All About School - The Complete Education Forum and Classifieds

Amazon CloudFormation is a service that makes it easy to set up your AWS resources. When you define your resources in JSON or YAML, CloudFormation takes care of provisioning and configuring the AWS resources, including Amazon Elastic Compute Cloud (Amazon EC2) instances, Amazon Elastic Block Store (Amazon EBS) volumes, Amazon Virtual Private Cloud (Amazon VPC), and Amazon Simple Storage Service (Amazon S3). You can also use CloudFormation to create other AWS resources like Amazon Simple Notification Service (Amazon SNS) topics and Amazon CloudWatch alarms.

What are the benefits of using Amazon Web Services (AWS)?

AWS provides the largest and most reliable cloud computing platform in the world and uses Amazon's new elastic systems, scale, data distribution systems, and durability.

The DevOps environment can be difficult with all the application environments, versions of operating systems, code sets, variable behaviors on different hardware, intermittent network connectivity and disconnected datasets among other headaches. 

As IT budgets get tighter organizations are being tasked with managing more frequent releases of their applications or doing more operations by themselves. Now for software developers this means planning projects better or adjusting how they deliver applications so as not to bring the company down because of an unexpected outage or system failure - but this can't happen when security needs are also increasing due to global cyber-attacks.

What if I don’t like coding?

You don't need any coding experience for the basic setup and maintenance of your Amazon Web Services. You can just point and click with simple templates that are already made for you in the Amazon CloudFormation console.

 Your resources are created, updated, or deleted automatically whenever you save changes to your template file without you having to do anything extra or different during those steps. The templates themselves are JSON documents where the contents map one-to-one with the properties and elements in AWS services such as EC2, S3, RDS, etc.

How do I get started with AWS?

CloudFormation provides configuration management and orchestration capabilities for architecting and automating the delivery of AWS cloud infrastructure with one stack, rather than managing discrete sets of resources. With minimal upfront engineering effort, you can create an entire logical data center within minutes.

*When you first deploy a stack in CloudFormation, every resource in the stack is deployed at once with minimal human interaction or waiting time; there's no need to deal with configuring instances one by one. The structure of your account tree (for example, the EC2 family) dictates how stacks are created and deleted so that all IAM resources in your account are replaced with the same type when new stacks are created from templates.

Where do I start?

The AWS console can be used to create or modify infrastructure, however creating large and complex stacks might become tedious. To work around this Amazon offers the command line tool aws cloudformation. This tool works as a way for automation of workflow using templates (simple JSON documents) throughout the stack creation process.

You firstly need to do some manual setup before you can use the aws clodform command: create an IAM user, grant appropriate permissions, and configure S3 bucket for storing template files. The following are steps from initializing a new stack: 

1. Create a template with desired configuration 

2. Configure lifecycle actions for what should happen when the stack is created, updated, deleted 

3. Use list-stacks to find and select the desired stack 

4. Create Stack from Template using `aws cloudformation deploy` with various flags such as --stack-name=LAMP_STACK_NAME -f LAMP_TEMPLATE_FILE

The Value of Autonomous Systems

For example, an autonomous Uber car could go from picking the user up at their home to dropping them off at the destination and simultaneously optimizes for fuel efficiency, number of people in the car, distance of travel and other parameters. 

By taking on this kind of orchestration and coordination responsibility at the micro level, companies can deliver higher quality service or drive more cost-efficient outcomes while also limiting some vulnerabilities or system risks. One major area where Amazon Web Services (AWS) has been a pioneer is in driving automation as applied to its own infrastructure with AWS Auto Scaling and Auto Healing services.

How Is an Auto Scaling Group Different From an Instance Group?

An Auto Scaling Group launches and terminates instances as needed using defined capacity settings, whereas an Instance Group provisions one or more instances with the same Amazon Machine Image (AMI). If you need more computing power temporarily, you can use Auto Scaling Groups for short-term needs. An Instance Group is useful if you want to create a number of instances all at once, but don't want them to be terminated automatically.

 For example, in order to test the performance of some new features, you might want to create 100 web servers in parallel. You could do this by creating an instance group with 100 EC2 instances running from the same AMI. The instance group will terminate these EC2 instances when they are no longer needed; however, since this process is done manually by an administrator there will be a delay between when these EC2 instances are no longer needed and when they are actually terminated.

Are There Any Limitations on Auto Scaling Groups?

Auto Scaling Groups provide you with the flexibility of scaling groups in or out dynamically and continuously, providing extra security by avoiding overprovisioning and underutilizing instances, as well as increasing agility by reducing waiting times for new instances. However, Auto Scaling Groups are subject to some limitations which include: Multi-AZ deployments can't be scaled in response to CPU demand; load balancing through ELB cannot be done for ASGs; ASGs only have one AZ configuration; and changing instance types will change the instance size from T2 to T1 (at no cost).

How Does Cross-Account Auto Scaling Work?

Cross-account auto scaling enables the creation of a single auto scaling group which automatically scales applications across all the accounts associated with that group. In other words, you don't have to worry about configuring or managing scaling in any of the accounts when using cross-account auto scaling - all the configuration and management happens in one place. 

There are two ways to create an Auto Scaling Group for use with cross-account auto scaling: either use the CreateAutoScalingGroup API call, or start by creating an Auto Scaling Template. Once created, you can associate this group (either created manually or from a template) with any number of AWS accounts. Then configure individual targets for each account within the Auto Scaling Group so that if there are unmet needs on some accounts, requests will be scaled out to other accounts as needed.

Final Thought

I would like to share some of the lessons I learned about Amazon CloudFormation from this new challenge in my life. The biggest lesson is the value in planning and how beneficial it can be, even if the plan changes significantly throughout execution, which mine did. Second, I found that reviewing documentation and tutorials often helped me understand what I was working on when struggling with a certain task. 

Finally, repetition proved very helpful because after doing something more than once you get used to doing things without having as much of an initial learning curve associated with them. Lastly, utilizing tools such as CloudFormation or other cloud-based automation tools help streamline your process so that you are able to spend less time managing those resources and more time focusing on your applications running in AWS.