← All case studiesSoftware Security · Cloud modernization

Case study

A Software Security Giant Migrates SaaS Platform to Google Cloud With Zero Application Changes

SquareShift converted the infrastructure code and billing pipeline, then validated both against the client's production data.

Book a session →
4Infrastructure layers ported to Google Cloud
55AWS Terraform modules mapped to Google Cloud
To the ByteBilling matched AWS in parity test

Client

A multi-tenant SaaS platform that ran on AWS and needed the same product on Google Cloud.

Industry

Software Security

Services

CMCloud Modernization

Stack

AWS · HashiCorp Terraform · Terraform Cloud · Google Cloud · Cloud Run · Cloud SQL · Cloud Storage · BigQuery · Cloud Armor · Pub/Sub

How SquareShift delivered it.

01

Project context

The client's roadmap needed the same product on Google Cloud, with unchanged application code. SquareShift converted the infrastructure code and billing pipeline, then validated both against the client's production data.

02

Project objectives

Move the platform to Google Cloud.

Keep billing, customer logs and provisioning identical, without changing the application.

03

Challenges

Usage billing had to match AWS exactly, with no drift.

AWS-specific billing and log services had no direct Google Cloud equivalents.

04

Solution delivery

SquareShift rebuilt the four Terraform layers natively on Google Cloud, matching the client's existing workspace naming and structure.

Replaced a five-service AWS egress meter with one built-in metric.

Rebuilt load-balancer billing on BigQuery with identical hourly, per-tenant billing files.

Proved billing matched AWS to the byte on production data. The Google Cloud meter reported 28,845,402 bytes, the identical figure, with the identical hourly breakdown.

Ported log delivery, firewall rules and disaster recovery design.

05

The outcome

The same four infrastructure layers now run native on Google Cloud. They applied cleanly in the client's environment, with zero drift on re-plan.

Load-balancer meter matched AWS logic to the byte (28,845,402) with identical hourly breakdown.

Customer log archives are identical in folder structure, file naming and record format, and are delivered event-driven through Pub/Sub.

“
Nothing is done until it is demonstrated end to end.

Senior Consultant, SquareShift

Related work