SlideShare a Scribd company logo

Advanced Task Scheduling with Amazon ECS (June 2017)

•
•
Julien SIMON
Julien SIMON

This document provides an overview of advanced task scheduling capabilities with Amazon ECS. It discusses the ECS placement engine which gives developers more control over task placement through constraints and strategies. Constraints allow targeting specific instance types, availability zones or custom attributes. Strategies like spread, binpack, and affinity can distribute tasks across instances. The document demonstrates how to use these features to optimize task placement and provides examples of companies using ECS for production workloads.

Related slideshows

Running Open Source Platforms on AWS (November 2016)
Running Open Source Platforms on AWS (November 2016)Running Open Source Platforms on AWS (November 2016)
Running Open Source Platforms on AWS (November 2016)
 
Disaster Recovery Site on AWS - Minimal Cost Maximum Efficiency (STG305) | AW...
Disaster Recovery Site on AWS - Minimal Cost Maximum Efficiency (STG305) | AW...Disaster Recovery Site on AWS - Minimal Cost Maximum Efficiency (STG305) | AW...
Disaster Recovery Site on AWS - Minimal Cost Maximum Efficiency (STG305) | AW...
 
Amazon EC2 - Masterclass - Pop-up Loft Tel Aviv
Amazon EC2 - Masterclass - Pop-up Loft Tel AvivAmazon EC2 - Masterclass - Pop-up Loft Tel Aviv
Amazon EC2 - Masterclass - Pop-up Loft Tel Aviv
 
1 of 43
Download to read offline
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Julien Simon Principal Technical Evangelist, AWS @julsimon Advanced Task Scheduling with Amazon ECS
Docker on Amazon Web Services Amazon EC2 Container Service (ECS) • https://aws.amazon.com/ecs/ • Launched in 04/2015 • No additional charge Amazon EC2 Container Registry (ECR) • https://aws.amazon.com/ecr/ • Launched in 12/2015 • Free tier: 500MB / month for a year • $0.10 / GB / month + outgoing traffic Both services are available in US, EU and APAC
ECS adoption https://www.datadoghq.com/blog/3-clear-trends-in-ecs-adoption/ November 2016
Companies Selected ECS customers
Container Partners https://aws.amazon.com/containers/partners/
ECS Scheduling
The problem Given a certain amount of computing power and memory, how can we best manage an arbitrary number of apps running in Docker containers? http://tidalseven.com
Amazon ECS: Under the Hood ALB ALB AZ 1 AZ 2 user / scheduler https://github.com/aws/amazon-ecs-agenthttp://www.allthingsdistributed.com/2015/07/under-the-hood-of-the-amazon-ec2-container-service.html
Scheduling on ECS: two options so far 1. Let ECS handle scheduling through Services • Task Definition • ECS transcription of the Docker Compose file • Versioning • cpu_shares, mem_limit • Number of desired containers 2. Implement a custom scheduler with the ECS API • Describe cluster state • Select a specific ECS instance according to custom logic • Run task on this instance • Coursera use case: www.youtube.com/watch?v=a45J6xAGUvA
ECS Placement Engine
Placement Engine: giving developers more control ALB ALB AZ 1 AZ 2 user / scheduler Placement Engine Constraints & Strategies
Placement Constraints Name Example AMI ID attribute:ecs.ami-id == ami-eca289fb Availability Zone attribute:ecs.availability-zone == us-east-1a Instance Type attribute:ecs.instance-type == t2.small Distinct Instances type=“distinctInstance” Custom attribute:stack == prod
Example: Constraint on Instance Family/Type
Example: Constraint on Availability Zone
Example: Combining Multiple Constraints
Placement Strategies Binpacking Spread Affinity Distinct Instance
Placement Strategy Chaining Spread tasks across Zones and Binpack within each Zone
Placing Tasks
Anatomy of Task Placement Cluster Constraints Custom Constraints Placement Strategies Apply Filter Satisfy CPU, memory, and port requirements Filter for location, instance-type, AMI, or custom attribute constraints Identify instances that meet spread or binpack placement strategy Select final container instances for placement
g2.2xlarge t2.small g2.2xlarge g2.2xlarge Placement: Targeting Instance Type
g2.2xlarge t2.small t2.micro t2.medium t2.medium t2.small g2.2xlarge t2.small t2.small t2.medium us-east-1aus-east-1d Placement: Targeting Instance Type & Zone
g2.2xlarge t2.small t2.micro t2.medium t2.medium t2.small g2.2xlarge t2.small us-east-1aus-east-1d g2.2xlarge t2.medium t2.micro t2.small us-east-1c Placement: Availability Zone Spread
g2.2xlarge t2.small t2.micro t2.medium t2.medium t2.small g2.2xlarge t2.small us-east-1aus-east-1d g2.2xlarge t2.medium t2.micro t2.small us-east-1c Placement: Spread across Zone and Binpack
g2.2xlarge t2.small t2.micro t2.medium t2.medium t2.small g2.2xlarge t2.small us-east-1aus-east-1d g2.2xlarge t2.medium t2.micro t2.small us-east-1c Placement: Affinity and Anti-Affinity
Running a Service
t2.medium t2.small t2.small us-east-1aus-east-1d t2.medium t2.micro t2.small us-east-1c Placement: Multiple Services on a Cluster
t2.medium g2.2xlarge t2.micro t2.small t2.small t2.small g2.2xlarge t2.small t2.small t2.small g2.2xlarge t2.small Placement: Services – Distinct Instances
Event Stream & Blox
Amazon ECS: Under the Hood ALB ALB AZ 1 AZ 2 user / scheduler Placement Engine Event Stream
Consuming Real-time Events
Handling ECS events with Blox scheduler cluster state service https://blox.github.io/
Amazon ECS: Under the Hood ALB ALB AZ 1 AZ 2 user / scheduler Scheduler Cluster State Service Placement Engine Event Stream
t2.small t2.small t2.small Blox: Daemon Scheduler t2.small t2.small t2.small scheduler cluster state service t2.small t2.small t2.small
Demo: Deploying Blox on AWS
Creating Clusters Create an ECS cluster for Blox CF template: https://github.com/blox/blox/blob/dev/deploy/aws/conf/cloudformation_template.json à CloudWatch Event Rule + SQS queue à 3 containers: Daemon Scheduler + Cluster State Service + etcd à REST API exposing the Daemon Scheduler API Create another ECS cluster managed by Blox $ ecs-cli configure --cluster WebCluster $ ecs-cli up --keypair admin --capability-iam --size 3 --instance-type t2.micro Invoke the scheduler API ‘demo-cli’ tool: https://github.com/blox/blox/tree/dev/deploy/demo-cli
Listing Task Definitions Grab the ARN for an nginx Task Definition, which the Daemon Scheduler will manage on ‘WebCluster’. $ ./list-task-definitions.py --region eu-central-1 == Blox Demo CLI - List Task Definitions == { "taskDefinitionArns": [ "arn:aws:ecs:eu-central-1:ACCOUNT:task-definition/BloxFramework:2", "arn:aws:ecs:eu-central-1:ACCOUNT:task-definition/nginx:1", "arn:aws:ecs:eu-central-1:ACCOUNT:task-definition/nginx:2" ] }
Creating an Environment $ ./blox-create-environment.py --environment WebEnvironment --cluster WebCluster --task-definition "arn:aws:ecs:eu-central-1:ACCOUNT:task- definition/nginx:2" --stack Blox --apigateway --region eu-central-1 == Blox Demo CLI - Create Blox Environment == HTTP Response Code: 200 { "taskDefinition": "arn:aws:ecs:eu-central-1:ACCOUNT:task-definition/nginx:2", "deploymentToken": "17248257-08ec-4438-888f-e0ac28397653", "health": "healthy", "name": "WebEnvironment", "instanceGroup": { "cluster": "arn:aws:ecs:eu-central-1:ACCOUNT:cluster/WebCluster" } }
Listing Environments $ ./blox-list-environments.py --stack Blox --apigateway --region eu-central-1 == Blox Demo CLI - List Blox Environments == HTTP Response Code: 200 { "items": [ { "taskDefinition": "arn:aws:ecs:eu-central-1:ACCOUNT:task-definition/nginx:2", "deploymentToken": "17248257-08ec-4438-888f-e0ac28397653", "health": "healthy", "name": "WebEnvironment", "instanceGroup": { "cluster": "arn:aws:ecs:eu-central-1:ACCOUNT:cluster/WebCluster" } } ] }
Creating a Deployment $ ./blox-create-deployment.py --environment WebEnvironment --deployment-token "17248257-08ec-4438-888f-e0ac28397653" --stack Blox --apigateway --region eu-central-1 == Blox Demo CLI - Create Blox Deployment == HTTP Response Code: 200 { "status": "pending", "environmentName": "WebEnvironment", "id": "7a05ea99-27a9-4339-a7a6-f4120065aea3", "failedInstances": [], "taskDefinition": "arn:aws:ecs:eu-central-1:613904931467:task-definition/nginx:2” }
Listing Deployments $ ./blox-list-deployments.py --environment WebEnvironment --stack Blox --apigateway --region eu-central-1 == Blox Demo CLI - List Blox Deployments == HTTP Response Code: 200 { "items": [ { "status": "completed", "environmentName": "WebEnvironment", "id": "7a05ea99-27a9-4339-a7a6-f4120065aea3", "failedInstances": [], "taskDefinition": "arn:aws:ecs:eu-central-1:ACCOUNT:task-definition/nginx:2" } ] }
Scaling a Deployment $ ecs-cli ps Name State Ports TaskDefinition 26313cbe-d929-49de-9cc3-873bf5f32a91/nginx RUNNING nginx:2 98442432-fd5c-434d-b93c-0737bd06aaab/nginx RUNNING nginx:2 ce9bf217-4b34-4f31-9c7b-a8c3402f1ffd/nginx RUNNING nginx:2 $ ecs-cli scale --size 4 --capability-iam $ ecs-cli ps Name State Ports TaskDefinition 26313cbe-d929-49de-9cc3-873bf5f32a91/nginx RUNNING nginx:2 98442432-fd5c-434d-b93c-0737bd06aaab/nginx RUNNING nginx:2 c404ac9a-0948-4cc8-b5b0-2238ccdf4035/nginx RUNNING nginx:2 ce9bf217-4b34-4f31-9c7b-a8c3402f1ffd/nginx RUNNING nginx:2
Additional resources ECS customers https://www.mapbox.com/blog/switch-to-ecs/ https://segment.com/blog/rebuilding-our-infrastructure/ Tech articles by Werner Vogels, CTO, Amazon.com http://www.allthingsdistributed.com/2014/11/amazon-ec2-container-service.html http://www.allthingsdistributed.com/2015/04/state-management-and-scheduling-with-ecs.html http://www.allthingsdistributed.com/2015/07/under-the-hood-of-the-amazon-ec2-container-service.html Blog articles & videos https://aws.amazon.com/blogs/compute/amazon-ec2-container-service-at-aws-reinvent-2016-wrap-up/ https://aws.amazon.com/blogs/compute/continuous-deployment-to-amazon-ecs-using-aws-codepipeline- aws-codebuild-amazon-ecr-and-aws-cloudformation/
Thank you! Julien Simon Principal Technical Evangelist, AWS @julsimon

More Related Content

Advanced Task Scheduling with Amazon ECS (June 2017)

  • 1. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Julien Simon Principal Technical Evangelist, AWS @julsimon Advanced Task Scheduling with Amazon ECS
  • 2. Docker on Amazon Web Services Amazon EC2 Container Service (ECS) • https://aws.amazon.com/ecs/ • Launched in 04/2015 • No additional charge Amazon EC2 Container Registry (ECR) • https://aws.amazon.com/ecr/ • Launched in 12/2015 • Free tier: 500MB / month for a year • $0.10 / GB / month + outgoing traffic Both services are available in US, EU and APAC
  • 7. The problem Given a certain amount of computing power and memory, how can we best manage an arbitrary number of apps running in Docker containers? http://tidalseven.com
  • 8. Amazon ECS: Under the Hood ALB ALB AZ 1 AZ 2 user / scheduler https://github.com/aws/amazon-ecs-agenthttp://www.allthingsdistributed.com/2015/07/under-the-hood-of-the-amazon-ec2-container-service.html
  • 9. Scheduling on ECS: two options so far 1. Let ECS handle scheduling through Services • Task Definition • ECS transcription of the Docker Compose file • Versioning • cpu_shares, mem_limit • Number of desired containers 2. Implement a custom scheduler with the ECS API • Describe cluster state • Select a specific ECS instance according to custom logic • Run task on this instance • Coursera use case: www.youtube.com/watch?v=a45J6xAGUvA
  • 11. Placement Engine: giving developers more control ALB ALB AZ 1 AZ 2 user / scheduler Placement Engine Constraints & Strategies
  • 12. Placement Constraints Name Example AMI ID attribute:ecs.ami-id == ami-eca289fb Availability Zone attribute:ecs.availability-zone == us-east-1a Instance Type attribute:ecs.instance-type == t2.small Distinct Instances type=“distinctInstance” Custom attribute:stack == prod
  • 13. Example: Constraint on Instance Family/Type
  • 14. Example: Constraint on Availability Zone
  • 16. Placement Strategies Binpacking Spread Affinity Distinct Instance
  • 17. Placement Strategy Chaining Spread tasks across Zones and Binpack within each Zone
  • 19. Anatomy of Task Placement Cluster Constraints Custom Constraints Placement Strategies Apply Filter Satisfy CPU, memory, and port requirements Filter for location, instance-type, AMI, or custom attribute constraints Identify instances that meet spread or binpack placement strategy Select final container instances for placement
  • 20. g2.2xlarge t2.small g2.2xlarge g2.2xlarge Placement: Targeting Instance Type
  • 21. g2.2xlarge t2.small t2.micro t2.medium t2.medium t2.small g2.2xlarge t2.small t2.small t2.medium us-east-1aus-east-1d Placement: Targeting Instance Type & Zone
  • 22. g2.2xlarge t2.small t2.micro t2.medium t2.medium t2.small g2.2xlarge t2.small us-east-1aus-east-1d g2.2xlarge t2.medium t2.micro t2.small us-east-1c Placement: Availability Zone Spread
  • 23. g2.2xlarge t2.small t2.micro t2.medium t2.medium t2.small g2.2xlarge t2.small us-east-1aus-east-1d g2.2xlarge t2.medium t2.micro t2.small us-east-1c Placement: Spread across Zone and Binpack
  • 24. g2.2xlarge t2.small t2.micro t2.medium t2.medium t2.small g2.2xlarge t2.small us-east-1aus-east-1d g2.2xlarge t2.medium t2.micro t2.small us-east-1c Placement: Affinity and Anti-Affinity
  • 26. t2.medium t2.small t2.small us-east-1aus-east-1d t2.medium t2.micro t2.small us-east-1c Placement: Multiple Services on a Cluster
  • 27. t2.medium g2.2xlarge t2.micro t2.small t2.small t2.small g2.2xlarge t2.small t2.small t2.small g2.2xlarge t2.small Placement: Services – Distinct Instances
  • 29. Amazon ECS: Under the Hood ALB ALB AZ 1 AZ 2 user / scheduler Placement Engine Event Stream
  • 31. Handling ECS events with Blox scheduler cluster state service https://blox.github.io/
  • 32. Amazon ECS: Under the Hood ALB ALB AZ 1 AZ 2 user / scheduler Scheduler Cluster State Service Placement Engine Event Stream
  • 33. t2.small t2.small t2.small Blox: Daemon Scheduler t2.small t2.small t2.small scheduler cluster state service t2.small t2.small t2.small
  • 35. Creating Clusters Create an ECS cluster for Blox CF template: https://github.com/blox/blox/blob/dev/deploy/aws/conf/cloudformation_template.json à CloudWatch Event Rule + SQS queue à 3 containers: Daemon Scheduler + Cluster State Service + etcd à REST API exposing the Daemon Scheduler API Create another ECS cluster managed by Blox $ ecs-cli configure --cluster WebCluster $ ecs-cli up --keypair admin --capability-iam --size 3 --instance-type t2.micro Invoke the scheduler API ‘demo-cli’ tool: https://github.com/blox/blox/tree/dev/deploy/demo-cli
  • 36. Listing Task Definitions Grab the ARN for an nginx Task Definition, which the Daemon Scheduler will manage on ‘WebCluster’. $ ./list-task-definitions.py --region eu-central-1 == Blox Demo CLI - List Task Definitions == { "taskDefinitionArns": [ "arn:aws:ecs:eu-central-1:ACCOUNT:task-definition/BloxFramework:2", "arn:aws:ecs:eu-central-1:ACCOUNT:task-definition/nginx:1", "arn:aws:ecs:eu-central-1:ACCOUNT:task-definition/nginx:2" ] }
  • 37. Creating an Environment $ ./blox-create-environment.py --environment WebEnvironment --cluster WebCluster --task-definition "arn:aws:ecs:eu-central-1:ACCOUNT:task- definition/nginx:2" --stack Blox --apigateway --region eu-central-1 == Blox Demo CLI - Create Blox Environment == HTTP Response Code: 200 { "taskDefinition": "arn:aws:ecs:eu-central-1:ACCOUNT:task-definition/nginx:2", "deploymentToken": "17248257-08ec-4438-888f-e0ac28397653", "health": "healthy", "name": "WebEnvironment", "instanceGroup": { "cluster": "arn:aws:ecs:eu-central-1:ACCOUNT:cluster/WebCluster" } }
  • 38. Listing Environments $ ./blox-list-environments.py --stack Blox --apigateway --region eu-central-1 == Blox Demo CLI - List Blox Environments == HTTP Response Code: 200 { "items": [ { "taskDefinition": "arn:aws:ecs:eu-central-1:ACCOUNT:task-definition/nginx:2", "deploymentToken": "17248257-08ec-4438-888f-e0ac28397653", "health": "healthy", "name": "WebEnvironment", "instanceGroup": { "cluster": "arn:aws:ecs:eu-central-1:ACCOUNT:cluster/WebCluster" } } ] }
  • 39. Creating a Deployment $ ./blox-create-deployment.py --environment WebEnvironment --deployment-token "17248257-08ec-4438-888f-e0ac28397653" --stack Blox --apigateway --region eu-central-1 == Blox Demo CLI - Create Blox Deployment == HTTP Response Code: 200 { "status": "pending", "environmentName": "WebEnvironment", "id": "7a05ea99-27a9-4339-a7a6-f4120065aea3", "failedInstances": [], "taskDefinition": "arn:aws:ecs:eu-central-1:613904931467:task-definition/nginx:2” }
  • 40. Listing Deployments $ ./blox-list-deployments.py --environment WebEnvironment --stack Blox --apigateway --region eu-central-1 == Blox Demo CLI - List Blox Deployments == HTTP Response Code: 200 { "items": [ { "status": "completed", "environmentName": "WebEnvironment", "id": "7a05ea99-27a9-4339-a7a6-f4120065aea3", "failedInstances": [], "taskDefinition": "arn:aws:ecs:eu-central-1:ACCOUNT:task-definition/nginx:2" } ] }
  • 41. Scaling a Deployment $ ecs-cli ps Name State Ports TaskDefinition 26313cbe-d929-49de-9cc3-873bf5f32a91/nginx RUNNING nginx:2 98442432-fd5c-434d-b93c-0737bd06aaab/nginx RUNNING nginx:2 ce9bf217-4b34-4f31-9c7b-a8c3402f1ffd/nginx RUNNING nginx:2 $ ecs-cli scale --size 4 --capability-iam $ ecs-cli ps Name State Ports TaskDefinition 26313cbe-d929-49de-9cc3-873bf5f32a91/nginx RUNNING nginx:2 98442432-fd5c-434d-b93c-0737bd06aaab/nginx RUNNING nginx:2 c404ac9a-0948-4cc8-b5b0-2238ccdf4035/nginx RUNNING nginx:2 ce9bf217-4b34-4f31-9c7b-a8c3402f1ffd/nginx RUNNING nginx:2
  • 42. Additional resources ECS customers https://www.mapbox.com/blog/switch-to-ecs/ https://segment.com/blog/rebuilding-our-infrastructure/ Tech articles by Werner Vogels, CTO, Amazon.com http://www.allthingsdistributed.com/2014/11/amazon-ec2-container-service.html http://www.allthingsdistributed.com/2015/04/state-management-and-scheduling-with-ecs.html http://www.allthingsdistributed.com/2015/07/under-the-hood-of-the-amazon-ec2-container-service.html Blog articles & videos https://aws.amazon.com/blogs/compute/amazon-ec2-container-service-at-aws-reinvent-2016-wrap-up/ https://aws.amazon.com/blogs/compute/continuous-deployment-to-amazon-ecs-using-aws-codepipeline- aws-codebuild-amazon-ecr-and-aws-cloudformation/
  • 43. Thank you! Julien Simon Principal Technical Evangelist, AWS @julsimon