Cloud cost-optimization simulator
Cloud cost-optimization simulator

Cloud computing offers the greatest benefits when strong FinOps capabilities are in place to continuously manage and optimize costs.1

This cloud cost-optimization simulator details the range of levers that can be used to substantially reduce costs for one illustrative scenario of an application on the cloud.2

In this scenario, a number of modules of a retail bank’s credit-scoring application have been migrated to cloud, primarily:

  • web UI for back-office personnel
  • workflow and business-logic applications
  • advanced analytics scoring models

Some of the core components and a substantial amount of customer data are still being hosted on-premises. For the purposes of this simulation, total annual cloud spend for the application is assumed to be roughly $1 million.

(A note on cloud prices shown: Cloud pricing is dynamic and changes frequently; the prices in this interactive are for illustrative purposes only, to show the relative impact of different optimization options.)

Compute
Total quantity: 150 instances
Average instance price: ~ $0.45/hour
Total annual cost: $600,000
Storage
Total quantity: 125 TB
Unit price: ~ $0.10/GB/month
Total annual cost: $150,000
Network
Total quantity: 1.5 TB/day
Unit price: ~ $0.10/GB transferred
Total annual cost: $50,000
Managed database
Total quantity: 14 instances
Unit price: ~ $0.80/hour
Total annual cost: $100,000
Other services
Total annual cost: $100,000

1Activating these levers can typically be implemented as configuration changes without substantial revision to the application architecture. Deriving full value from these levers also requires companies to have the right enablers in place.

2This is a generalized illustrative example. The specific cloud services and pricing information are based on offerings from major cloud service providers and do not refer to any particular cloud service provider. The list of optimization levers is based on our experience serving companies across industries and geographies on the topic of cloud cost optimization and does not represent all possible optimization options. The impact for each of the levers would heavily depend on specifics of application architecture and environment configuration; ranges provided represent the most common scenarios.