TERM) { consumer.stop }
+
+metadata = Net::HTTP.start("169.254.169.254", 80) do |http|
+  {
+    instance_id: http.get("/latest/meta-data/instance-id").body,
+    instance_type: http.get("/latest/meta-data/instance-type").body,
+    availability_zone: http.get("/latest/meta-data/placement/availability-zone").body,
+  }
+end
+
+client = Mysql2::Client.new(
+  host: ENV["MYSQL_HOST"] || "localhost",
+  port: 3306,
+  username: "admin",
+  password: ENV["MYSQL_PASSWORD"],
+)
+
+logger.info "[#{metadata}] Consuming messages..."
+
+start_time = Process.clock_gettime(Process::CLOCK_MONOTONIC)
+processed_count = 0
+partition_count = 0
+end_time = nil
+consumer.each_message do |message|
+  if message.value == "FIN"
+    end_time = Process.clock_gettime(Process::CLOCK_MONOTONIC)
+    partition_count += 1
+    logger.info "[#{metadata}] Received FIN message"
+    next
+  end
+
+  JSON.parse(message.value)["events"].each do |event|
+    Time.iso8601(event["timestamp"])
+  end
+  client.query("SELECT * FROM ruby_kafka_ec2_benchmark.events").to_a
+
+  processed_count += 1
+  logger.info "[#{metadata}] #{processed_count} messages were consumed" if (processed_count % 10_000).zero?
+end
+
+duration = end_time - start_time
+logger.info "[#{metadata}] Complete (duration: #{duration}, partition_count: #{partition_count}, processed_count: #{processed_count})"

data/benchmark/consume_messages.sh

@@ -0,0 +1,47 @@
+#!/bin/bash
+
+set -eo pipefail
+
+REPOSITORY=ruby-kafka-ec2/benchmark
+TASK_FAMILY=ruby-kafka-ec2-benchmark-consumer
+
+repo_uri=$((aws ecr describe-repositories --repository-names $REPOSITORY | jq -r '.repositories[] | .repositoryUri') || true)
+if [ -z "$repo_uri" ]; then
+  echo "The repository '$REPOSITORY' is not found. Execute register_docker_image.sh first, please." >&2
+  exit
+fi
+
+aws ecs register-task-definition --cli-input-json "$(cat <<JSON
+{
+  "family": "$TASK_FAMILY",
+  "containerDefinitions": [
+    {
+      "name": "consume_messages",
+      "image": "$repo_uri",
+      "command": ["consume_messages.rb"],
+      "essential": true,
+      "environment": [
+        {"name": "KAFKA_BROKERS", "value": "$KAFKA_BROKERS"},
+        {"name": "KAFKA_TOPIC", "value": "$KAFKA_TOPIC"},
+        {"name": "MYSQL_HOST", "value": "$MYSQL_HOST"},
+        {"name": "MYSQL_PASSWORD", "value": "$MYSQL_PASSWORD"},
+        {"name": "USE_KAFKA_EC2", "value": "$USE_KAFKA_EC2"}
+      ],
+      "logConfiguration": {
+        "logDriver": "awslogs",
+        "options": {
+          "awslogs-group": "/aws/ecs/ruby-kafka-ec2/benchmark",
+          "awslogs-region": "ap-northeast-1",
+          "awslogs-stream-prefix": "consume-messages"
+       }
+      }
+    }
+  ],
+  "cpu": "1024",
+  "memory": "2048"
+}
+JSON
+)"
+
+aws ecs run-task --cluster $CLUSTER --task-definition $TASK_FAMILY --count 10
+aws ecs run-task --cluster $CLUSTER --task-definition $TASK_FAMILY --count 2

data/benchmark/produce_messages.rb

@@ -0,0 +1,96 @@
+require "json"
+require "time"
+
+require "concurrent"
+require "kafka"
+require "mysql2"
+
+KAFKA_BROKERS = ENV.fetch("KAFKA_BROKERS", "localhost:9092").split(/\p{Space}*,\p{Space}*/)
+KAFKA_CLIENT_ID = "net.abicky.ruby-kafka-ec2"
+KAFKA_TOPIC = ENV.fetch("KAFKA_TOPIC") do
+  raise 'The environment variable "KAFKA_TOPIC" must be specified'
+end
+PARTITION_COUNT = 200
+MAX_BUFFER_SIZE = 1_000
+MESSAGE_COUNT = 1_000_000
+
+$stdout.sync = true
+logger = Logger.new($stdout)
+
+client = Mysql2::Client.new(
+  host: ENV["MYSQL_HOST"] || "localhost",
+  port: 3306,
+  username: "admin",
+  password: ENV["MYSQL_PASSWORD"],
+)
+client.query("CREATE DATABASE IF NOT EXISTS ruby_kafka_ec2_benchmark")
+client.query(<<~SQL)
+  CREATE TABLE IF NOT EXISTS ruby_kafka_ec2_benchmark.events (
+    id bigint(20) NOT NULL AUTO_INCREMENT,
+    name varchar(255) NOT NULL,
+    created_at datetime NOT NULL,
+    PRIMARY KEY (id)
+  )
+SQL
+client.query(<<~SQL)
+  INSERT INTO ruby_kafka_ec2_benchmark.events (name, created_at) VALUES ('page_view', '#{Time.now.strftime("%F %T")}')
+SQL
+
+kafka = Kafka.new(KAFKA_BROKERS, client_id: KAFKA_CLIENT_ID)
+
+unless kafka.topics.include?(KAFKA_TOPIC)
+  logger.info "Create the kafka topic '#{KAFKA_TOPIC}'"
+  kafka.create_topic(KAFKA_TOPIC, num_partitions: PARTITION_COUNT, replication_factor: 3)
+end
+
+now = Time.now.iso8601(3)
+message = {
+  events: [{ name: "page_view", timestamp: now }] * 10,
+}.to_json
+
+
+logger.info "Producing messages..."
+
+start_time = Process.clock_gettime(Process::CLOCK_MONOTONIC)
+pool = Concurrent::FixedThreadPool.new(4)
+producers = {}
+current_processed_count = Concurrent::AtomicFixnum.new
+futures = []
+MESSAGE_COUNT.times do |i|
+  futures << Concurrent::Future.execute(executor: pool) do
+    producers[Thread.current.object_id] ||= kafka.producer(max_buffer_size: MAX_BUFFER_SIZE)
+    producer = producers[Thread.current.object_id]
+    producer.produce(message,
+      topic: KAFKA_TOPIC,
+      key: i.to_s,
+      partition: i % PARTITION_COUNT,
+    )
+    if producer.buffer_size == MAX_BUFFER_SIZE
+      producer.deliver_messages
+    end
+    processed_count = current_processed_count.increment
+    logger.info "#{processed_count} messages were produced" if (processed_count % 10_000).zero?
+  end
+
+  if futures.size == 10_000
+    futures.each(&:wait!)
+    futures.clear
+  end
+end
+futures.each(&:wait!)
+
+producers.each_value(&:deliver_messages)
+
+logger.info "Produce FIN messages"
+producer = kafka.producer
+PARTITION_COUNT.times do |i|
+  producer.produce("FIN",
+    topic: KAFKA_TOPIC,
+    key: "fin_#{i}",
+    partition: i % PARTITION_COUNT,
+  )
+end
+producer.deliver_messages
+
+duration = Process.clock_gettime(Process::CLOCK_MONOTONIC) - start_time
+logger.info "Complete (duration: #{duration})"

data/benchmark/produce_messages.sh

@@ -0,0 +1,45 @@
+#!/bin/bash
+
+set -eo pipefail
+
+REPOSITORY=ruby-kafka-ec2/benchmark
+TASK_FAMILY=ruby-kafka-ec2-benchmark-producer
+
+repo_uri=$((aws ecr describe-repositories --repository-names $REPOSITORY | jq -r '.repositories[] | .repositoryUri') || true)
+if [ -z "$repo_uri" ]; then
+  echo "The repository '$REPOSITORY' is not found. Execute register_docker_image.sh first, please." >&2
+  exit
+fi
+
+aws ecs register-task-definition --cli-input-json "$(cat <<JSON
+{
+  "family": "$TASK_FAMILY",
+  "containerDefinitions": [
+    {
+      "name": "produce_messages",
+      "image": "$repo_uri",
+      "command": ["produce_messages.rb"],
+      "essential": true,
+      "environment": [
+        {"name": "KAFKA_BROKERS", "value": "$KAFKA_BROKERS"},
+        {"name": "KAFKA_TOPIC", "value": "$KAFKA_TOPIC"},
+        {"name": "MYSQL_HOST", "value": "$MYSQL_HOST"},
+        {"name": "MYSQL_PASSWORD", "value": "$MYSQL_PASSWORD"}
+      ],
+      "logConfiguration": {
+        "logDriver": "awslogs",
+        "options": {
+          "awslogs-group": "/aws/ecs/ruby-kafka-ec2/benchmark",
+          "awslogs-region": "ap-northeast-1",
+          "awslogs-stream-prefix": "produce-messages"
+        }
+      }
+    }
+  ],
+  "cpu": "2048",
+  "memory": "4096"
+}
+JSON
+)"
+
+aws ecs run-task --cluster $CLUSTER --task-definition $TASK_FAMILY

data/benchmark/register_docker_image.sh

@@ -0,0 +1,20 @@
+#!/bin/bash
+
+set -eo pipefail
+
+REPOSITORY=ruby-kafka-ec2/benchmark
+
+cd $(dirname $(cd $(dirname $0) && pwd))
+repo_uri=$((aws ecr describe-repositories --repository-names $REPOSITORY | jq -r '.repositories[] | .repositoryUri') || true)
+if [ -z "$repo_uri" ]; then
+  echo "Create ECR repository '$REPOSITORY'"
+  repo_uri=$(aws ecr create-repository --repository-name $REPOSITORY | jq -r '.repository.repositoryUri')
+fi
+
+echo "Build docker image"
+DOCKER_BUILDKIT=1 docker build -t $REPOSITORY . -f benchmark/Dockerfile
+
+echo "Push $repo_uri:latest"
+aws ecr get-login-password | docker login --username AWS --password-stdin $repo_uri
+docker tag $REPOSITORY:latest $repo_uri:latest
+docker push $repo_uri:latest

data/benchmark/ruby-kafka-ec2.log

@@ -0,0 +1,12 @@
+I, [2020-06-21T21:02:27.530399 #1]  INFO -- : [{:instance_id=>"i-03d54244a9ca1b766", :instance_type=>"r5.large", :availability_zone=>"ap-northeast-1c"}] Complete (duration: 94.20265760399889, partition_count: 27, processed_count: 135000)
+I, [2020-06-21T21:02:28.532064 #1]  INFO -- : [{:instance_id=>"i-0277b51635cc60328", :instance_type=>"r4.large", :availability_zone=>"ap-northeast-1c"}] Complete (duration: 106.07931412400012, partition_count: 22, processed_count: 110000)
+I, [2020-06-21T21:02:28.787485 #1]  INFO -- : [{:instance_id=>"i-0aae2c827f2e13c54", :instance_type=>"m5.large", :availability_zone=>"ap-northeast-1c"}] Complete (duration: 98.62195983100173, partition_count: 30, processed_count: 150000)
+I, [2020-06-21T21:02:28.838145 #1]  INFO -- : [{:instance_id=>"i-03d54244a9ca1b766", :instance_type=>"r5.large", :availability_zone=>"ap-northeast-1c"}] Complete (duration: 95.73462492099952, partition_count: 27, processed_count: 135000)
+I, [2020-06-21T21:02:29.113526 #1]  INFO -- : [{:instance_id=>"i-042ca61bd40d8692c", :instance_type=>"r4.large", :availability_zone=>"ap-northeast-1a"}] Complete (duration: 79.04582024299998, partition_count: 5, processed_count: 25000)
+I, [2020-06-21T21:02:29.150897 #1]  INFO -- : [{:instance_id=>"i-0c0b6259f2e5ec3d8", :instance_type=>"r5.large", :availability_zone=>"ap-northeast-1a"}] Complete (duration: 102.10359391399925, partition_count: 7, processed_count: 35000)
+I, [2020-06-21T21:02:29.181021 #1]  INFO -- : [{:instance_id=>"i-0386bf32a565e5bff", :instance_type=>"m5.large", :availability_zone=>"ap-northeast-1a"}] Complete (duration: 101.41423055399991, partition_count: 7, processed_count: 35000)
+I, [2020-06-21T21:02:29.250973 #1]  INFO -- : [{:instance_id=>"i-0aae2c827f2e13c54", :instance_type=>"m5.large", :availability_zone=>"ap-northeast-1c"}] Complete (duration: 101.59283073200277, partition_count: 31, processed_count: 155000)
+I, [2020-06-21T21:02:29.257868 #1]  INFO -- : [{:instance_id=>"i-042ca61bd40d8692c", :instance_type=>"r4.large", :availability_zone=>"ap-northeast-1a"}] Complete (duration: 92.3937842589985, partition_count: 6, processed_count: 30000)
+I, [2020-06-21T21:02:29.255276 #1]  INFO -- : [{:instance_id=>"i-0c0b6259f2e5ec3d8", :instance_type=>"r5.large", :availability_zone=>"ap-northeast-1a"}] Complete (duration: 102.14144639099959, partition_count: 7, processed_count: 35000)
+I, [2020-06-21T21:02:29.381202 #1]  INFO -- : [{:instance_id=>"i-0277b51635cc60328", :instance_type=>"r4.large", :availability_zone=>"ap-northeast-1c"}] Complete (duration: 105.87376681400019, partition_count: 23, processed_count: 115000)
+I, [2020-06-21T21:02:29.386796 #1]  INFO -- : [{:instance_id=>"i-0386bf32a565e5bff", :instance_type=>"m5.large", :availability_zone=>"ap-northeast-1a"}] Complete (duration: 114.40461669500019, partition_count: 8, processed_count: 40000)

data/benchmark/ruby-kafka.log

@@ -0,0 +1,12 @@
+I, [2020-06-21T20:46:00.436023 #1]  INFO -- : [{:instance_id=>"i-0aae2c827f2e13c54", :instance_type=>"m5.large", :availability_zone=>"ap-northeast-1c"}] Complete (duration: 57.33392923200154, partition_count: 17, processed_count: 85000)
+I, [2020-06-21T20:46:00.539574 #1]  INFO -- : [{:instance_id=>"i-0c0b6259f2e5ec3d8", :instance_type=>"r5.large", :availability_zone=>"ap-northeast-1a"}] Complete (duration: 238.05382897799973, partition_count: 17, processed_count: 85000)
+I, [2020-06-21T20:46:00.607154 #1]  INFO -- : [{:instance_id=>"i-0c0b6259f2e5ec3d8", :instance_type=>"r5.large", :availability_zone=>"ap-northeast-1a"}] Complete (duration: 238.03822113800015, partition_count: 17, processed_count: 85000)
+I, [2020-06-21T20:46:00.988614 #1]  INFO -- : [{:instance_id=>"i-0aae2c827f2e13c54", :instance_type=>"m5.large", :availability_zone=>"ap-northeast-1c"}] Complete (duration: 56.78356462699958, partition_count: 17, processed_count: 85000)
+I, [2020-06-21T20:46:01.308462 #1]  INFO -- : [{:instance_id=>"i-0386bf32a565e5bff", :instance_type=>"m5.large", :availability_zone=>"ap-northeast-1a"}] Complete (duration: 224.05910681800015, partition_count: 16, processed_count: 80000)
+I, [2020-06-21T20:46:01.377471 #1]  INFO -- : [{:instance_id=>"i-0386bf32a565e5bff", :instance_type=>"m5.large", :availability_zone=>"ap-northeast-1a"}] Complete (duration: 237.7210560049998, partition_count: 17, processed_count: 85000)
+I, [2020-06-21T20:46:01.431828 #1]  INFO -- : [{:instance_id=>"i-03d54244a9ca1b766", :instance_type=>"r5.large", :availability_zone=>"ap-northeast-1c"}] Complete (duration: 55.10812590900059, partition_count: 16, processed_count: 80000)
+I, [2020-06-21T20:46:01.929500 #1]  INFO -- : [{:instance_id=>"i-042ca61bd40d8692c", :instance_type=>"r4.large", :availability_zone=>"ap-northeast-1a"}] Complete (duration: 257.61435206200076, partition_count: 17, processed_count: 85000)
+I, [2020-06-21T20:46:01.980481 #1]  INFO -- : [{:instance_id=>"i-042ca61bd40d8692c", :instance_type=>"r4.large", :availability_zone=>"ap-northeast-1a"}] Complete (duration: 258.9417281019996, partition_count: 17, processed_count: 85000)
+I, [2020-06-21T20:46:02.120333 #1]  INFO -- : [{:instance_id=>"i-0277b51635cc60328", :instance_type=>"r4.large", :availability_zone=>"ap-northeast-1c"}] Complete (duration: 76.57086077900021, partition_count: 17, processed_count: 85000)
+I, [2020-06-21T20:46:02.319151 #1]  INFO -- : [{:instance_id=>"i-0277b51635cc60328", :instance_type=>"r4.large", :availability_zone=>"ap-northeast-1c"}] Complete (duration: 74.19314250700063, partition_count: 16, processed_count: 80000)
+I, [2020-06-21T20:46:02.947805 #1]  INFO -- : [{:instance_id=>"i-03d54244a9ca1b766", :instance_type=>"r5.large", :availability_zone=>"ap-northeast-1c"}] Complete (duration: 54.1334114199999, partition_count: 16, processed_count: 80000)

data/bin/console

@@ -0,0 +1,14 @@
+#!/usr/bin/env ruby
+
+require "bundler/setup"
+require "kafka/ec2"
+
+# You can add fixtures and/or initialization code here to make experimenting
+# with your gem easier. You can also use a different console, if you like.
+
+# (If you use this, don't forget to add pry to your Gemfile!)
+# require "pry"
+# Pry.start
+
+require "irb"
+IRB.start(__FILE__)

data/bin/setup

@@ -0,0 +1,8 @@
+#!/usr/bin/env bash
+set -euo pipefail
+IFS=$'\n\t'
+set -vx
+
+bundle install
+
+# Do any other automated setup that you need to do here

data/lib/kafka/ec2.rb

@@ -0,0 +1,23 @@
+require "kafka/ec2/ext/consumer_group"
+require "kafka/ec2/ext/protocol/join_group_request"
+require "kafka/ec2/mixed_instance_assignment_strategy_factory"
+require "kafka/ec2/version"
+
+module Kafka
+  class EC2
+    class << self
+      attr_reader :assignment_strategy_factory
+
+      def with_assignment_strategy_factory(factory)
+        @assignment_strategy_factory = factory
+        yield
+      ensure
+        @assignment_strategy_factory = nil
+      end
+
+      def assignment_strategy_classes
+        @assignment_strategy_classes ||= {}
+      end
+    end
+  end
+end

data/lib/kafka/ec2/ext/consumer_group.rb

@@ -0,0 +1,33 @@
+# frozen_string_literal: true
+
+require "kafka/consumer_group"
+require "kafka/ec2/mixed_instance_assignment_strategy"
+
+module Kafka
+  class EC2
+    module Ext
+      module ConsumerGroup
+        def initialize(*args, **kwargs)
+          super
+          if Kafka::EC2.assignment_strategy_factory
+            @assignment_strategy = Kafka::EC2.assignment_strategy_factory.create(cluster: @cluster)
+          end
+          Kafka::EC2.assignment_strategy_classes[@group_id] = @assignment_strategy.class
+        end
+
+        def join_group
+          super
+          if Kafka::EC2.assignment_strategy_classes[@group_id] == Kafka::EC2::MixedInstanceAssignmentStrategy
+            @assignment_strategy.member_id_to_metadata = @members
+          end
+        end
+      end
+    end
+  end
+end
+
+module Kafka
+  class ConsumerGroup
+    prepend Kafka::EC2::Ext::ConsumerGroup
+  end
+end

data/lib/kafka/ec2/ext/protocol/join_group_request.rb

@@ -0,0 +1,39 @@
+# frozen_string_literal: true
+
+require "net/http"
+
+require "kafka/protocol/consumer_group_protocol"
+require "kafka/protocol/join_group_request"
+
+module Kafka
+  class EC2
+    module Ext
+      module Protocol
+        module JoinGroupRequest
+          def initialize(*args, topics:, **kwargs)
+            super
+            if Kafka::EC2.assignment_strategy_classes[@group_id] == Kafka::EC2::MixedInstanceAssignmentStrategy
+              user_data = Net::HTTP.start("169.254.169.254", 80) do |http|
+                instance_id = http.get("/latest/meta-data/instance-id").body
+                instance_type = http.get("/latest/meta-data/instance-type").body
+                az = http.get("/latest/meta-data/placement/availability-zone").body
+                "|#{instance_id},#{instance_type},#{az}"
+              end
+              @group_protocols = {
+                "mixedinstance" => Kafka::Protocol::ConsumerGroupProtocol.new(topics: topics, user_data: user_data),
+              }
+            end
+          end
+        end
+      end
+    end
+  end
+end
+
+module Kafka
+  module Protocol
+    class JoinGroupRequest
+      prepend Kafka::EC2::Ext::Protocol::JoinGroupRequest
+    end
+  end
+end

data/lib/kafka/ec2/mixed_instance_assignment_strategy.rb

@@ -0,0 +1,95 @@
+# frozen_string_literal: true
+
+require "kafka"
+require "kafka/protocol/member_assignment"
+
+module Kafka
+  class EC2
+    class MixedInstanceAssignmentStrategy
+      # metadata is a byte sequence created by Kafka::Protocol::ConsumerGroupProtocol.encode
+      attr_accessor :member_id_to_metadata
+
+      # @param cluster [Kafka::Cluster]
+      # @param instance_family_weights [Hash{String => Numeric}, Proc] a hash whose the key
+      #   is the instance family and whose value is the weight. If the object is a proc,
+      #   it must returns such a hash and the proc is called every time the method "assign"
+      #   is called.
+      # @param availability_zone_weights [Hash{String => Numeric}, Proc] a hash whose the key
+      #   is the availability zone and whose value is the weight. If the object is a proc,
+      #   it must returns such a hash and the proc is called every time the method "assign"
+      #   is called.
+      def initialize(cluster:, instance_family_weights:, availability_zone_weights:)
+        @cluster = cluster
+        @instance_family_weights = instance_family_weights
+        @availability_zone_weights = availability_zone_weights
+      end
+
+      # Assign the topic partitions to the group members.
+      #
+      # @param members [Array<String>] member ids
+      # @param topics [Array<String>] topics
+      # @return [Hash{String => Protocol::MemberAssignment}] a hash mapping member
+      #   ids to assignments.
+      def assign(members:, topics:)
+        group_assignment = {}
+        instance_id_to_capacity = Hash.new(0)
+        instance_id_to_member_ids = Hash.new { |h, k| h[k] = [] }
+        total_capacity = 0
+
+        instance_family_to_capacity = @instance_family_weights.is_a?(Proc) ? @instance_family_weights.call() : @instance_family_weights
+        az_to_capacity = @availability_zone_weights.is_a?(Proc) ? @availability_zone_weights.call() : @availability_zone_weights
+        members.each do |member_id|
+          group_assignment[member_id] = Protocol::MemberAssignment.new
+
+          instance_id, instance_type, az = member_id_to_metadata[member_id].split(",")
+          instance_id_to_member_ids[instance_id] << member_id
+          capacity = calculate_capacity(instance_type, az, instance_family_to_capacity, az_to_capacity)
+          instance_id_to_capacity[instance_id] += capacity
+          total_capacity += capacity
+        end
+
+        topic_partitions = topics.flat_map do |topic|
+          begin
+            partitions = @cluster.partitions_for(topic).map(&:partition_id)
+          rescue UnknownTopicOrPartition
+            raise UnknownTopicOrPartition, "unknown topic #{topic}"
+          end
+          Array.new(partitions.count) { topic }.zip(partitions)
+        end
+
+        partition_count_per_capacity = topic_partitions.size / total_capacity
+        last_index = 0
+        instance_id_to_capacity.sort_by { |_, capacity| -capacity }.each do |instance_id, capacity|
+          partition_count = (capacity * partition_count_per_capacity).round
+          member_ids = instance_id_to_member_ids[instance_id]
+          topic_partitions[last_index, partition_count]&.each_with_index do |(topic, partition), index|
+            member_id = member_ids[index % member_ids.size]
+            group_assignment[member_id].assign(topic, [partition])
+          end
+
+          last_index += partition_count
+        end
+
+        if last_index < topic_partitions.size
+          member_ids = instance_id_to_member_ids.values.flatten
+          topic_partitions[last_index, topic_partitions.size].each_with_index do |(topic, partition), index|
+            member_id = member_ids[index % member_ids.size]
+            group_assignment[member_id].assign(topic, [partition])
+          end
+        end
+
+        group_assignment
+      rescue Kafka::LeaderNotAvailable
+        sleep 1
+        retry
+      end
+
+      private
+
+      def calculate_capacity(instance_type, az, instance_family_to_capacity, az_to_capacity)
+        instance_family, _ = instance_type.split(".")
+        instance_family_to_capacity.fetch(instance_family, 1) * az_to_capacity.fetch(az, 1)
+      end
+    end
+  end
+end

data/lib/kafka/ec2/mixed_instance_assignment_strategy_factory.rb

@@ -0,0 +1,24 @@
+# frozen_string_literal: true
+
+require "kafka/ec2/mixed_instance_assignment_strategy"
+
+module Kafka
+  class EC2
+    class MixedInstanceAssignmentStrategyFactory
+      # @param instance_family_weights [Hash, Proc]
+      # @param availability_zone_weights [Hash, Proc]
+      def initialize(instance_family_weights: {}, availability_zone_weights: {})
+        @instance_family_weights = instance_family_weights
+        @availability_zone_weights = availability_zone_weights
+      end
+
+      def create(cluster:)
+        Kafka::EC2::MixedInstanceAssignmentStrategy.new(
+          cluster: cluster,
+          instance_family_weights: @instance_family_weights,
+          availability_zone_weights: @availability_zone_weights,
+        )
+      end
+    end
+  end
+end

data/lib/kafka/ec2/version.rb

@@ -0,0 +1,5 @@
+module Kafka
+  class EC2
+    VERSION = "0.1.0"
+  end
+end

data/ruby-kafka-ec2.gemspec

@@ -0,0 +1,29 @@
+require_relative 'lib/kafka/ec2/version'
+
+Gem::Specification.new do |spec|
+  spec.name          = "ruby-kafka-ec2"
+  spec.version       = Kafka::EC2::VERSION
+  spec.authors       = ["abicky"]
+  spec.email         = ["takeshi.arabiki@gmail.com"]
+
+  spec.summary       = %q{An extension of ruby-kafka for EC2}
+  spec.description   = %q{Kafka::EC2 is an extension of ruby-kafka that provides useful features for EC2 like Kafka::EC2::MixedInstanceAssignmentStrategy.}
+  spec.homepage      = "https://github.com/abicky/ruby-kafka-ec2"
+  spec.license       = "MIT"
+  spec.required_ruby_version = Gem::Requirement.new(">= 2.3.0")
+
+  spec.metadata["homepage_uri"] = spec.homepage
+  spec.metadata["source_code_uri"] = spec.homepage
+
+  spec.files         = Dir.chdir(File.expand_path('..', __FILE__)) do
+    `git ls-files -z`.split("\x0").reject { |f| f.match(%r{^(test|spec|features)/}) }
+  end
+  spec.bindir        = "exe"
+  spec.executables   = spec.files.grep(%r{^exe/}) { |f| File.basename(f) }
+  spec.require_paths = ["lib"]
+
+  spec.add_runtime_dependency "ruby-kafka", ">= 0.7", "< 2"
+
+  spec.add_development_dependency "webmock"
+  spec.add_development_dependency "concurrent-ruby" # cf. https://github.com/zendesk/ruby-kafka/pull/835
+end

metadata

@@ -0,0 +1,123 @@
+--- !ruby/object:Gem::Specification
+name: ruby-kafka-ec2
+version: !ruby/object:Gem::Version
+  version: 0.1.0
+platform: ruby
+authors:
+- abicky
+autorequire: 
+bindir: exe
+cert_chain: []
+date: 2020-06-21 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: ruby-kafka
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0.7'
+    - - "<"
+      - !ruby/object:Gem::Version
+        version: '2'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0.7'
+    - - "<"
+      - !ruby/object:Gem::Version
+        version: '2'
+- !ruby/object:Gem::Dependency
+  name: webmock
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: concurrent-ruby
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+description: Kafka::EC2 is an extension of ruby-kafka that provides useful features
+  for EC2 like Kafka::EC2::MixedInstanceAssignmentStrategy.
+email:
+- takeshi.arabiki@gmail.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- ".github/workflows/test.yml"
+- ".gitignore"
+- ".rspec"
+- Gemfile
+- LICENSE.txt
+- README.md
+- Rakefile
+- benchmark/.gitignore
+- benchmark/Dockerfile
+- benchmark/Dockerfile.dockerignore
+- benchmark/Gemfile
+- benchmark/README.md
+- benchmark/aws.tf
+- benchmark/consume_messages.rb
+- benchmark/consume_messages.sh
+- benchmark/produce_messages.rb
+- benchmark/produce_messages.sh
+- benchmark/register_docker_image.sh
+- benchmark/ruby-kafka-ec2.log
+- benchmark/ruby-kafka.log
+- bin/console
+- bin/setup
+- lib/kafka/ec2.rb
+- lib/kafka/ec2/ext/consumer_group.rb
+- lib/kafka/ec2/ext/protocol/join_group_request.rb
+- lib/kafka/ec2/mixed_instance_assignment_strategy.rb
+- lib/kafka/ec2/mixed_instance_assignment_strategy_factory.rb
+- lib/kafka/ec2/version.rb
+- ruby-kafka-ec2.gemspec
+homepage: https://github.com/abicky/ruby-kafka-ec2
+licenses:
+- MIT
+metadata:
+  homepage_uri: https://github.com/abicky/ruby-kafka-ec2
+  source_code_uri: https://github.com/abicky/ruby-kafka-ec2
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: 2.3.0
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubygems_version: 3.1.2
+signing_key: 
+specification_version: 4
+summary: An extension of ruby-kafka for EC2
+test_files: []