Cloud Regions: How to Make the Smartest Choice

One of the advantages of cloud computing is that public cloud vendors offer dozens of cloud regions to choose from when you are deciding where to host your workloads. But that can also create a challenge in that you have to figure out which cloud region (or cloud regions) is best for your needs. Here’s a primer on how to decide.

What Are Cloud Regions?

A cloud region is the geographic area in which a cloud data center is located. Public cloud providers maintain data centers in a number of different locations and allow customers to choose among them when deploying a workload.

In fact, not only can you choose from among different cloud regions, but you have to. In other words, the cloud provider will require you to select a specific region when you are deploying a workload.

Why Do Cloud Regions Matter?

The main reason why cloud regions are important is that the closer your users are to the data center where your workloads live, the better the user experience will be. It’s harder to optimize page load time when your cloud region is geographically distant from your end users.

Selecting the right cloud region is also important because the cost of many cloud services varies depending on which region your workload is hosted in. For example, AWS’s S3 data storage service costs $0.025 per gigabyte (for the first 50 terabytes on the standard storage tier) in the AWS Hong Kong region, compared to $0.025 in Ohio.

The cloud region you use can also have consequences for compliance, reliability and more, as explained below.

Factors to Consider When Selecting a Cloud Region

Many companies default to choosing to host their workloads in whichever cloud region is closest to their headquarters. But that approach is not always ideal.

Instead, weigh the following considerations when selecting from among cloud regions.

1. Where are your end users?

If most of your end users are located in a specific region, hosting your workloads in the cloud region closest to them is the obvious thing to do. It’s a key step toward optimizing performance.

Of course, if you are serving users who are spread across multiple geographical areas, you’ll need to consider other factors when selecting a cloud region.

2. Do you have data sovereignty requirements?

If compliance rules or internal data privacy policies require you to keep data within a specific geographic jurisdiction, you’ll need to select a cloud region that meets that need. This is a situation where the decision about which cloud region to use is more or less made for you.

3. Where are your other workloads?

If the workload you are deploying in one cloud region needs to integrate with or connect to workloads running on-prem, in a different cloud or in a different cloud region, that is a factor to consider as well. In general, the closer your various workloads are in a geographic sense, the better the overall performance will be.

For example, if you are building an application that will be accessed by users in Japan but that will need to ingest data hosted in a private data center that you own in the eastern United States, you may want to choose a cloud region that is halfway between those points. Choosing a cloud region close to Japan may not deliver the best overall performance because it will take longer to move data from your data center to the Japan-based cloud region.

4. What are your SLA needs?

In certain cases, the service level agreement (SLA) that cloud providers offer for a cloud service differs between cloud regions. If the availability guarantees of SLAs are a key priority for you, check whether you can obtain better SLAs in one cloud region than another for whichever cloud service or services you will be using.

5. Which cloud features do you need?

The features available from cloud services may also vary between regions. For example, not all AWS EC2 instance types are available in all AWS regions. Sometimes, an entire cloud service may not be available at all in a given region.

Make sure, then, that the specific configuration or functionality you require from a cloud service is supported in the region you intend to use.

6. Which region costs the least?

As noted above, costs can vary somewhat between cloud regions. Comparing prices between regions for the cloud services you intend to use can go a long way toward optimizing your cloud costs.

7. How many availability zones do you need?

Public cloud providers divide each of their cloud regions into multiple availability zones. (Some clouds call them just zones.) An availability zone is an isolated data center within a given cloud region. Although you don’t have to use more than one availability zone, some organizations choose to do so to increase the reliability of their workloads. If one availability zone fails, the workload will remain up as long as it is mirrored on a second availability zone.

All cloud regions should offer at least two availability zones, but some offer more. If you want to use more than two availability zones, select a cloud region that supports that approach.

Using Multiple Cloud Regions at Once

If you’re having trouble committing to a single cloud region, remember that there is nothing stopping you from using more than one cloud region at one time.

You can host some workloads in one region while running others in another region within the same cloud. That approach can work well if you need to cater to user bases that are concentrated in two distinct regions.

Likewise, if one of the cloud services you need to use costs less in one region, and another service is cheaper in a different region, you can run each service in whichever region is most cost-effective.

Just bear in mind that using multiple regions to improve reliability is usually not a cost-effective strategy. Use multiple availability zones for that purpose.

Conclusion

Choosing the right cloud region is important for optimizing costs, performance, reliability and more. Instead of defaulting to whichever region is closest to you or whichever one your cloud provider suggests, do your research to determine which region (or regions) will provide the overall best value and performance.