Content Domain 3: Cloud Technology and Services

Task Statement 3.2: Define the AWS global infrastructure

Let's get started with the second task statement to define the AWS global infrastructure. Understanding the concepts behind the AWS global infrastructure will help you know how to connect, design, build, and deploy your architectures in AWS. Remember, we mentioned the AWS global infrastructure and its components such as Availability Zones, Regions, and edge locations in an earlier lesson. For this task statement you need to ensure you understand how they function, both individually and together, what choices do you have, and how do you determine when services use a specific Availability Zone, Region, or edge location? Let's dive a bit deeper than we dove in our earlier lesson. AWS offers globally resilient services, regional resilient services, and zonal resilient service. A globally resilient service is a service that operates globally, for example, a single database, and that data is then replicated across AWS Regions. So a Region can fail, but the service will continue to run. IAM, CloudFront, and Amazon Route 53 are examples of globally resilient services. Region Resilient services operate in one Region, with one set of that data in that Region, and then replicate their data to multiple Availability Zones inside that Region. If you lose one Availability Zone in that Region, the service will continue to operate, but if the whole Region fails, then so does that service. Amazon EFS and AWS Batch are examples of regionally resilient services. Availability Zone resilient services are run in a single Availability Zone, so if that Availability Zone fails, then so does that service. So an important exam tip is to know what level of resilience an AWS service has because it will help you answer exam questions, but also give you a much deeper understanding. Amazon EBS is an example of an Availability Zone service. There are also zonal services, such as Amazon EC2, that are tied to a particular Availability Zone. So what is the AWS global infrastructure? Well, AWS created their global infrastructure as a collection of individual infrastructures located all over the world. So globally, AWS offers Regions and also edge locations, and AWS is designed to have multiple Regions, multiple Availability Zones, and multiple edge locations around the world. The AWS Global infrastructure is made up of AWS Regions around the world, and then on top of the infrastructure, we have a high level of services. So you will see there is a level of compute, and examples of compute are Amazon EC2 and AWS Lambda. There's a level of storage and examples of AWS storage services are Amazon S3 or Amazon EFS. There's a level of database services such as Amazon RDS, Amazon Aurora, Amazon DynamoDB. There's a level of migration and transfer, network and content delivery, developer tools, robotics, blockchain, satellite, management and governance, machine learning, security, identity and compliance, application integration, cost management, and more. And an AWS Region is a geographical area that consists of two or more Availability Zones, and all of the Regions are under that Region's rules. If your data sits in a particular Region, your data will not leave that Region unless you specifically move that data. AWS designs their Regions to be geographically spread across the world. This way you can use different Regions to design your infrastructure that can withstand global disasters. So there is separation between Regions so each Region is fault tolerant. Here's an exam tip and one for the real world too. When you are selecting Regions for your applications in AWS, remember to check if there are compliance requirements and also choose the closest Region to reduce latency for your users. Regions also have Local Zones which are an extension of an AWS Region in closed geographic proximity to your users. AWS Wavelength is similar. It helps you to build applications that deliver ultra-low latency to mobile devices and end users. Wavelength has a wavelength zone which is an isolated zone in the carrier location where the Wavelength infrastructure is deployed. Wavelength zones are tied to an AWS Region and are extensions like the Local Zones. An Availability Zone is one or more data centers with redundant power, networking connectivity and so on. And the data centers are in separate facilities inside different AWS Regions. AWS has multiple data centers across the world. You can think of an Availability Zone as one or more data centers. And a data center is a building that's filled with servers, sands, switches, load balancers, firewalls, storage and more. You can have more than one data center in each Availability Zone, and each Availability Zone is an isolated compute storage network, and so on. You can distribute your infrastructure across multiple Availability Zones in your Region for high availability. So if one Availability Zone fails, then your other Availability Zones should remain operational because they are isolated from each other, but they are connected with high speed redundant networking. Your services can be placed across multiple Availability Zones to add resilience and high availability. And an AWS edge location, again, is a global service, and it is an endpoint for AWS that is used for caching content. AWS has a content delivery network or CDN, which is CloudFront. And CloudFront offers, that if a user requests certain information, that information is then cached at the edge location, so the next time another user requests that same information, that information is already available and delivered to the user much faster than if you had to go all the way back to the database and search for that specific information. An example of this is how Netflix uses edge locations to store content as close to their customers as possible. So customers get very low latency when requesting certain movies and shows. The further the data that is being requested is located from your customers, the slower the transfer of that data becomes. Another service of focus here is AWS Global Accelerator, which is a global service that supports endpoints in multiple AWS Regions to improve the performance of your applications for local and global users. For your exam, know the difference between CloudFront and Global Accelerator, because they both use the AWS Global Network and edge locations. Both also integrate with AWS Shield for DDoS protection. However, CloudFront will improve content for cache content, both static and dynamic content, and content is then served from that edge location most of the time. Global Accelerator improves the performance for applications over TCP and UDP because the packets are being proxied from the edge locations to the applications running in one or more Regions. So all the requests that are making it to the edge, but there is no caching available. Global Accelerator is great for HTTP use cases that need a static IP or need fast regional failover. Make sure you understand the differences between a Region, Availability Zone, and edge locations, because you will most likely see questions on these on your exam. Also ensure you know the models for cloud. For example, cloud computing has infrastructure as a service, platform as a service, and software as a service, but there is also database as a service and more. Next, look at the components separately. With Availability Zones, study which services are bound by their borders, and how the Availability Zones can help you when building a highly available environment. Also, look at how the Availability Zones come into play when you're determining points of failure in your architectures. Additionally, how is communication handled when going from one Availability Zone to another? This is gonna differ based on the services, so you'll need to look at which services deal specifically with the Availability Zones, and then how they connect and communicate across those Availability Zones. When considering Regions, you wanna look into a lot of the same stuff as you will with Availability Zones, but on top of that, you should look into ways to use Regions for disaster recovery and business continuity. What can you do to avoid catastrophic failures and what services help you utilize your use of the Regions? Also, understand why you would use the different Regions and what benefits you might find from the use of a particular Region or the use of multiple Regions. Lastly, study how Region usage of affects your compliance requirements. Oftentimes, data management is determined by sovereign borders. It is important that you know how such laws and requirements will affect your usage of the AWS Cloud. The final category for this task statement and for the global infrastructure you need to be mindful of is edge locations. Look into what services take advantage of these edge locations and how those services differ in their utilization. Can you choose which edge location to use? How are edge locations utilized by CloudFront? What about for AWS Global Accelerator? Do the edge locations come into play only with certain services, or are there other times when you're using them? What benefits are provided by the edge locations, and what level of management can you dictate? Availability Zones, Regions, and edge locations are foundational components to understanding and utilizing AWS. Understand how they help and affect your architectures, what you can do to take advantage of their benefits, and the services that are bound or directly interact with these components. Let's get started with the third task statement and talk about AWS compute services.