Cloud Economics for Migrations and Modernizations

4 min readAug 27, 2022

Introduction

In many cloud transformation business cases the overall cost reduction will be the primary business benefit that most companies will expect as a major rational for the investment in it. Whether it is Cloud Migration or Cloud Modernization, the calculation and estimation of Total Cost of Ownership (TCO) will be utmost important and significant to analyze and produce when it comes to Cloud Transformation. When there is estimated TCO for the Cloud Transformation then it becomes easy to visualize the target cost of IT and hence compare that against the cost models or TCO of on-premise estate if they are available for the Enterprises.

Based on experience of working on different cloud transformation business case proposals, the various cost aspects that need to be considered to model the TCO comparison between onpremise and cloud have been covered in this article as a blueprint for any enterprise to consider.

Generic TCO Framework

At high level there are different cost types involved when it comes to Cloud Transformation such as below

Hosting Cost

In general Cloud CoE team and respective App Team from the given (Line of Business) Organization will own the Cloud Hosting Cost which will involve various cost elements such as below

Software & ADMO Cost

In most of the cases, the Software and ADMO cost is owned by the respective App team & Platform Team of the given Line of Business Organization which includes the following cost elements

Enterprise Services Cost

Typically, Enterprise services are owned and managed by Cloud CoE which are generic, standard and common services as required for all the Line Of Business organizations across the enterprise. This involves following cost elements.

Uncategorized Cost

There will be some other costs which are incurred as a pre-requisite preparation, post migration aspects which will be owned by the given Line of Business Organization team or App team. Some of those costs will be owned by the enterprise but will not be managed by the Enterprise themselves directly such as below cost elements.

Cloud Cost Management KPIs

In IT operations, metrics and KPIs help the enterprises to measure, monitor and manage the performance and align appropriately and typically it will be a journey to achieve the benchmarks in Metrics and KPIs. Organizations typically start this journey with unit metrics focusing on cloud costs and eventually progress toward a set of clearly defined business value metrics.

Traditionally, enterprises resorted to using traditional P&L allocation models, such as using business revenue or headcount of the associated business units while allocating the CapEx & OpEx budgets for the onpremise data centers cost management.

When it comes to Cloud Cost management, in order to establish a full chargeback of typical cloud services, enterprises will need to attribute costs associated with 3 types of cloud resources.

Taggable cloud resources that are aligned to a specific P&L of a given line of business.
Taggable cloud resources that are shared across multiple lines of business
Untaggable cloud resources (eg. networking/connectivity cost, support cost, common security solutions, guardrails, auditing, hybrid cloud data integrations, marketplace costs, etc)

For all the above types, it is absolute necessary for enterprises to use unit cost models (eg. API calls).

Best Practices for Cloud Cost Control

As a general practice, budget limits in Cloud are set per account or organization but not specific to a given App or Team. As long as the cost is in budget limits for the given account then governance will assume all is good but that poses a key threat of missing the granular level aspects at App level.

Code Quality

There are number of instances where the cost overshooting happens due to unknown quality issue or defect in the code of the application which will be unnoticed during test and deployment but will be realized when it is operated in the production.

Hence teams have to cultivate new practices and methods to discover the issues in the code that will affect the cost and budget limits. When the application is onpremise, then the cost specific test and assurance is not significant and necessary. But when the application is moved to Cloud, then cost control will be a new dimension to be considered in quality assurance.

Following practices are required for cost control and assurance:

Specific Cost related KPIs per Application to test, measure, monitor and control as per the given budget limits allocated for the LoB or App Team
Explicit Test Cases to test cost aspects of Cloud

Unit Cost Modelling

Cost Mapping to different units
- business value
- product feature/functionality/capability
- number of users
- number of API calls/transactions/queries/page hits (throughput)
- end customer volume
- volume of data objects
- data size