Amazon Web Services is available to use all over the world.
It has it's data centers in 24 different regions of the
world and adding more every year. All of the regions that
AWS provides contains multiple Availability Zones, which is
a group of data centers within these regions.
All of the zones are completely independet of another and they are isolated, they also have their independent cooling and power source, security and network - which makes the fault-tolerance sky-high.
So, the question arises, which AWS region should you choose for your work and why should you even select one in the first place? Should you choose the one nearest to you, or is it not that simple? Well, it depends on your requirements. Let's see what's the difference between the AWS region and how you should choose one.
Although some AWS services. such as S3 are available in all regions, most of the services are region-dependent. Therefore, if you want a specific service, you must check beforehand its availability in different regions. Amazon provides the list of services supported by each region on its website.
Another aspect you should look out for is the cost of
computing. The cost varies between different regions. For
example, the cost of computing in the US regions is almost
15% cheaper than in the EU regions.
Therefore, if your workload requires long computing hours, make sure to check which regions have the best pricing alternative for you. This would save you thousands of dollars by choosing the right region. You can find out the estimated pricing with AWS calculator which is a great tool to get an estimated cost.
Here's one example of two price comparisons between the hourly cost of a hourly cost of an EC2 in California compared to Sao Paolo, which has a 26% higher hourly cost!
That's not enought to consider which service you want to
use. Even though you've found a region that provide the
services you need and the pricing that you prefer, there is
one more thing to examine - the region latency to each
region. First of all, you need to figure out the
geographical location of your target audience. Then make
sure to select the AWS region that is closest to them.
This will reduce the latency significantly and provide users with a much better experience. There are of course other solutions such as Cloudfront to solve latency, but not for all services. You can also use websites such as CloudPing to check the latency to different AWS regions.
Even though there are a couple more aspects to choosing an AWS region, you should be good to go if you only consider the three factors mentioned above to have a good grasp of which regions you should choose!