Senior Cloud Engineer GCP Interview Questions: Complete Guide

# Create managed instance group with autoscaling
gcloud compute instance-groups managed create my-mig \
  --base-instance-name=my-app \
  --template=my-template \
  --size=3 \
  --zone=us-central1-a

# Configure autoscaling
gcloud compute instance-groups managed set-autoscaling my-mig \
  --max-num-replicas=10 \
  --min-num-replicas=3 \
  --target-cpu-utilization=0.7 \
  --cool-down-period=90

# Create load balancer
gcloud compute backend-services create my-backend \
  --protocol=HTTP \
  --health-checks=my-health-check \
  --global

# Add instance group to backend
gcloud compute backend-services add-backend my-backend \
  --instance-group=my-mig \
  --instance-group-zone=us-central1-a \
  --global

# Create GKE cluster
gcloud container clusters create my-cluster \
  --num-nodes=3 \
  --machine-type=e2-medium \
  --zone=us-central1-a \
  --enable-autoscaling \
  --min-nodes=3 \
  --max-nodes=10

# Get credentials
gcloud container clusters get-credentials my-cluster \
  --zone=us-central1-a

# Deploy application
kubectl apply -f - <<EOF
apiVersion: apps/v1
kind: Deployment
metadata:
  name: myapp
spec:
  replicas: 3
  selector:
    matchLabels:
      app: myapp
  template:
    metadata:
      labels:
        app: myapp
    spec:
      containers:
      - name: myapp
        image: gcr.io/my-project/myapp:v1
        ports:
        - containerPort: 8080
        resources:
          requests:
            cpu: 100m
            memory: 128Mi
          limits:
            cpu: 500m
            memory: 512Mi
---
apiVersion: v1
kind: Service
metadata:
  name: myapp-service
spec:
  type: LoadBalancer
  selector:
    app: myapp
  ports:
  - port: 80
    targetPort: 8080
EOF

# Cloud Function example
def hello_pubsub(event, context):
    """Triggered by Pub/Sub message"""
    import base64
    
    if 'data' in event:
        message = base64.b64decode(event['data']).decode('utf-8')
        print(f'Received message: {message}')
        
        # Process message
        process_data(message)

# Deploy Cloud Function
gcloud functions deploy hello_pubsub \
  --runtime=python39 \
  --trigger-topic=my-topic \
  --entry-point=hello_pubsub

# Deploy Cloud Run
gcloud run deploy myapp \
  --image=gcr.io/my-project/myapp:v1 \
  --platform=managed \
  --region=us-central1 \
  --allow-unauthenticated

# Enable Shared VPC in host project
gcloud compute shared-vpc enable my-host-project

# Attach service project
gcloud compute shared-vpc associated-projects add my-service-project \
  --host-project=my-host-project

# Grant permissions
gcloud projects add-iam-policy-binding my-host-project \
  --member=serviceAccount:[email protected] \
  --role=roles/compute.networkUser

# Use Recommender API
gcloud recommender recommendations list \
  --project=my-project \
  --location=us-central1 \
  --recommender=google.compute.instance.MachineTypeRecommender

# Create preemptible instance
gcloud compute instances create my-preemptible \
  --preemptible \
  --machine-type=e2-medium

# Set lifecycle policy
gsutil lifecycle set lifecycle.json gs://my-bucket

# lifecycle.json
{
  "lifecycle": {
    "rule": [
      {
        "action": {"type": "SetStorageClass", "storageClass": "NEARLINE"},
        "condition": {"age": 30}
      },
      {
        "action": {"type": "Delete"},
        "condition": {"age": 365}
      }
    ]
  }
}

# Use service accounts with minimal permissions
gcloud iam service-accounts create my-app-sa \
  --display-name="My App Service Account"

# Grant specific role
gcloud projects add-iam-policy-binding my-project \
  --member=serviceAccount:[email protected] \
  --role=roles/storage.objectViewer \
  --condition='expression=resource.name.startsWith("projects/_/buckets/my-bucket"),title=bucket-access'

# Customer-managed encryption keys
gcloud kms keyrings create my-keyring \
  --location=global

gcloud kms keys create my-key \
  --location=global \
  --keyring=my-keyring \
  --purpose=encryption

# Use with Cloud Storage
gsutil -o 'GSUtil:encryption_key=...' cp file.txt gs://my-bucket/

-- Create partitioned table
CREATE TABLE mydataset.events
(
  event_id STRING,
  user_id STRING,
  event_type STRING,
  event_data JSON,
  event_timestamp TIMESTAMP
)
PARTITION BY DATE(event_timestamp)
CLUSTER BY user_id, event_type
OPTIONS(
  partition_expiration_days=90,
  require_partition_filter=true
);

-- Create materialized view
CREATE MATERIALIZED VIEW mydataset.daily_summary
AS
SELECT
  DATE(event_timestamp) as event_date,
  event_type,
  COUNT(*) as event_count,
  COUNT(DISTINCT user_id) as unique_users
FROM mydataset.events
GROUP BY event_date, event_type;

-- Time-based partitioning
CREATE TABLE mydataset.sales
PARTITION BY DATE(sale_date)
AS SELECT * FROM source_table;

-- Integer range partitioning
CREATE TABLE mydataset.user_data
PARTITION BY RANGE_BUCKET(user_id, GENERATE_ARRAY(0, 1000000, 10000))
AS SELECT * FROM source_table;

-- Query with partition filter (cost-effective)
SELECT *
FROM mydataset.events
WHERE DATE(event_timestamp) BETWEEN '2024-01-01' AND '2024-01-31'
  AND event_type = 'purchase';

-- Cluster by frequently filtered columns
CREATE TABLE mydataset.logs
PARTITION BY DATE(log_timestamp)
CLUSTER BY user_id, region, status
AS SELECT * FROM source_logs;

-- Queries benefit from clustering
SELECT *
FROM mydataset.logs
WHERE DATE(log_timestamp) = '2024-11-26'
  AND user_id = '12345'
  AND region = 'us-east1';

-- Bad: SELECT * (scans all columns)
SELECT * FROM mydataset.large_table;

-- Good: SELECT specific columns
SELECT user_id, event_type, event_timestamp
FROM mydataset.large_table;

-- Use approximate aggregation for large datasets
SELECT APPROX_COUNT_DISTINCT(user_id) as unique_users
FROM mydataset.events;

-- Avoid self-joins, use window functions
SELECT
  user_id,
  event_timestamp,
  LAG(event_timestamp) OVER (PARTITION BY user_id ORDER BY event_timestamp) as prev_event
FROM mydataset.events;

# Set maximum bytes billed
bq query \
  --maximum_bytes_billed=1000000000 \
  --use_legacy_sql=false \
  'SELECT COUNT(*) FROM mydataset.large_table'

# Dry run to estimate costs
bq query \
  --dry_run \
  --use_legacy_sql=false \
  'SELECT * FROM mydataset.large_table'

# Load from Cloud Storage
bq load \
  --source_format=NEWLINE_DELIMITED_JSON \
  --autodetect \
  mydataset.mytable \
  gs://mybucket/data/*.json

# Load with schema
bq load \
  --source_format=CSV \
  --skip_leading_rows=1 \
  mydataset.mytable \
  gs://mybucket/data.csv \
  schema.json

# Streaming inserts (real-time)
from google.cloud import bigquery

client = bigquery.Client()
table_id = "my-project.mydataset.mytable"

rows_to_insert = [
    {"user_id": "123", "event_type": "click", "timestamp": "2024-11-26T10:00:00"},
    {"user_id": "456", "event_type": "purchase", "timestamp": "2024-11-26T10:05:00"},
]

errors = client.insert_rows_json(table_id, rows_to_insert)
if errors:
    print(f"Errors: {errors}")

-- Create Spanner instance
gcloud spanner instances create my-instance \
  --config=regional-us-central1 \
  --nodes=3 \
  --description="Production instance"

-- Create database
gcloud spanner databases create my-database \
  --instance=my-instance \
  --ddl='CREATE TABLE Users (
    UserId INT64 NOT NULL,
    Username STRING(100),
    Email STRING(255),
    CreatedAt TIMESTAMP
  ) PRIMARY KEY (UserId)'

-- Insert data
gcloud spanner databases execute-sql my-database \
  --instance=my-instance \
  --sql="INSERT INTO Users (UserId, Username, Email, CreatedAt)
        VALUES (1, 'alice', '[email protected]', CURRENT_TIMESTAMP())"

-- Query with strong consistency
gcloud spanner databases execute-sql my-database \
  --instance=my-instance \
  --sql="SELECT * FROM Users WHERE UserId = 1"

from google.cloud import spanner

# Create client
spanner_client = spanner.Client()
instance = spanner_client.instance('my-instance')
database = instance.database('my-database')

# Read with strong consistency
def read_user(user_id):
    with database.snapshot() as snapshot:
        results = snapshot.execute_sql(
            "SELECT UserId, Username, Email FROM Users WHERE UserId = @user_id",
            params={"user_id": user_id},
            param_types={"user_id": spanner.param_types.INT64}
        )
        for row in results:
            print(f"User: {row[0]}, {row[1]}, {row[2]}")

# Write with transaction
def create_user(user_id, username, email):
    def insert_user(transaction):
        transaction.execute_update(
            "INSERT INTO Users (UserId, Username, Email, CreatedAt) "
            "VALUES (@user_id, @username, @email, CURRENT_TIMESTAMP())",
            params={
                "user_id": user_id,
                "username": username,
                "email": email
            },
            param_types={
                "user_id": spanner.param_types.INT64,
                "username": spanner.param_types.STRING,
                "email": spanner.param_types.STRING
            }
        )
    
    database.run_in_transaction(insert_user)

# Create Cloud SQL instance
gcloud sql instances create my-instance \
  --database-version=POSTGRES_14 \
  --tier=db-n1-standard-2 \
  --region=us-central1 \
  --root-password=mypassword

# Create database
gcloud sql databases create mydatabase \
  --instance=my-instance

# Connect
gcloud sql connect my-instance --user=postgres

# Create read replica
gcloud sql instances create my-replica \
  --master-instance-name=my-instance \
  --tier=db-n1-standard-1 \
  --region=us-east1

# Create access policy
gcloud access-context-manager policies create \
  --organization=123456789 \
  --title="Production Policy"

# Create access level
gcloud access-context-manager levels create CorpNetwork \
  --policy=accessPolicies/123456789 \
  --title="Corporate Network" \
  --basic-level-spec=access_level.yaml

# access_level.yaml
conditions:
  - ipSubnetworks:
    - 203.0.113.0/24  # Corporate IP range
  - members:
    - user:[email protected]

# Create perimeter
gcloud access-context-manager perimeters create production_perimeter \
  --policy=accessPolicies/123456789 \
  --title="Production Perimeter" \
  --resources=projects/123456789012 \
  --restricted-services=storage.googleapis.com,bigquery.googleapis.com \
  --access-levels=accessPolicies/123456789/accessLevels/CorpNetwork

# Add project to perimeter
gcloud access-context-manager perimeters update production_perimeter \
  --policy=accessPolicies/123456789 \
  --add-resources=projects/987654321098

# ingress_rule.yaml
ingressPolicies:
  - ingressFrom:
      sources:
        - accessLevel: accessPolicies/123456789/accessLevels/CorpNetwork
      identities:
        - serviceAccount:[email protected]
    ingressTo:
      resources:
        - '*'
      operations:
        - serviceName: storage.googleapis.com
          methodSelectors:
            - method: '*'

# Apply ingress rule
gcloud access-context-manager perimeters update production_perimeter \
  --policy=accessPolicies/123456789 \
  --set-ingress-policies=ingress_rule.yaml

# egress_rule.yaml
egressPolicies:
  - egressFrom:
      identities:
        - serviceAccount:[email protected]
    egressTo:
      resources:
        - projects/external-project-id
      operations:
        - serviceName: storage.googleapis.com
          methodSelectors:
            - method: 'google.storage.objects.create'

# Apply egress rule
gcloud access-context-manager perimeters update production_perimeter \
  --policy=accessPolicies/123456789 \
  --set-egress-policies=egress_rule.yaml

# Test access from within perimeter
from google.cloud import storage

def test_access():
    try:
        client = storage.Client()
        bucket = client.bucket('my-protected-bucket')
        blobs = list(bucket.list_blobs())
        print(f"Access granted: {len(blobs)} objects")
    except Exception as e:
        print(f"Access denied: {e}")

# This will succeed from authorized network
# This will fail from unauthorized network
test_access()

# View VPC SC logs
gcloud logging read \
  'protoPayload.metadata.@type="type.googleapis.com/google.cloud.audit.VpcServiceControlAuditMetadata"' \
  --limit=50 \
  --format=json