Immersion Day

Using EC2 Auto Scaling to Build Highly Available

Environments on AWS

JUNE 2019
Table of Contents
Overview.......................................................................................................................... 3
Task 1: Creating a Launch Template............................................................................... 5
Task 2: Creating an Auto Scaling Group ....................................................................... 13
Task 3: Creating Scaling Policies .................................................................................. 17
Task 4: Testing the Auto Scaling Group ........................................................................ 19
Task 5: Clean Up ........................................................................................................... 24
Appendix – Additional Reading...................................................................................... 25

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 2
Overview
Amazon EC2 Auto Scaling

Amazon EC2 Auto Scaling helps you to maintain application availability by scaling
your infrastructure as the need or demand arises. It helps support your application’s
workload by making sure you have the right number of EC2 instances available. You
can create Auto Scaling Groups, which are collections of EC2 instances that have the
same characteristics and applications.

The number of EC2 instances can be scaled in or out as Auto Scaling responds to the
metrics you define when creating these groups.

• You can specify the minimum number of instances in each Auto Scaling Group,
so that your group never goes below this size.
• You can specify the maximum number of instances in each Auto Scaling
Group, so that your group never goes above this size.
• You can specify a desired capacity so that Auto Scaling ensures your group
always has a certain number of instances.
• You can specify scaling policies so that Auto Scaling will modify the desired
target capacity mentioned in the previous point. It will launch or terminate
instances as demand on your application increases or decreases.

There are multiple components of Auto Scaling on AWS. They are:

1. Groups:

Your EC2 instances are organized into groups so that they can be treated as a
logical unit for the purposes of scaling and management. When you create a
group, you can specify its minimum, maximum, and desired number of EC2
instances.

2. Launch Template:

A Launch Template is a capability of EC2 Auto Scaling that allows a way to

templatize your launch requests. It enables you to store launch parameters so
that you do not have to specify them every time you launch an instance. For
example, a launch template can contain an Amazon Machine Image, instance
type, storage, and networking settings that you typically use to launch
instances. You can also specify advanced configurations like user data, as well

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 3
 as the ability to choose T2/T3 Unlimited EC2 instances, where you can decide
whether to enable applications to burst beyond the baseline CPU performance
for as long as needed. For each Launch Template, you can create one or more
numbered Launch Template Versions. Each version can have different launch
parameters. When you create an Auto Scaling Group that is backed by a Launch
Template, you also have the option of launching one type of instance, or a
combination of instance types and purchase options.

3. Scaling Policies:

A Scaling Policy tells Auto Scaling when and how to scale. Scaling can occur
manually, on a schedule, on demand or you can use Auto Scaling to maintain a
specific number of instances.

Auto Scaling is well suited for applications that have unpredictable demand patterns
that can experience hourly, daily, or weekly variability in usage. This helps you to
manage your cost and eliminate over-provisioning of capacity during times when it is
not needed. Auto Scaling can also find an unhealthy instance, terminate that instance,
and launch a new one based on the scaling plan

Getting Started

This lab will walk you through the process of building out these components. For the
purposes of this lab, you will be using the AWS Management Console. Auto Scaling
can also be configured from the AWS CLI or Windows PowerShell if you prefer a
command line interface. You can also do this via an available SDK.

This lab assumes that you have an AWS account and the appropriate permissions to
use the services involved.

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 4
Task 1: Creating a Launch Template
When you create an Auto Scaling Group, you must specify a Launch Template. The
first step in this lab is to create the Launch Template for an EC2 Auto Scaling Group.

1. Sign into the AWS Management Console and on the Services menu, click EC2.

2. In the left navigation pane, find AUTO SCALING and click Auto Scaling Groups.

3. Click Create Auto Scaling group.

The console explains that the first step is to define a Launch Template and
the second step is to create the Auto Scaling Group.

4. Click Get Started if prompted.

5. Select Launch Template, and then click Create a new launch template.

a. If you do not see the option to create a Launch Template at this step,

look for INSTANCES in the left navigation pane and select Launch

Templates to get started instead

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 5
 6. This takes you to the Create launch template page where you can define the

following configurations for your launch template:

a. What would you like to do? Create a new template

b. Launch template name: scaling-template

c. Template version description: Launch template for my Auto Scaling

lab!

7. Next, select an AMI ID by clicking the Search for AMI link to the right of this

configuration field.

a. For the AMI Catalog, select Quick Start

b. For the AMI, select the Amazon Linux AMI.

You are selecting an Amazon Machine Image (AMI), which provides the
information required to launch an instance. It is a template for the root
volume for the instance that can contain an operating system,
application server, and applications. You use the AMI to launch an EC2
instance, which is a copy of the AMI running as a virtual server in the
cloud.

AMIs are available for various versions of Windows and Linux. In this lab,
you are going to launch an instance running Amazon Linux.

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 6
 8. Next, select the instance type to be t2.micro.

EC2 offers a wide selection of instance types that are optimized to fit
different use cases. For example, there are memory optimized, compute
optimized, and storage optimized instances, among others. For this lab,
you are going to use a t2.micro instance, which is a general purpose
instance. This instance type is also covered under EC2 Free Tier pricing,
which includes 750 hours of Linux and Windows t2.micro instances each
month for one year.

9. Your configurations up to this point should look like this:

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 7
 10. Now, you are going to create a security group. Open up the AWS Management

Console in a new tab and on the Services menu, click EC2.

11. On the left navigation pane, find NETWORK & SECURITY and select Security

Groups.

12. Click Create Security Group and enter the following configurations:

a. Security group name: scaling-lab-sg

b. Description: Security group for my Auto Scaling lab!

c. Add an Inbound rule with Type HTTP

Your final configurations should look like this:

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 8
 By creating this security group and adding it to your Launch Template,
you are allowing inbound HTTP traffic on port 80 to the EC2 instances in
your Auto Scaling Group so that you can access them through your web
browser.

13. After this, go back to the first tab where you are configuring your Launch

Template.

14. Hit the refresh button next to Security Groups. Find and select the Security

Group that you just created.

15. Scroll down to Storage Volumes and select Add new volume and configure:

a. Device Name: /dev/xvda

b. Size: 8

c. Volume type: General Purpose

d. Delete on termination: Yes

In a production environment, you might want to add persistent block

storage based on your requirements. You can specify an EBS volume

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 9
 type, size, and termination requirement. You can also enable encryption
on the EBS volume.

16. Under Instance Tags, select Add Tag and configure:

a. Key: Name

b. Value: Scaling Lab Instances

You can add tags for better organization and management of your AWS
resources.

17. Your configurations up to this point should look like this:

18. Expand Advanced details

a. Monitoring: Enabled

Here, you are enabling CloudWatch Detailed Monitoring. By default,

your instance is enabled for basic monitoring with a 5-minute period for
the instances. You can optionally enable detailed monitoring for an
added cost, however 10 Detailed Monitoring Metrics at a 1-minute
frequency are covered under the Free Tier. After you enable detailed
monitoring, the Amazon EC2 console displays monitoring graphs with a
1-minute period for the instance.

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 10
 b. Enter the following user data as text:

#!/bin/sh
yum -y install httpd php mysql php-mysql
chkconfig httpd on
/etc/init.d/httpd start
cd /tmp
wget http://us-east-1-aws-
training.s3.amazonaws.com/self-paced-lab-
4/examplefiles-as.zip
unzip examplefiles-as.zip
mv examplefiles-as/* /var/www/html

When you launch EC2 instances, you have the option to pass user data
to the instance which can be used to perform common automated
configuration tasks and even run scripts after the instance starts. Here,
you are installing scripts needed for this lab when the instance is created
and launched.

19. Your final configurations for your Launch Template should be similar to this:

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 11
 20. When you are sure the configurations are correct, click Create launch template

and now you are finished creating your Launch Template!

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 12
Task 2: Creating an Auto Scaling Group
You have officially created a Launch Template, which defines what should be
launched. Now, it is time to create an Auto Scaling Group so that you can define how
many EC2 instances should be launched and where to launch them.

After you finish creating the Launch Template, follow these steps:

21. Go back to the Services menu and click EC2.

22. In the left navigation pane, find AUTO SCALING and click Auto Scaling Groups

23. Click Create Auto Scaling Group.

24. Select Launch Template and choose the one you have just created.

25. Click Next Step.

26. On the Configure Auto Scaling group details page, configure the following:

a. Group name: auto-scaling-lab

b. Fleet composition: combine purchase options and instances

This is a feature that lets you create a mixed Auto Scaling Group with
different instance types as well as different purchasing options. This
way, you can harness the power of multiple EC2 instance types to

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 13
 support your workloads. You can also cost optimize by having the option
to use On-Demand, Reserved, and Spot instances.

c. Try adding a new instance type by selecting a different type, like

t3.micro

d. Group size: start with 1 instance

e. Network: keep this the default VPC

f. Subnet: select the first subnet that appears

27. Expand Advanced Details.

Here you can see further configurations that you can make with your
Auto Scaling Group, like the ability to associate your group with a load
balancer. You can use an Elastic Load Balancer with your Auto Scaling
Group to help automatically distribute incoming application traffic
across multiple targets, like your EC2 instances. It can handle varying
load of your application traffic across multiple Availability Zones.

You can use an Elastic Load Balancer with your Auto Scaling Group to
create a fault tolerant and highly available environment. If you associate
an Elastic Load Balancer to the Auto Scaling Group, then all newly
instantiated instances will register themselves with that load balancer
and then it will distribute incoming traffic across those instances. For
now, you will leave this unselected and move on.

a. Select Enable CloudWatch detailed monitoring under Advanced

Details.

28. Your configurations should be similar to:

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 14
 29. Select Next: Configure scaling policies.

You are going to skip configuring scaling policies for now. You will do
this in the next section of the lab.

30. Continue on by selecting Next: Configure Notifications.

You can configure your Auto Scaling Group to send notifications to an

endpoint that you choose, such as an email address or even an SQS
Queue or SNS topic, so that you can create automations that work as
part of your larger workload. You can receive notifications whenever a
specified event takes place, including the successful launch of an
instance, failed instance launch, instance termination, and failed
instance termination.

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 15
 For now, you are going to skip this step.

31. Click Next: Configure Tags.

32. Add a tag and configure the following:
a. Key: Name
b. Value: Auto Scaling Group

33. Select Review and then Create Auto Scaling group and now you have
successfully created your Auto Scaling Group!

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 16
Task 3: Creating Scaling Policies
Now that you have created an Auto Scaling Group, you are going to create a scaling
policy so that your Auto Scaling Group knows when to add capacity to your
applications when they need it, and remove capacity when they don’t.

You can use AWS Auto Scaling to do this. This is a service that monitors your
applications and automatically adjusts capacity to maintain steady, predictable
performance. It offers different built-in scaling strategies that you can choose from,
as well as predictive and dynamic scaling. Predictive scaling uses machine learning to
predict future traffic, including regularly occurring spikes, and provisions the right
number of EC2 instances in advance of these predicted changes. Dynamic scaling lets
you define how to scale in response to changing demand by using target tracking,
which is a feature that lets you set a target value and specify a scaling metric.

34. To start, go to the Services menu and select AWS Auto Scaling

35. Click Get Started if prompted.

36. Under Find Scalable Resources, select Choose EC2 Auto Scaling groups and

select the one you have just created.

37. Click Next so that you can specify a scaling strategy and define:

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 17
 a. Name: scaling-plans

Here you can choose a built-in scaling strategy. There are ones
optimized for availability, cost, or both. You can also define your own
custom metrics for optimization. You can also enable predictive and
dynamic scaling here.

38. Select the Optimized for availability strategy

39. Disable predictive scaling for now and leave the rest of the configuration

details the same

The default settings are configured to maintain an average CPU utilization of

40%. Your resources will scale in and out as needed to achieve this.

40. Hit next twice and create the scaling plan. You have successfully created your
scaling strategy!

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 18
Task 4: Testing the Auto Scaling Group

Now that you have created your Auto Scaling Group and scaling strategy, you can test
it to ensure that it works correctly.

41. Hit Close to return back to the Auto Scaling Groups page under Auto Scaling
listed in the navigation pane on the left.
42. Select the Auto Scaling Group you have just created.

This brings up details about your Auto Scaling Group, like the active history,
scaling policies, instances, and allows you to monitor different metrics.

43. Select the Instances tab to see that your Auto Scaling Group is in the process
of spinning up an EC2 instance, as shown below.

44. Select the Instance ID, which will take you to the EC2 Management Console.

On this page, you can see details about specific EC2 instances, like the
public DNS name, IPv4 address, and status checks. Since you have
enabled detailed monitoring, you can also monitor metrics at a 1-minute

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 19
 interval under the Monitoring tab. These metrics include CPU utilization,
disk reads, disk writes, and much more.

45. Wait for the instance state to say running and status checks to say 2/2 before
moving forward.

46. Copy the Public DNS name.

47. Open a new tab in your web browser and go to that public DNS name. You
should see this:

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 20
 48. Click Generate Load. You might have to do this a couple times to ensure you
are keeping a constant load on your instance.

Now that you are maxing out the load on that EC2 instance to 100%,
your Auto Scaling Group should respond by spinning up more instances
to help support the increase in load.

49. Go back to the EC2 console, make sure your instance is selected, and click the
Monitoring tab so you can see CloudWatch detailed monitoring metrics.

It will take a minute for data to start populating in CloudWatch. Since

you are maxing out the load on your EC2 instance, your CPU utilization
should look similar to this:

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 21
 You can also view metrics in the AWS Auto Scaling page if you select
your scaling plan:

Based on the scaling policy you have set in your Auto Scaling Group,
your group should spin up a new instance to help support this increase
in demand.

This might take a couple of minutes. Refresh the EC2 instances page and
you should soon see a new instance spinning up automatically.

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 22
 You can also see this in the Auto Scaling Group page. If you look at the
details under the Active History tab, you can see that the new instance is
warming up. You can look at the Instances tab to see how many
instances there are in your group currently. The monitoring tab shows
you different metrics like group size, pending instances, total instances,
and much more.

Congratulations! You have successfully created an EC2 Auto Scaling Group!

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 23
Task 5: Clean Up
Now that you have successfully created your Auto Scaling Group and tested it to
ensure that it works, you can clean up your environment by spinning down these
resources.

50. First, delete your Auto Scaling Group by selecting your group, hitting Actions,
and then Delete.

Deleting your Auto Scaling Group also deletes all the EC2 instances
associated with it.

51. Finally, delete your Launch Template by going to the Launch Templates in the
left navigation pane, selecting the template, hitting Actions, and then Delete.

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 24
Appendix – Additional Reading
What is Amazon EC2 Auto Scaling?
https://docs.aws.amazon.com/autoscaling/ec2/userguide/what-is-amazon-ec2-auto-
scaling.html

What is Elastic Load Balancing?

https://docs.aws.amazon.com/elasticloadbalancing/latest/userguide/what-is-load-
balancing.html

What is Amazon VPC?

https://docs.aws.amazon.com/vpc/latest/userguide/what-is-amazon-vpc.html

Whitepaper: Building Fault-Tolerant Applications on AWS

https://media.amazonwebservices.com/AWS_Building_Fault_Tolerant_Applications.p
df

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 25