Google Cloud Platform
Google Cloud Platform (GCP) is Google’s hyperscale Cloud offering. We adopt it as the multi-cloud spearhead for container workloads, data, AI, and DevSecOps-aligned platforms. It is not a wholesale replacement for AWS or Azure where those estates already exist. GCP’s service catalog reflects Google’s internal scale problems more than a feature checklist against other hyperscalers.
Blurb
The new way to cloud.
Summary
Garden stance: Use GCP as the preferred greenfield cloud for Kubernetes (Google GKE), big data, ML/AI services, and security-first platform engineering. Keep Hybrid Cloud in play: run each workload where the fit is strongest rather than mirroring every service across vendors.
Strengths: Independent CPU, memory, and GPU billing for right-sized compute. Global networking and storage foundations exposed as managed services. First-class observability through Cloud Operations (formerly Stackdriver). Strong IaC and policy guardrail story across the control plane.
Caveats: Smaller third-party marketplace than AWS. Not every AWS or Azure service has a direct peer. Teams need explicit multi-cloud boundaries and Terraform discipline so GCP stays the spearhead, not another silo.
Details
| Topic | Notes |
|---|---|
| Position | Third-largest hyperscaler by market share; strongest fit for K8s-native and data/AI greenfield |
| Compute | Decoupled vCPU, RAM, and GPU pricing vs fixed instance families on AWS |
| K8s | Google GKE is the default managed cluster choice; Anthos for multi-cloud fleet ops |
| Observability | Cloud Operations integrates metrics, logging, and tracing; GCP metrics included |
| Security | BeyondCorp lineage; org/folder/project IAM; policy and IaC-first setup patterns |
| Data | BigQuery, Spanner, and shared storage layer for analytics and global consistency |
| Multi-cloud | Hybrid Cloud pattern: GCP spearhead, not full migration from AWS/Azure |
| Tooling | Terraform and Google-native deployment tools; GoLang ecosystem alignment |