How does AWS manage its Global Infrastructure? | PART 1

How does AWS manage its Global Infrastructure? | PART 1

Today People access the Application all over the world, from Africa to Europe and Asia to America. AWS needs to provide proper functionality of the application in all the regions with the lowest latency possible. But How does AWS achieve this?

Discussing all the Topics of AWS Global Infrastructure cant be covered in one article, so in this article, we shall discuss AWS regions and further information in upcoming articles

Why We Need A Global Infrastructure?

                Suppose we are running a Web Server in France, and most of the traffic comes from France users. But if somehow your website got popular and people from other regions or even countries want to use the site does your server can deliver data with low latency? this problem is related to high latency delivery. Another major problem can be if you have deployed all your servers in a single location, during times of disaster or power outages it will be a tremendous loss in the business. This is the common problem which we need to solve for cloud architects, and AWS has solved this problem with its global infrastructure. For more clarity let's start with what does global infrastructure means in cloud computing.

For more Updated Regions Visit: AWS Global Infrastructure

                As we discussed earlier Disaster management and low latency delivery are 2 major factors that will affect the working, let's consider a day-to-day example to solve this problem. Suppose you run a hotel and some fine day there is a fair going on in the lane of the hotel, and there is no way for the customers to reach the hotel, so in this situation most of the customers might drop the idea and goto other hotels, the same happens in the case of website the users will skip to the next website if your site is not working properly or delivering with high latency. But if there are multiple branches of the hotel the traffic can be diverted to the other hotel which is nothing but a global infrastructure, in a small region, we as a hotel managers can do this in the city, but AWS does this all over the regions. AWS runs multiple data centers all over the world to provide low latency, data delivery. We can understand this Global Infrastructure by the following terms.

AWS Regions

                In the hotel example, we have deployed multiple branches in the city for dividing the traffic, but there is one more advantage which we get by doing this which is if one branch has some power outage or was unable to provide service traffic can be diverted to another branch, in the same way, if data centers face some issues the request can be transferred to the next data center. A data center in a more broad way is called an AWS Regions. AWS Regions are spread across the globe with 2-3 regions in each major country. Each region has its own power supply and is geographically separated by huge distances. The terminology of naming the regions is mostly based on the directions example us-east, us-west, us-north, etc. Each region again has 2-3 availability zones, which we will discuss in the article later.

There are also some Major decisions we need to make before choosing the region we want to deliver the web application or service, they are

Proximity Range for the Customers.

                The Regions which you choose should be nearer to the location where the customers are, for example, if the users of your company are mostly from France then you should select France region and not Spain, Italy, or other regions which will increase the latency for the customers. There is also a concept of edge location where we transfer data to other countries or continents if your web applications are on a huge range, we shall see about edge location in the next part of the article.

Should Check the Available Services

                The point to be noted about AWS is all the services provided by AWS may not be available in all regions, so it's very important to think about the regions also while using the service. Suppose you use some advanced Deep learning services which are not available in the current region it will be a big issue. But AWS is growing and it's increasing to provide its services all over the regions

Cost and Pricing

                In all regions, the pricing system might not be the same, for example, the prices of AWS services in brazil are more compared to other regions so it's necessary to see the pricing plans also before selecting the regions so we will be not paying more. There are multiple AWS pricing services also which we will be covering in upcoming articles.

Legal Requirements and Data Governance

                This is not as much as important other decisions if the users of your web application are from certain regions then it's best to deploy the website in that region. Which will reduce the cost, latency, and other requirements. If the users are from different regions of the world then we need to consider the above 3 decisions for further deployment.

Post a Comment

Previous Post Next Post