Setup Autoscaling

• In real life you're load on the websites can change over time. So for example, say your users are doing shopping, they're most likely doing shopping during the day and not at night. So you expect more load during the day and less load during the night.
• So in the cloud we know we can create and get rid of servers very quickly and so the goal of an auto scaling group is to scale out. That means add EC2 instances to match an increased load or scale in, that means remove EC2 instances to match a decreased load.
• With this we can ensure that we have, also as well, a minimum, and a maximum number of machines running at any point of time and once the auto scaling group does create, or remove EC2 instances we can make sure that these instances will be registered, or de registered into our load balancer.
• So these two things work hand in hand. Finally, in case one of our servers becomes unhealthy, maybe there's an application bug, then the auto scaling group can detect it and say, yeah, you know what, I don't need an unhealthy instance. I'm going to de register it. I'm going to terminate it and replace it by a new healthy one.

So with an auto scaling group we get a lot of benefits and another benefit we get, is that we have huge cost savings, because we are only running all the time at the optimal capacity and that is one of the guiding principle of the cloud, elasticity.

Autoscaling mainly comprise of 4 important steps:

1. Launch an instance.

2. Create an image.

3. Launch Configurations

4. Autoscaling Groups

Launch an Instance

• Logging in to AWS, select EC2 in the AWS services. To launch a new configuration, click on?Launch Instances.

• Choose an Amazon Machine Image(AMI). (?Microsoft Windows Server 2016 Base)?to have Graphical Interface.

• Choose Instance type t2.micro. Click on?Next.

• We will launch only 1 instance and let Autoscaling group launch instances according to requirements. Click on?Next: Add Storage.

• By keeping?Add Storage?and?Add Tags?to default.

• In Configure Security Group, We will create a new security group named?demoASG, keeping Description to default. Choose?RDP?for remote connection to Windows system and?HTTP?and?HTTPS?to allow internet traffic to reach our instances. Select?Source?to Anywhere for all the 3 types as shown below.

• Choose a key-pair. Create a new key-pair and give it a name of your choice. Download key-pair and launch instances.

• Give name to the launched instance to avoid any confusion.

Creating Image

• Select the running instance and Click on?Actions?and select?Images and Templates?and?Create Image.

• Enter any name of your choice in Image name. I will choose demoASG. Keeping all other settings to default, Click on?create instance image.

• AMI image is successfully created.

Creating Launch Configuration

• In the left navigation Panel, under?Autoscaling?select?Launch Configurations. Click on?Create Launch Configuration.

• Give a Launch Configuration name and Choose AMI name of the one we created above. Here will select?demoASG.

• Select Instance type as?t2.micro. Keeping?Additional Configuration?and?Storage?to default, we will choose existing security group, the one we created while launching our instance.

• Use existing key-pair, same as one created while launching an instance. Click on?Launch Configurations.

Creating Auto Scaling Group

DEMO

• Enter?group name?and?Switch to launch configurations?and select the configuration name we created in Step 3. Click on?Next.
• Choose?VPC?as default and select all the available?subnets?to balance the load. Click on?Next.

• We are not using Load Balancers. In Health Check,?Health Check type?by default will be EC2 because we have launched an instance and will select?Health Check grace period?as 120 seconds.?Click on?Next.

• We will keep?Desired Capacity?2 as we want Autoscaling group to make 2 instances available to us.?Minimum Capacity?as 1, available instances should never be less than this size and?Maximum Capacity?as 5, number of instances should never exceed this size.

• Selecting?Target Tracking Scaling Policy?and using?Average CPU utilization?as Metric Size and setting?Target Value?as 50, an instance will be launched if CPU utilization exceeds Target Value. Instances need 120 seconds warm up before including in Metric. Click on?Next.

• Keeping other settings by default. Click on?Create Autoscaling Group.
• Autoscaling group is created. We will see new instances being launched. If we increase load on server, more instances will be launched according to need.

Connecting to original instance using RDP.

• Select the instance and click on?Connect.

• In RDP client, click?Get Password.
• Browse the key-pair you used while launching the instance and click on?Decrypt Password.
• Download the Remote Desktop file and copy the password.
• Open Remote Desktop file and proceed. Enter the copied password when asked for. Windows Server is connected to our system.

• To increase CPU utilization, create a .bat file and open the file multiple times. When CPU utilization increase to a limit, new instances will be launched.