SlideShare a Scribd company logo

A 60-mn tour of AWS compute (March 2016)

•
•
Julien SIMON
Julien SIMON

This document summarizes a 60-minute talk on AWS compute technologies including EC2, ECS, Lambda, and Elastic Beanstalk. The talk provides an introduction to each service, demos of launching EC2 instances, deploying apps with Elastic Beanstalk and ECS, and implementing APIs with Lambda. It also lists upcoming user group events and a new book on AWS Lambda.

1 of 27
Download to read offline
© 2015, Amazon Web Services, Inc. or its Affiliates. All rights reserved. A 60-minute tour of AWS Compute Julien Simon, Principal Technical Evangelist @julsimon Meetup AWS User Group Rennes #1 29/03/2016 EC2 EC2 Container Service LambdaElastic Beanstalk
What to expect from this talk •  An introduction to all four AWS Compute technologies •  A good understanding on when to use them best •  Demos •  Launching an EC2 instance from the CLI •  Deploying a Ruby on Rails web app with Elastic Beanstalk •  Deploying a PHP web app with ECS •  Implementing an API with API Gateway and a Lambda function in Python •  Reacting to S3 events with a Lambda function in Java •  Answers to your questions J
AWS Compute technologies
Amazon EC2 •  Infrastructure as a Service, launched in 2006 •  Based on virtual machines (“EC2 instances”) and images (“Amazon Machine Image”, “AMI”) •  Many instance types for different needs: general purpose, compute, memory, GPU, etc. •  Users can pick from Amazon-supported AMIs, vendor-supported AMIs (“EC2 Marketplace”) or they can build their own •  All-inclusive: networking (“Virtual Private Cloud”), storage (“Elastic Block Storage”), firewalling (“Security Group”), load balancing (“Elastic Load Balancing”), high availability (“Availability Zones”), automatic scaling (“Auto-scaling groups”), monitoring (“Cloudwatch”) •  Pay on an hourly basis The best option if you need full control over your instances Use Reserved Instances and Spot instances for massive savings
Amazon EC2 demo Launch an Amazon Linux instance in the default VPC with the default security group $ aws ec2 run-instances --image-id ami-e1398992 --instance-type t2.micro --key-name lab2 --security-group-ids sg-09238e6d --region eu-west-1 This is the most important command ;) Take some time to experiment with the ‘aws ec2’ command line
Amazon Elastic Beanstalk •  Platform as a Service, launched in 2011 •  Supports PHP, Java, .NET, Node.js, Python, Go, Ruby IIS, Tomcat and Docker containers •  Developer-friendly CLI : ‘eb’ •  Uses AWS Cloudformation to build all required resources •  Built-in monitoring (Amazon Cloudwatch), networking (Amazon VPC), load balancing (Amazon ELB) and scaling (Auto-scaling) •  Relational data tier is available through Amazon Relational Data Service (RDS) •  No charge for the service itself The simplest and most intuitive way to deploy your applications This should really be your default option for deployment
Amazon Elastic Beanstalk demo 1.  Create a new Rails application 2.  Add a resource to the application 3.  Declare a new Rails application in Amazon Elastic Beanstalk 4.  Create an environment and launch the application
Create a new Rails application $ rails new blog $ cd blog $ git init $ git add . $ git commit -m "Initial version"
Add a ‘post’ resource to the application $ rails generate scaffold post title:string body:text $ bundle exec rake db:migrate $ git add . $ git commit -m "Add post resource" $ rails server $ open http://localhost:3000/posts
Initialize a Ruby application $ eb init blog --platform Ruby --region eu-west-1 $ git add .gitignore $ git commit -m "Ignore .elasticbeantalk directory"
Create a ‘blog-dev’ environment Single instance (no auto scaling, no load balancing), t2.micro instance size (default value) $ eb create blog-dev --single --keyname aws-eb --envvars SECRET_KEY_BASE=`rake secret` $ eb deploy $ eb terminate blog-dev –force For more information on Elastic Beanstalk (load balancing, high availability, RDS with Postgres) http://www.slideshare.net/JulienSIMON5/deploying-a-simple-rails-application-with-aws-elastic-beanstalk
Amazon EC2 Container Service •  Container as a Service, launched in 2015 •  Built-in clustering, state management, scheduling and high availability •  EC2 Container Registry (ECR): private Docker registry hosted in AWS •  Developer-friendly CLI : ‘ecs-cli’ •  Uses AWS Cloudformation to build all required resources •  Supports Docker 1.9.1, including Docker Compose files •  No charge for the service itself A simple and scalable way to manage your Dockerized applications
Amazon ECS demo 1.  Build a Docker image for a simple PHP web app 2.  Push it to an ECR repository 3.  Create an ECS cluster 4.  Deploy and scale the containerized web app
Amazon ECR demo: build and push container $ git clone https://github.com/awslabs/ecs-demo-php-simple-app.git $ cd ecs-demo-php-simple-app $ docker build -t php-simple-app . $ docker tag php-simple-app:latest ACCOUNT_ID.dkr.ecr.us-east-1.amazonaws.com/php-simple-app:latest $ aws ecr get-login --region us-east-1 <run docker login command provided as output> $ docker push ACCOUNT_ID.dkr.ecr.us-east-1.amazonaws.com php-simple-app:latest
Amazon ECS demo: write a Compose file php-demo: image: ACCOUNT_ID.dkr.ecr.us-east-1.amazonaws.com/php-simple-app:latest cpu_shares: 100 mem_limit: 134217728 ports: - "80:80" entrypoint: - "/usr/sbin/apache2" - "-D" - "FOREGROUND"
Amazon ECS demo: launch cluster & service $ ecs-cli configure --cluster myCluster --region eu-west-1 $ ecs-cli up --keypair lab2 --capability-iam --size 1 --instance-type t2.micro $ ecs-cli compose service up $ ecs-cli scale --size 3 --capability-iam $ ecs-cli compose service scale 3 $ ecs-cli compose service delete $ ecs-cli down myCluster --force For more information on ECS: http://www.slideshare.net/JulienSIMON5/amazon-ecs-january-2016
AWS Lambda •  Code as a Service, launched in 2014 •  Supports Java 8, Python 2.7 and Node.js v0.10.36 •  Build event-driven applications •  Build APIs in conjunction with Amazon API Gateway •  Interact with other AWS services (S3, DynamoDB, etc) •  Log automatically to CloudWatch Logs •  Pay as you go: number of requests + execution time (100ms slots) The future: serverless applications and NoOps J
AWS Lambda demo (Python) 1.  Write a simple Lambda function adding two integers 2.  Create a REST API with API Gateway (resource + method) 3.  Create a new stage 4.  Deploy our API to the stage 5.  Invoke the API with ‘curl’
A simple Lambda function in Python def lambda_handler(event,context):    result = event['value1'] + event['value2']    return result $ curl -H "Content-Type: application/json" -X POST -d "{"value1":5, "value2":7}" https://API_ENDPOINT/STAGE/RESOURCE 12%
AWS Lambda in Java with Eclipse https://java.awsblog.com/post/TxWZES6J1RSQ2Z/Testing-Lambda-functions-using-the- AWS-Toolkit-for-Eclipse
AWS Lambda demo (Java) 1.  In Eclipse, write a simple Lambda function triggered by an S3 event 2.  Unit-test the function with Junit 3.  Using the AWS Eclipse plug-in, upload and run the function in AWS 4.  Run the function again in the AWS Console
AWS Lambda in Node.js with Serverless framework https://github.com/serverless/serverless •  Run/test AWS Lambda functions locally, or remotely •  Auto-deploys & versions your Lambda functions •  Auto-deploys your REST API to AWS API Gateway •  Auto-deploys your Lambda events •  Support for multiple stages •  Support for multiple regions within stages •  Manage & deploy AWS CloudFormation resources
Upcoming book on AWS Lambda Written by AWS Technical Evangelist Danilo Poccia Early release available at: https://www.manning.com/books/aws- lambda-in-action
And now the trip begins. Time to explore! https://aws.amazon.com/fr/documentation/gettingstarted/ https://docs.aws.amazon.com https://aws.amazon.com/fr/blogs/compute/
April 20-22April 6-7 (Lyon) April 25 May 31st June 28 September 27 December 6 Next events
AWS User Groups AWS Lille Paris Rennes Nantes Bordeaux Lyon Montpellier facebook.com/groups/AWSFrance/ @aws_actus AWS User Groups
Merci ! Julien Simon Principal Technical Evangelist, AWS julsimon@amazon.fr @julsimon

More Related Content

A 60-mn tour of AWS compute (March 2016)

  • 1. © 2015, Amazon Web Services, Inc. or its Affiliates. All rights reserved. A 60-minute tour of AWS Compute Julien Simon, Principal Technical Evangelist @julsimon Meetup AWS User Group Rennes #1 29/03/2016 EC2 EC2 Container Service LambdaElastic Beanstalk
  • 2. What to expect from this talk •  An introduction to all four AWS Compute technologies •  A good understanding on when to use them best •  Demos •  Launching an EC2 instance from the CLI •  Deploying a Ruby on Rails web app with Elastic Beanstalk •  Deploying a PHP web app with ECS •  Implementing an API with API Gateway and a Lambda function in Python •  Reacting to S3 events with a Lambda function in Java •  Answers to your questions J
  • 4. Amazon EC2 •  Infrastructure as a Service, launched in 2006 •  Based on virtual machines (“EC2 instances”) and images (“Amazon Machine Image”, “AMI”) •  Many instance types for different needs: general purpose, compute, memory, GPU, etc. •  Users can pick from Amazon-supported AMIs, vendor-supported AMIs (“EC2 Marketplace”) or they can build their own •  All-inclusive: networking (“Virtual Private Cloud”), storage (“Elastic Block Storage”), firewalling (“Security Group”), load balancing (“Elastic Load Balancing”), high availability (“Availability Zones”), automatic scaling (“Auto-scaling groups”), monitoring (“Cloudwatch”) •  Pay on an hourly basis The best option if you need full control over your instances Use Reserved Instances and Spot instances for massive savings
  • 5. Amazon EC2 demo Launch an Amazon Linux instance in the default VPC with the default security group $ aws ec2 run-instances --image-id ami-e1398992 --instance-type t2.micro --key-name lab2 --security-group-ids sg-09238e6d --region eu-west-1 This is the most important command ;) Take some time to experiment with the ‘aws ec2’ command line
  • 6. Amazon Elastic Beanstalk •  Platform as a Service, launched in 2011 •  Supports PHP, Java, .NET, Node.js, Python, Go, Ruby IIS, Tomcat and Docker containers •  Developer-friendly CLI : ‘eb’ •  Uses AWS Cloudformation to build all required resources •  Built-in monitoring (Amazon Cloudwatch), networking (Amazon VPC), load balancing (Amazon ELB) and scaling (Auto-scaling) •  Relational data tier is available through Amazon Relational Data Service (RDS) •  No charge for the service itself The simplest and most intuitive way to deploy your applications This should really be your default option for deployment
  • 7. Amazon Elastic Beanstalk demo 1.  Create a new Rails application 2.  Add a resource to the application 3.  Declare a new Rails application in Amazon Elastic Beanstalk 4.  Create an environment and launch the application
  • 8. Create a new Rails application $ rails new blog $ cd blog $ git init $ git add . $ git commit -m "Initial version"
  • 9. Add a ‘post’ resource to the application $ rails generate scaffold post title:string body:text $ bundle exec rake db:migrate $ git add . $ git commit -m "Add post resource" $ rails server $ open http://localhost:3000/posts
  • 10. Initialize a Ruby application $ eb init blog --platform Ruby --region eu-west-1 $ git add .gitignore $ git commit -m "Ignore .elasticbeantalk directory"
  • 11. Create a ‘blog-dev’ environment Single instance (no auto scaling, no load balancing), t2.micro instance size (default value) $ eb create blog-dev --single --keyname aws-eb --envvars SECRET_KEY_BASE=`rake secret` $ eb deploy $ eb terminate blog-dev –force For more information on Elastic Beanstalk (load balancing, high availability, RDS with Postgres) http://www.slideshare.net/JulienSIMON5/deploying-a-simple-rails-application-with-aws-elastic-beanstalk
  • 12. Amazon EC2 Container Service •  Container as a Service, launched in 2015 •  Built-in clustering, state management, scheduling and high availability •  EC2 Container Registry (ECR): private Docker registry hosted in AWS •  Developer-friendly CLI : ‘ecs-cli’ •  Uses AWS Cloudformation to build all required resources •  Supports Docker 1.9.1, including Docker Compose files •  No charge for the service itself A simple and scalable way to manage your Dockerized applications
  • 13. Amazon ECS demo 1.  Build a Docker image for a simple PHP web app 2.  Push it to an ECR repository 3.  Create an ECS cluster 4.  Deploy and scale the containerized web app
  • 14. Amazon ECR demo: build and push container $ git clone https://github.com/awslabs/ecs-demo-php-simple-app.git $ cd ecs-demo-php-simple-app $ docker build -t php-simple-app . $ docker tag php-simple-app:latest ACCOUNT_ID.dkr.ecr.us-east-1.amazonaws.com/php-simple-app:latest $ aws ecr get-login --region us-east-1 <run docker login command provided as output> $ docker push ACCOUNT_ID.dkr.ecr.us-east-1.amazonaws.com php-simple-app:latest
  • 15. Amazon ECS demo: write a Compose file php-demo: image: ACCOUNT_ID.dkr.ecr.us-east-1.amazonaws.com/php-simple-app:latest cpu_shares: 100 mem_limit: 134217728 ports: - "80:80" entrypoint: - "/usr/sbin/apache2" - "-D" - "FOREGROUND"
  • 16. Amazon ECS demo: launch cluster & service $ ecs-cli configure --cluster myCluster --region eu-west-1 $ ecs-cli up --keypair lab2 --capability-iam --size 1 --instance-type t2.micro $ ecs-cli compose service up $ ecs-cli scale --size 3 --capability-iam $ ecs-cli compose service scale 3 $ ecs-cli compose service delete $ ecs-cli down myCluster --force For more information on ECS: http://www.slideshare.net/JulienSIMON5/amazon-ecs-january-2016
  • 17. AWS Lambda •  Code as a Service, launched in 2014 •  Supports Java 8, Python 2.7 and Node.js v0.10.36 •  Build event-driven applications •  Build APIs in conjunction with Amazon API Gateway •  Interact with other AWS services (S3, DynamoDB, etc) •  Log automatically to CloudWatch Logs •  Pay as you go: number of requests + execution time (100ms slots) The future: serverless applications and NoOps J
  • 18. AWS Lambda demo (Python) 1.  Write a simple Lambda function adding two integers 2.  Create a REST API with API Gateway (resource + method) 3.  Create a new stage 4.  Deploy our API to the stage 5.  Invoke the API with ‘curl’
  • 19. A simple Lambda function in Python def lambda_handler(event,context):    result = event['value1'] + event['value2']    return result $ curl -H "Content-Type: application/json" -X POST -d "{"value1":5, "value2":7}" https://API_ENDPOINT/STAGE/RESOURCE 12%
  • 20. AWS Lambda in Java with Eclipse https://java.awsblog.com/post/TxWZES6J1RSQ2Z/Testing-Lambda-functions-using-the- AWS-Toolkit-for-Eclipse
  • 21. AWS Lambda demo (Java) 1.  In Eclipse, write a simple Lambda function triggered by an S3 event 2.  Unit-test the function with Junit 3.  Using the AWS Eclipse plug-in, upload and run the function in AWS 4.  Run the function again in the AWS Console
  • 22. AWS Lambda in Node.js with Serverless framework https://github.com/serverless/serverless •  Run/test AWS Lambda functions locally, or remotely •  Auto-deploys & versions your Lambda functions •  Auto-deploys your REST API to AWS API Gateway •  Auto-deploys your Lambda events •  Support for multiple stages •  Support for multiple regions within stages •  Manage & deploy AWS CloudFormation resources
  • 23. Upcoming book on AWS Lambda Written by AWS Technical Evangelist Danilo Poccia Early release available at: https://www.manning.com/books/aws- lambda-in-action
  • 24. And now the trip begins. Time to explore! https://aws.amazon.com/fr/documentation/gettingstarted/ https://docs.aws.amazon.com https://aws.amazon.com/fr/blogs/compute/
  • 25. April 20-22April 6-7 (Lyon) April 25 May 31st June 28 September 27 December 6 Next events
  • 26. AWS User Groups AWS Lille Paris Rennes Nantes Bordeaux Lyon Montpellier facebook.com/groups/AWSFrance/ @aws_actus AWS User Groups
  • 27. Merci ! Julien Simon Principal Technical Evangelist, AWS julsimon@amazon.fr @julsimon