Why Data Center Location Matters

"Where do you want your data housed? Please choose from the list below. You can't migrate data between regions."

When I read those words, I was signing up for a SolarWinds® Observability SaaS trial. I froze, and I didn't know where to start. If you are similarly perplexed, let me share what I learned.  

Location Matters  

Cloud platform providers like Amazon Web Services (AWS) offer multiple geographic regions for hosting workloads. Each region contains numerous Availability Zones (AZs), the physical data centers where platform providers cluster data centers. Data center location matters for regions because it can impact performance, cost, and AWS service availability. Understanding the tradeoffs can be helpful when selecting the best region to support your organizational goals. 

The Proximity Priority  

You can optimize your region selection according to your specific business objectives. If your organization is concerned about service delivery to a specific group of users, you can select a region to help you best serve that target audience. You are not required to select the region where you or your organization are physically located.  

There are some use cases where choosing the closest region to you would make sense. For example, selecting the region closest to your data center works if you plan to use SolarWinds Observability SaaS to monitor hosts running on-premises and in a software-as-a-service (SaaS) environment. It would reduce latency between the host agent and SolarWinds Observability SaaS dashboards, metrics, and alerts.  

If on-premises host monitoring is not an issue, a general rule of thumb is to select the region closest to your monitoring and operations team. Doing so reduces any lag between data collection and performance metric generation in SolarWinds Observability SaaS. A region close to your monitoring and operations teams provides a near real-time view of performance and availability, enabling them to respond more quickly to any emerging issues. 

The general guidance is to select a US region if you are part of a large global organization and the stakeholders are distributed across regions. US regions are the most connected to other regions, which helps reduce latency and provides the best overall experience. 

Resilient Observability 

Suppose your primary business objective is to ensure the resilient observability of your environment. In that case, consider deploying SolarWinds Observability SaaS to a region other than the region where your cloud resources are located. Disasters can be one of the most complex challenges to plan for. Weather events, including tornados, floods, and ice storms, can impact the availability of the data center resources supporting an AWS region. Unplanned or unauthorized changes to data center resources can also prevent a workload from running, which would be another type of disaster.  

For example, a 2023 fire at a Paris data center brought down the entire  Google Cloud Platform (GCP) europe-west9 region. A cooling system water pump failure caused water to leak into the battery room, which sparked a fire.  Though the fire was limited to a single room at one of the two buildings at the data center, the water leak caused a multi-cluster failure that impacted the entire GCP europe-west9. The GCP europe-west9 region was offline for three weeks. 

If you were running SolarWinds Observability SaaS and your cloud resources out of the GCP europe-west9 region, your service delivery to your users and your ability to monitor your resources would have been impacted.  

Cost Considerations 

There can be significant cost and feature availability differences across regions. Factors like customer demand, infrastructure, energy, and labor costs can result in some regions being more expensive than others. The cost of the same instance configuration with the same data input and output volume across 20 AWS regions can vary as much as 50%. AWS provides rich guidance on price differences in their documentation. Other pricing variables, such as the instance type, configuration, on-demand, or reserved, can also vary from region to region and impact the price.  

AWS typically rolls out new features slowly across regions, with larger regions being the first to receive the latest capabilities, so not all services and features are available in every AWS region. The differences in AWS features offered in each region could be of particular importance if you have specific security or compliance concerns or are considering locating your SolarWinds Observability SaaS application in the same region as a workload requiring advanced hardware, such as NVIDIA L40S Tensor Core GPUs. For example, the EC2 G6e instance type is only available in AWS US East, US West, Asia Pacific, and Europe (Frankfurt, Spain) regions as of November 2024.  

Clearing the Confusion 

SolarWinds Observability SaaS is available in multiple AWS regions, and selecting the region that best aligns with your overall needs is essential. Once you have determined your organizational priorities and evaluated the impact of the AWS region selection on each, you should be able to make an informed choice. When you’ve selected a region and started your trial, you’re on your way to enjoying unified and comprehensive visibility across cloud-native, on-premises, and hybrid custom and commercial applications. 

SolarWinds Observability SaaS offers a holistic performance view across the stack. Operations teams can efficiently manage and optimize their IT resources with near real-time performance metrics, anomaly detection, and smart automation in a single pane of glass. Teams can proactively identify bottlenecks, plan capacity, and identify issues before they result in poor user experiences.   

Ready to experience the power of full-stack observability? Start a free 30-day SolarWinds Observability SaaS trial and unlock critical insights into your system health, performance, and availability today.