Understanding AWS Infrastructure Costs: A FinOps Case Study

INDUSTRY

Real Estate

CUSTOMER OVERVIEW

Our customer is a global online real estate advertising company, and operates one of Australia’s leading residential, commercial and share property websites, as well as a number of leading property portals in Asia.

Case Study

BUSINESS CHALLENGE

Within this large and very mature agile agency, multiple business units use and rely heavily on Amazon Web Services (AWS) to support their development. However the finance team had no breakdown of AWS infrastructure costs by department.

Case Study

SOLUTION

At that time there were no ‘off the shelf’ applications available manage organisation spend on AWS infrastructure. Cevo consultants were engaged to help identify and build a bespoke solution which would allow the data to be pulled through to an easy-to-read dashboard giving a full break down of the AWS costs by department.

AWS allows each resource to contain an individual tag. Setting this up correctly, by department, became a key requirement to ensure the dashboard displayed accurate figures. A consistent naming convention was agreed upon, and the business units were approached to ensure their resources were appropriately tagged (existing and new).

While managing the tagging process Cevo also assisted in developing a tool which reviewed all services in the AWS account, and helped the client teams decommission any that did not respect the tagging requirements. This created a standard procedure moving forward, ensuring all services were measurable and leaving the numbers as accurate as possible.

A management decision was made to forego the initial dashboard so long as the information could be viewed in a meaningful way. The last phase of the project involved creating a spreadsheet which fed in all financial information broken down by department and AWS resource.

Case Study_1

BENEFITS

The solution provided met all the key requirements of the finance team, allowing for the first time a proper allocation of current costs as well as giving visibility over expected future spend.

Teams also have a better understanding of their infrastructure costs which allows for better decisions to be made around required servers and the most appropriate ways to configure code and environments.