Case Study: GCP Platform Uplift, Automation and Gitlab Runner Migration

By intelia | @intelia | Dec 12, 2024

intelia was engaged by a leading Retail and Logistics organisation to enhance and manage their Google Cloud infrastructure, which supports hundreds of workloads for daily data analytics and computational needs. Our engagement focused on ensuring optimal performance, scalability, and security while streamlining cloud operations. Additionally, we migrated their Gitlab deployment process from AWS across to Google Cloud.

The challenge

The client faced challenges with their existing project creation process, where developers were managing it without adhering to best practices. This lack of governance led to security risks, resource mismanagement, and inconsistencies across their cloud environment. Recognising these risks, the client emphasised the need for a well-defined cloud governance framework that aligns with security standards, IAM policy best practices, and Google Cloud’s recommended guidelines.

The solution

intelia addressed the customer’s challenge by redesigning the CI/CD pipeline using Google’s Cloud Foundation Toolkit while enforcing the principle of least privilege for enhanced security. Additionally, all workloads were transitioned to an Infrastructure-as-Code (IaC) managed environment using Terraform, ensuring consistency, scalability, and governance. Throughout the migration, intelia provided technical support and operational oversight to minimise disruptions and ensure seamless business continuity.

The results

intelia’s solution solution streamlined the GCP project provisioning process, enhancing governance, auditability, and security. By implementing policy constraints, disabling unused services, and enforcing best practices aligned with Google Cloud’s foundational framework, we significantly improved the security posture of the customer’s platform. As a result, the customer now has a compliant, well-governed GCP infrastructure with the flexibility to seamlessly provision and manage additional workloads.