AWS Partner Network (APN) Blog

Using TSO Logic Data Analytics and Data Modeling to Demonstrate Cost Advantages of AWS Cloud Migration

By Aaron Rallo, CEO, TSO Logic.

This is a guest post from TSO Logic, an Advanced APN Technology Partner and AWS Migration Competency Partner.

Let’s say you’re considering migrating to the AWS Cloud, but first want to get a solid grasp on the economics. Maybe your team hasn’t had the time to identify how much compute you actually are using so they can evaluate potential migration costs. Or maybe your team is running numbers based on limited, static data and they have concluded that migration would be more expensive. If you are planning a migration to AWS then take a look at the solutions offered by TSO Logic. TSO Logic is an Advanced Technology Partner and AWS Migration Competency Partner whose platform can help you analyze the performance and financial characteristics of every workload you are running.

Over the past few months, while considering large-scale data center migration to AWS, six customers used TSO Logic to discover the scope of their environment and analyze utilization patterns. The outcome was a data-driven business plan and cost model for cloud to fast-track cloud transformation. The following sections show how TSO Logic was used to analyze the real-time data from these customers with a focus on compute usage, utilization, and instance sizing.

Let’s take a look at some insights from the analysis for the initial cost-modeling done with TSO Logic for these six customers.

High-level Insights and Takeaways

The first step to cloud migration is to have a good plan in place, but it’s important to note that cost-modeling should not be solely a one-time exercise. Customers who migrate to AWS are able to continue to optimize and save costs long-term because of their ability to continuously innovate and take advantage of new services and technology (such as building a serverless architecture, improving automation, and so forth) that enable them to continue reducing their total cost of ownership.

Insight from the TSO Logic Analysis: Common Oversights When Initially Calculating the Cost of the AWS Cloud:

  1. As previously noted (but well worth repeating), cost modeling is not a one-and-done exercise because the variables, such as compute patterns, applications, and cloud service catalogs, are always changing. Modeling on a consistent basis will ensure that you are always getting the most compute for your investment.
  2. Overprovisioning happens. For many enterprises, compute was intentionally overprovisioned to meet the demands of unexpected spikes in workload. With the elasticity of the cloud, there is no longer a need to overprovision to that extent. It is important to take this into consideration when creating on-premises to cloud cost models.
  3. AWS is continuously innovating so you don’t have to. Hardware is frequently updated and new instance families and types are often being added. When you compare your current on premises costs to cloud costs it’s important to consider the differences in compute power.
  4. Rightsizing is key to an accurate model. When the rightsizing of environments was accurately accounted for (e.g., for utilization levels, processing power, etc.) with the six customers who used the TSO Logic Platform, AWS migration led to at least a 26% reduction in annual cost savings for compute and up to 60% savings compared to on-premises. This includes hardware amortization, maintenance, OS licensing, and facility (it doesn’t include savings related to labor).

A Real-World Field Study:
Leveraging the TSO Logic Platform to Analyze and Rightsize On-Premises OS Instances

The Raw Numbers

AWS/TSO Logic Customers: 6
On-premises OS Instances Evaluated: 33,936 Operating Systems Instances
Current On-premises Costs: $58,224,000/yr.
TSO Logic Rightsized to AWS: $42,816,000/yr. (26% savings over on-premises)

The Findings Using the TSO Logic Platform

  • Rightsizing for economic sense. When cloud instances can be rightsized based on historical usage and utilization patterns, TSO Logic demonstrated that AWS is more economical and the customers whose data was analyzed could save 26% or more over their current costs.
  • Older hardware, server refresh. When TSO Logic looked at the 15,270 instances that were running on hardware that was more than five years old, it was determined that migrating these instances to the AWS Cloud could reduce costs by 74%. Overall, one of the most immediate economic benefits of migration comes from instances that are running on older hardware — the older the hardware on premises, the more economical the cloud becomes according to the analysis done. Cloud also has the added benefit of reducing upfront capital investments in new hardware. In addition, since AWS is always innovating and releasing new instance types, you get the benefit of the latest hardware without any additional capital expenditure.
  • Environments. When TSO Logic looked at the 2,883 instances in the sample set running in test-dev environments on premises, it was determined that migrating would save $2.8 million/yr. for these customers — a minimum savings of 42%.

The Bottom Line

To accurately evaluate the cost of cloud migration, you first have to know the actual historical usage and utilization patterns of your workloads and your true provisioning needs. Take a look at how TSO Logic’s solution can help you as you look to understand your business case for migration and how you can take advantage of the cost benefits of running on AWS.

Want to learn more? Learn how TSO Logic can help you plan your migration to AWS during our upcoming webinar. Register here.

About the Data

The economic models for rightsizing and right costing the environments detailed in this post were created using data points ingested into the TSO Logic agentless software platform. Data included provisioned compute and historical usage and utilization data, along with a pre-existing benchmark catalog of on-premises costs. The models identified the direct-match and the rightsized options along with the associated costs from the AWS catalog.

The content and opinions in this blog are those of the third party author and AWS is not responsible for the content or accuracy of this post.