- [Seph] To look at how the technology topics fit into the real-world utilization of AWS, I'll take you through a few of the topics individually. To start, let's look at the deployment and operational aspects of AWS. Imagine a group of components that are all part of a larger architecture. You have components like a load balancer, some web servers, some application servers, and your backend databases. It's a fairly typical setup, but there are several things to consider with this. First is the how, that is, how do you manage the various components? Yes, the AWS Management Console can be of some help, but it makes it very difficult to document processes with the interface updates. And automation of tasks become nearly impossible. What if you need to perform operations on several servers at once? In the Management Console, the operations you can distribute like that are minimal, at best. This is where it would be important to understand your other options. If you needed to manage your infrastructure as a whole, the use of the Management Console will almost be a hindrance. So, what options would you use? The SDKs? Or maybe the AWS Command Line Interface? Or maybe even a service like AWS CloudFormation, or another infrastructure as code option? Well, it will depend on what needs to be done, and how it needs to be handled. You could just be using the CLI, or in actuality, you would likely be using a mixture of all of these. And once you understand how you'll be handling things, you start looking into the where. The AWS global infrastructure, among other things, helps you to determine where you'll be building. It helps to ask questions like, where are my end users? Do I have any data-sovereignty regulations to consider? Will caching content improve my end-user experience? How will I handle high availability and disaster recovery? These and many other questions are answered through understanding of the AWS global infrastructure. Keeping the previous architecture in mind, how would you handle disaster recovery for that application? Do you need high availability options within the same Region, or maybe failover options in another Region? Could you need both? How would the use of edge location services help with your customer impact, should you encounter an event? Those are just some of the starting questions. Knowing how to utilize the AWS global infrastructure will provide many of your answers, and go a long way to clarifying your needs. Last is the what of your architecture. This consists of the actual tools and services you'll be implementing in your builds. What combination of components will help you meet the goals you've set? If you're migrating, what are the needs from your on-premises environment that you have to meet, and which services can help with that? Let's just look at the database needs from the application architecture I mentioned earlier. One option you have is to install and run your database engine on an Amazon EC2 instance yourself. This will give you full access and control to the instance, the database engine, all configurations, and complete design power over your database setup. The downside is that this will require much more operational overhead. You would be responsible for OS and database-engine updates, high availability mechanisms, downtime management, and data consistency. The other option could be to use a managed service, such as Amazon RDS. This managed service automatically patches the database software and backs up your database, stores backups for a user-defined retention period, and also supports point-in-time recovery. You benefit from the flexibility of being able to scale the compute resources or storage capacity associated with your relational database instance by making a single API call. The other side to this is that you are limited in what database engines you can utilize, and you do not have root-level access. You maintain a lot of control, but not full control. Neither of these options is worse than the other, but they both have scenarios where one would be a better fit. Understanding what service you'll use will be where you spend a lot of your decision-making time. And it will also provide the most confusion, if you don't have a foundational understanding of the services available. I hope you see how important it is to build that foundational knowledge. While, yes, this will help you to pass the certification exam, it will also help you to best utilize AWS and speed up your ability to operate in the cloud. That's all I have for now. Cheers.
