HOW MUCH WOULD COST ME TO BUILD A SYSTEM TO RUN ON TWO CLOUD VENDORS SIDE-BY-SIDE?

 One common question that I hear lately is: 

HOW MUCH WOULD COST ME TO BUILD A SYSTEM TO RUN ON TWO CLOUD VENDORS SIDE-BY-SIDE?

It is a simple question that a business person has. Nevertheless, the response is not simple. Without running a workshop and invest time to understand the business requirements, the current technology stack and expected quality attributes, providing a cost of building the same solution on another CSP (Cloud Service Provider) is hard.

There are tools on the market that provide the mechanism to assess costs on each CSP or what would be the cost of running the same on-premises payloads inside a CSP. Analogies and mapping between different CSP services can be made, making running cost estimation an easy job.

Useful tools:

• Azure Pricing Calculator
• AWS Pricing Calculator
• Total Cost of Ownership (TCO) Calculator
• AWS to Azure service mapping

The challenge is not the running cost! It is THE EFFORT COST to make a system run on another CSP. It is possible? How much ($100k, $1M, 10% of the initial effort)? 

The complexity of building such a system is higher and, in most cases, can be compared with migration from one cloud vendor to another. Trade-offs like using fewer SaaS services and cloud features of a CSP to provide that cloud interoperability level can impact the quality of the final product. Still, they help us to keep costs under control. All the juice that is provided by a specific CSP is diluted when you use less CSP specific features.

When you want to run a solution side-by-side on two CSP, additional effort is required on:

1. Architecture
2. DevOps and Automation
3. Application Development
4. Operational and Management
5. Infrastructure
6. Testing

The main affected layers of a system are:

1. Application
2. Database
3. Infrastructure
4. Communication

The actual cost of a side-by-side solution

I had the opportunity to discuss directly or indirectly with 51 cloud experts with solid experience with the cloud (Azure, AWS and GCP). Each of them provided input related to the additional cost of building a system that side-by-side VS a system that runs only on one CSP.

The additional effort per layer:

• 30% for the Application layer
• 25% for the DB layer 
• 10% for the Communication layer
• 15% for the Infrastructure layer

The additional work per work type:

• 30% for the Architecture
• 20% for the DevOps using Terraform
• 70% for the DevOps with cloud-native solutions (e.g. ARM, CloudFormation)
• 50% for the Operational and Management
• 10% for the Testing
• 30% for the Development
• 25% for the Database

The monthly running cost of a solution side-by-side is not 2 x 100% because tradeoffs need to be done. Taking this into account, the running costs are:

• 215% for the cloud running costs in the best case
• 250% for the cloud running costs on average
• 300% for the cloud running costs when you don't use SaaS and PaaS services anymore, and you need to use only IaaS services

The discussion and estimations were built for the following type of system: A standard business application, built using Kubernetes and Docker, using an RDMS DB, where an ESB and REST API endpoints are used for external communication. In front of the system, an API Gateway and a Load Balancer is required. A VPN connection exists between the cloud and the on-premise office together with basic cloud-native monitoring tools. 

Conclusion

Taking all the previous effort estimations into consideration and the effort required to develop a cloud solution, we expect that to develop a side-by-side system that runs on AWS and Azure would require:

• 35% additional effort to build and design the solution
• 50% additional effort cost for Operational and Management
• 215% additional cost to run the solution on AWS and Azure, side-by-side