If you already hold AWS SAA-C03 or higher, the GCP Associate Cloud Engineer (ACE) is the fastest second-cloud cert you can add. The mental model transfers; the syntax and a few core concepts do not. Below is the AWS-to-GCP translation table you actually need, the genuinely different parts to focus on, and a realistic time estimate.
Exam facts
- Cost: $200 USD via Webassessor (Kryterion proctoring)
- Format: 50 multiple choice and multiple select questions, 120 minutes
- Passing score: not officially published; community estimate around 70%
- Validity: 2 years; renewal requires a full re-sit
- Recommended experience: 6+ months on GCP
Source: cloud.google.com/learn/certification/cloud-engineer, accessed May 2026.
AWS to GCP service translation
| AWS service | GCP equivalent | Conceptual mapping |
|---|---|---|
| EC2 | Compute Engine | Direct equivalent; instance types differ by name |
| S3 | Cloud Storage | Direct equivalent; bucket naming and storage classes differ |
| RDS | Cloud SQL | Direct equivalent; supports MySQL, PostgreSQL, SQL Server |
| DynamoDB | Firestore (NoSQL) or Bigtable (wide-column) | Use case dependent; GCP has separate services |
| Lambda | Cloud Functions or Cloud Run | Cloud Functions for events; Cloud Run for containers |
| ECS / EKS | GKE (Google Kubernetes Engine) | GKE is the Kubernetes path; no direct ECS analogue |
| VPC | VPC Network | Different model: GCP VPCs are global, not regional |
| IAM | Cloud IAM | Very different model; see below |
| CloudWatch | Cloud Monitoring + Cloud Logging | Two separate services in GCP |
| CloudFormation | Deployment Manager or Terraform | GCP shipped Deployment Manager; community standard is Terraform |
| Route 53 | Cloud DNS | Direct equivalent |
| SQS | Pub/Sub | GCP Pub/Sub is a hybrid of SQS plus SNS |
| SNS | Pub/Sub | Same Pub/Sub service for fan-out patterns |
| Kinesis Data Streams | Pub/Sub or Dataflow | Use case dependent |
| Athena | BigQuery | BigQuery is broader; covers Athena and Redshift use cases |
| Redshift | BigQuery | Same; BigQuery is the canonical GCP analytics warehouse |
| Secrets Manager | Secret Manager | Direct equivalent |
| KMS | Cloud KMS | Direct equivalent |
| CloudTrail | Cloud Audit Logs | Built into Cloud Logging |
What is genuinely different
Project hierarchy
AWS organises resources by account; you can have many accounts under an Organization. GCP organises by project, with a separate Organization at the top and Folders in between. Every GCP resource lives inside a project. The project is the unit of billing, IAM scope, and API enablement.
Practical impact: you cannot just "spin up a GCP resource." You first create or select a project, enable the relevant API on the project, configure IAM on the project, and only then create the resource. This is friction that AWS engineers underestimate at the start.
IAM model
AWS IAM grants permissions through policies attached to users, groups, or roles. GCP IAM grants roles to identities (users, groups, service accounts) at a specific scope (project, folder, or organization). The same role bound at different scopes produces very different effective permissions.
Key concepts that the ACE exam tests:
- Predefined roles vs custom roles: GCP ships hundreds of predefined roles. The exam expects you to know roles/owner, roles/editor, roles/viewer, and the principle of least privilege.
- Service accounts: in GCP, service accounts are first-class identities used by applications running on Compute Engine or GKE to authenticate to other GCP services. AWS has IAM roles for similar purposes, but the GCP service account model is more explicit.
- Workload Identity: the modern way GKE pods authenticate to GCP APIs. Comparable to IRSA in AWS EKS but with different setup.
Networking
The biggest mental model shift: GCP VPCs are global by default, with subnets per region. You can have a single VPC that spans all GCP regions. AWS VPCs are regional; cross-region traffic requires VPC peering or Transit Gateway.
Practical impact: networking design in GCP is simpler for global apps and harder to mentally separate for engineers used to AWS regional thinking. The exam tests Shared VPC, VPC peering, Cloud NAT, Cloud Router, and Cloud Interconnect.
Billing and quotas
GCP has per-project quotas that you frequently bump into during lab work. AWS has account-level service quotas. The exam tests how to view and request quota increases through the GCP console and gcloud CLI.
Time to prep
From AWS Pro level (SAP-C02 or DOP-C02): 4-6 weeks of evening study. The mental model transfers; you spend most prep time learning gcloud syntax and the GCP-specific concepts above.
From AWS Associate (SAA-C03 or equivalent): 6-9 weeks. Add time for deeper hands-on with GCP-specific services like BigQuery, Pub/Sub, and GKE.
From AWS Cloud Practitioner only: 10-14 weeks. The cert is achievable but you are essentially learning cloud concepts and GCP simultaneously.
Recommended study resources
- Coursera Google Cloud Associate Cloud Engineer Professional Certificate (free to audit). Six-course series covering exam topics. Source: coursera.org/professional-certificates/cloud-engineering-gcp.
- Google Cloud Skills Boost (formerly Qwiklabs). Hands-on labs in real GCP environments. Around $29/month subscription. The fastest way to get gcloud muscle memory.
- Dan Sullivan's GCP Associate Cloud Engineer Study Guide (Sybex / Wiley). Covers exam objectives systematically; useful as a structured reference.
- Practice exams: Whizlabs and ExamTopics for volume; Tutorials Dojo for quality (smaller GCP catalogue than AWS but the questions that exist are well-written).
- Free Google Cloud documentation: the cloud.google.com docs are unusually well-written. Read the IAM, VPC, and Compute Engine sections directly.
The honest career framing
Adding GCP ACE to an AWS-heavy resume is most useful in three scenarios: you are pivoting to a multi-cloud or platform engineering role, your employer is starting GCP adoption, or you are targeting data-engineering roles where BigQuery is the standard analytics warehouse. For pure AWS-only career trajectories, GCP ACE adds limited value compared to deepening AWS expertise.
Salary impact: holding both AWS SAA-C03 and GCP ACE correlates with roughly a 5-8% additional salary uplift versus AWS-only at the same experience level, per Dice Tech Salary Report 2025-2026 data. The signal is strongest at companies running multi-cloud or evaluating cloud migration.
Quick gcloud commands AWS engineers should learn first
# Authenticate and set the active project
gcloud auth login
gcloud config set project <PROJECT_ID>
gcloud config list
# List Compute Engine VMs (the EC2 equivalent)
gcloud compute instances list
# List Cloud Storage buckets (the S3 equivalent)
gcloud storage buckets list
# Check IAM bindings on the current project
gcloud projects get-iam-policy <PROJECT_ID>
# Enable an API on the project (no AWS analogue, this is GCP-specific)
gcloud services enable compute.googleapis.com
The "enable API" step is the one that catches AWS engineers most often. In GCP, services are off by default per project, and you cannot use them until you enable the corresponding API. There is no equivalent step in AWS.