GCP Core Services

Compute Services

Compute Engine

Compute Engine provides IaaS virtual machines — equivalent to AWS EC2. You select a machine family (general-purpose, compute-optimized, memory-optimized), an image (Ubuntu, Debian, CentOS, Windows Server), and a zone.

gcloud compute instances create web-server \
  --zone us-central1-a \
  --machine-type e2-micro \
  --image-family ubuntu-2204-lts \
  --image-project ubuntu-os-cloud \
  --tags http-server

Key differentiators:

• Sole-tenant nodes — Dedicated physical servers for licensing or compliance
• Preemptible / Spot VMs — Up to 91% discount for interruptible workloads
• Live migration — GCP migrates VMs between hosts during maintenance without reboot (most machine types)

Google Kubernetes Engine (GKE)

GKE is the managed Kubernetes service that originated from Google’s internal Borg system — the same technology that spawned Kubernetes itself. GKE is widely considered the most mature managed Kubernetes offering.

gcloud container clusters create my-cluster \
  --zone us-central1-a \
  --num-nodes 3 \
  --enable-autopilot

Autopilot mode offloads node management entirely — GKE manages the underlying compute, scaling, and security. Standard mode gives you control over node pools and configuration.

Cloud Run

Cloud Run is a serverless container platform that executes stateless HTTP containers on a fully managed infrastructure. You package your app as a container image, push it to Artifact Registry, and Cloud Run scales from zero to thousands of requests.

gcloud run deploy my-service \
  --image gcr.io/my-project/my-app:latest \
  --region us-central1 \
  --allow-unauthenticated

Cloud Run charges only for the time your container handles requests (down to 100ms increments) and scales to zero when idle — the most cost-effective option for intermittent workloads.

Storage Services

Cloud Storage

Cloud Storage is GCP’s object storage service, equivalent to AWS S3. It offers four storage classes:

# Create a bucket and upload a file
gcloud storage buckets create gs://my-demo-bucket --location us-central1
gcloud storage cp index.html gs://my-demo-bucket

Cloud Storage supports object versioning, lifecycle policies, and uniform bucket-level access control (IAM) or fine-grained ACLs.

Networking Services

VPC

GCP’s VPC is a global networking resource — unlike AWS VPCs which are regional. A single VPC can span multiple continents without peering. Subnets are regional, and resources in any zone within that region can use them.

gcloud compute networks create my-vpc --subnet-mode custom
gcloud compute networks subnets create web-subnet \
  --network my-vpc \
  --region us-central1 \
  --range 10.0.1.0/24

GCP also supports Shared VPC (manage networking from a central project while allowing other projects to use the subnets) and VPC Network Peering for connecting across projects or organizations.

Cloud Load Balancing

Cloud Load Balancing is a single, global, anycast-based load balancing service. Unlike AWS (which requires separate ALB/NLB per region), a single Cloud Load Balancer can route traffic globally to backends in multiple regions.

gcloud compute forwarding-rules create http-rule \
  --region us-central1 \
  --ports 80 \
  --target-http-proxy my-proxy

It supports HTTP(S), TCP/SSL, and UDP protocols, with built-in CDN integration (Cloud CDN) and DDoS protection (Cloud Armor).

Management Tools

# List all compute instances across projects
gcloud compute instances list

# View resource hierarchy
gcloud resource-manager folders list
gcloud projects list

GCP’s core services follow a philosophy of global resources, managed services, and API-first design. This foundation enables the data and AI capabilities covered in the next section.