Azure interviews test your understanding of cloud fundamentals, Azure-specific services, networking, security, and architecture patterns. This guide covers the 50 most common questions — with concise answers, comparisons, and real examples.
Quick reference
| Topic | Most asked questions |
|---|---|
| Core concepts | Regions, AZs, resource groups, subscriptions |
| IAM | Azure AD, RBAC, managed identities, service principals |
| Compute | VMs, App Service, AKS, Azure Functions, Container Apps |
| Storage | Blob, File, Queue, Table, Azure Data Lake |
| Networking | VNet, NSG, Azure Firewall, Load Balancer, Front Door |
| Databases | Azure SQL, Cosmos DB, Synapse, Cache for Redis |
| Security | Key Vault, Defender, Sentinel, Policy |
| DevOps | Azure DevOps, GitHub Actions, pipelines |
| AI/ML | Azure OpenAI, Cognitive Services, ML Studio |
| Architecture | Well-Architected Framework, landing zones |
Core Concepts
1. What is the Azure global infrastructure?
| Component | Description | Example |
|---|---|---|
| Geography | Area containing ≥2 Regions with data-residency boundaries | Europe, US, Asia Pacific |
| Region | Set of data centres in a latency-defined perimeter | East US, West Europe |
| Availability Zone (AZ) | Physically separate data centres within a Region | Zone 1, Zone 2, Zone 3 |
| Region Pair | Two Regions ≥300 miles apart for DR replication | East US ↔ West US |
| Edge Zone | Micro data centre at telecom operators for low-latency | AT&T, Vodafone |
| Sovereign Cloud | Isolated environments for government/compliance | Azure Government, China |
Azure currently has 60+ regions worldwide — more than any other cloud provider.
2. What is an Azure Resource Group?
A Resource Group is a logical container for Azure resources that share the same lifecycle, permissions, and billing context.
Key properties:
- All resources must belong to exactly one Resource Group
- Resources in different regions can share one Resource Group
- Deleting a Resource Group deletes all contained resources
- RBAC and tags applied to a Resource Group cascade to child resources
- Resources can be moved between Resource Groups (
Move-AzResource)
Subscription
└── Resource Group (rg-prod-eastus)
├── VM (vm-api-01)
├── Storage Account (stproddata)
├── Virtual Network (vnet-prod)
└── Key Vault (kv-prod-secrets)
3. What is the Azure hierarchy?
Management Group (org-wide governance)
└── Subscription (billing unit)
└── Resource Group (lifecycle container)
└── Resource (VM, Storage, etc.)
| Level | Purpose | Scope for policy/RBAC |
|---|---|---|
| Management Group | Group subscriptions for governance | Inherited by all below |
| Subscription | Billing boundary, service limits | Inherited by Resource Groups |
| Resource Group | Deployment and lifecycle unit | Inherited by Resources |
| Resource | Individual service instance | Resource only |
4. What is the difference between Azure Regions and Availability Zones?
| Feature | Region | Availability Zone |
|---|---|---|
| Definition | Geographic area | Physical data centre in a Region |
| Distance | 1,000s of km apart | ≤10 ms latency between zones |
| Failure isolation | Cross-region DR | Zone failure (power, cooling, network) |
| Typical use | Data residency, latency | High-availability apps |
| Replication | Manual/async | ZRS, zone-redundant services |
| Not all regions have | — | Only 30+ regions have AZs |
5. What is an Azure Subscription?
A subscription is a billing and access boundary in Azure:
- Contains resources and is billed separately
- Has limits (quotas) per service (e.g., 20 vCPUs per region for free tier)
- Linked to an Azure AD tenant for identity
- Common patterns: dev/staging/prod subscriptions, per-team subscriptions
Subscription types: Free, Pay-As-You-Go, Enterprise Agreement (EA), CSP, Visual Studio.
IAM & Security
6. What is Azure Active Directory (Azure AD / Microsoft Entra ID)?
Azure AD (now rebranded as Microsoft Entra ID) is Azure's cloud-native identity platform:
| Feature | Description |
|---|---|
| Users & Groups | Identity management for humans |
| Service Principals | Application identities |
| Managed Identities | Auto-managed credentials for Azure resources |
| Conditional Access | Policy-based access control (MFA, device compliance) |
| B2B | Collaborate with external users |
| B2C | Customer identity for apps |
| SSO | Single sign-on for 3,000+ SaaS apps |
| PIM | Privileged Identity Management — just-in-time access |
Azure AD ≠ on-premises Windows Active Directory. Azure AD is HTTP/S based (OAuth 2.0, OIDC, SAML), not LDAP/Kerberos.
7. What is Azure RBAC?
Role-Based Access Control manages who can do what on Azure resources:
| Component | Description | Example |
|---|---|---|
| Security Principal | Who gets access | User, Group, Service Principal, Managed Identity |
| Role Definition | Set of allowed actions | Owner, Contributor, Reader, Storage Blob Data Contributor |
| Scope | Where role applies | Management Group / Subscription / Resource Group / Resource |
| Role Assignment | Binding principal + role + scope | Assign Contributor to dev-team on rg-dev |
Built-in roles:
- Owner — full access + can manage access
- Contributor — full access but cannot manage access
- Reader — view only
- User Access Administrator — manage access only
8. What is a Managed Identity?
A Managed Identity is an automatically managed service principal in Azure AD for an Azure resource — no credentials to manage.
| Type | Description | Use case |
|---|---|---|
| System-assigned | Tied to one resource; deleted with resource | VM needs access to Key Vault |
| User-assigned | Standalone resource; shared across multiple resources | Multiple VMs sharing same Key Vault access |
# Assign system-assigned managed identity to VM
az vm identity assign --resource-group rg-prod --name vm-api
# Grant Key Vault access to the managed identity
az keyvault set-policy --name kv-prod \
--object-id <identity-principal-id> \
--secret-permissions get list
9. What is Azure Key Vault?
Azure Key Vault is a secrets management service for storing keys, secrets, and certificates:
| Feature | Description |
|---|---|
| Secrets | Connection strings, passwords, API keys |
| Keys | Cryptographic keys (RSA, EC) for encryption/signing |
| Certificates | TLS/SSL certificates with auto-renewal |
| HSM | Hardware Security Module backing (Premium tier) |
| Soft-delete | 7–90 day recovery window |
| Purge protection | Prevent permanent deletion |
| Access | RBAC or Access Policies |
from azure.identity import DefaultAzureCredential
from azure.keyvault.secrets import SecretClient
credential = DefaultAzureCredential()
client = SecretClient(vault_url="https://kv-prod.vault.azure.net", credential=credential)
secret = client.get_secret("db-connection-string")
print(secret.value)
10. What is Microsoft Defender for Cloud?
Defender for Cloud (formerly Azure Security Center + Azure Defender) provides:
| Feature | Description |
|---|---|
| Secure Score | Aggregated security posture metric (0–100) |
| Recommendations | Actionable items to improve score |
| Alerts | Real-time threat detection |
| Regulatory compliance | NIST, PCI-DSS, ISO 27001 dashboards |
| CWPP | Cloud Workload Protection for VMs, containers, SQL |
| CSPM | Cloud Security Posture Management |
Compute
11. What are Azure VM sizes and families?
| Family | Optimised for | Example sizes |
|---|---|---|
| A-series | Entry-level, dev/test | A2, A4 |
| B-series | Burstable, variable CPU | B2s, B4ms |
| D-series | General purpose, balanced | D4s_v5, D8as_v5 |
| E-series | Memory-optimised | E8s_v5, E32s_v5 |
| F-series | Compute-optimised | F8s_v2 |
| L-series | Storage-optimised (NVMe) | L8s_v3 |
| M-series | Memory-heavy (up to 11.4 TB RAM) | M416ms_v2 |
| N-series | GPU (NVIDIA) | NC6, ND96asr_A100 |
| HB/HC | HPC — MPI workloads | HB120rs_v3 |
12. What is Azure App Service?
App Service is a fully managed PaaS for hosting web apps, APIs, and mobile backends:
| Feature | Description |
|---|---|
| Runtimes | .NET, Node.js, Python, Java, PHP, Ruby, custom Docker |
| Deployment slots | Blue-green with traffic splitting |
| Auto-scaling | Manual, scheduled, metrics-based |
| Custom domains | Free managed TLS via App Service Certificate |
| Hybrid connections | Access on-prem resources |
| Pricing tiers | Free/Shared/Basic/Standard/Premium/Isolated |
| App Service Plan | Defines region, VM family, scale |
Multiple apps share one App Service Plan (shared compute).
13. What is the difference between Azure Functions and Azure Logic Apps?
| Feature | Azure Functions | Logic Apps |
|---|---|---|
| Type | Code-first serverless | Designer-first workflow |
| Language | C#, JS/TS, Python, Java, PowerShell | JSON workflow + 400+ connectors |
| Trigger types | HTTP, timer, queue, event, blob, custom | HTTP, schedule, connectors |
| Execution | Short-lived (5–10 min default, 60 min max) | Minutes to days (stateful) |
| Cold start | Yes (Consumption plan) | Yes |
| State | Stateless (use Durable Functions for state) | Built-in stateful |
| Pricing | Per execution + GB-s | Per action execution |
| Best for | Code logic, data transforms | Orchestration, integrations |
14. What is Azure Kubernetes Service (AKS)?
AKS is managed Kubernetes on Azure:
- Azure manages the control plane (API server, etcd, scheduler) — free
- You manage node pools (VM Scale Sets)
- Supports system node pools (system components) + user node pools (workloads)
- Integration: Azure CNI, Azure AD workload identity, Azure Monitor, Azure Policy (Gatekeeper)
- Virtual Nodes — burst to Azure Container Instances serverlessly
- KEDA — event-driven autoscaling (built-in)
- Spot node pools — cost savings for fault-tolerant workloads
# Create AKS cluster
az aks create \
--resource-group rg-prod \
--name aks-prod \
--node-count 3 \
--enable-addons monitoring \
--generate-ssh-keys
# Get credentials
az aks get-credentials --resource-group rg-prod --name aks-prod
15. What is Azure Container Apps?
Container Apps is a serverless container platform built on Kubernetes + KEDA + Dapr:
| Feature | Container Apps | AKS |
|---|---|---|
| Kubernetes knowledge | Not required | Required |
| Control plane | Fully managed (invisible) | Managed control plane, visible |
| Scaling | 0–N via KEDA (HTTP, queue, cron) | Manual, HPA, KEDA |
| Dapr integration | Built-in | Manual install |
| Complexity | Low | High |
| Best for | Microservices, APIs, event processing | Custom K8s, complex networking |
Storage
16. What are Azure Storage account types and their use cases?
| Storage Type | Use case | Access tier |
|---|---|---|
| Blob Storage | Unstructured objects (images, videos, backups) | Hot / Cool / Cold / Archive |
| Azure Files | Managed SMB/NFS file shares (lift-and-shift) | — |
| Queue Storage | Reliable messaging between components | — |
| Table Storage | NoSQL key-attribute store (legacy) | — |
| Azure Data Lake Storage Gen2 | Big data analytics (HDFS-compatible) | — |
| Azure Disk Storage | Managed disks for VMs | Premium SSD / Standard SSD / HDD |
17. What are Blob Storage access tiers?
| Tier | Storage cost | Access cost | Minimum days | Use case |
|---|---|---|---|---|
| Hot | High | Low | — | Frequently accessed data |
| Cool | Medium | Medium | 30 days | Infrequent access, backup |
| Cold | Low | High | 90 days | Long-term backup |
| Archive | Lowest | Highest + rehydration time | 180 days | Compliance, rarely accessed |
Rehydration from Archive: Standard (up to 15 hrs) or High priority (under 1 hr, expensive).
18. What is Azure Blob Storage redundancy?
| Option | Full name | Copies | Failover |
|---|---|---|---|
| LRS | Locally Redundant Storage | 3 in one data centre | No |
| ZRS | Zone-Redundant Storage | 3 across AZs in one Region | Automatic (zone failure) |
| GRS | Geo-Redundant Storage | 3 LRS + 3 in paired Region | Manual |
| GZRS | Geo-Zone-Redundant Storage | 3 ZRS + 3 in paired Region | Manual |
| RA-GRS | Read-Access GRS | GRS + read from secondary | Read available, failover manual |
| RA-GZRS | Read-Access GZRS | GZRS + read from secondary | Read available, failover manual |
19. What is Azure Data Lake Storage Gen2 (ADLS Gen2)?
ADLS Gen2 = Blob Storage + hierarchical namespace (HNS):
- Enables directory semantics (rename/move in O(1), not O(n))
- Compatible with HDFS API (Hadoop, Spark, Databricks)
- Fine-grained POSIX-style ACLs at file/directory level
- Used with Azure Synapse Analytics, Azure Databricks, HDInsight
- Tiering still available (Hot/Cool/Archive)
20. What is the difference between Azure Blob Storage and Azure Files?
| Feature | Blob Storage | Azure Files |
|---|---|---|
| Protocol | HTTP/HTTPS, REST | SMB 3.0, NFS 4.1, REST |
| Mount on OS | No (except FUSE) | Yes (Windows, Linux, macOS) |
| Best for | Object storage, streaming, CDN | Lift-and-shift, shared config |
| Access control | SAS, RBAC, ACL | RBAC, on-prem AD integration |
| Snapshots | Blob snapshots | Share snapshots |
| Max size | 4.75 TB per blob (block) | 100 TiB per share |
Networking
21. What is an Azure Virtual Network (VNet)?
A VNet is a logically isolated network in Azure:
- You define the IP address space (CIDR) e.g.
10.0.0.0/16 - Divided into subnets (e.g.
10.0.1.0/24for web tier) - Resources within a VNet communicate privately by default
- Connects to on-premises via VPN Gateway or ExpressRoute
- Connects to other VNets via VNet Peering
- Default outbound internet access (can be restricted)
az network vnet create \
--resource-group rg-prod \
--name vnet-prod \
--address-prefix 10.0.0.0/16 \
--subnet-name snet-web \
--subnet-prefix 10.0.1.0/24
22. What is the difference between NSG and Azure Firewall?
| Feature | Network Security Group (NSG) | Azure Firewall |
|---|---|---|
| Type | L4 stateful packet filter | L4–L7 stateful firewall (managed) |
| Applied to | Subnet or NIC | VNet hub (centrally) |
| Rules | Source/dest IP, port, protocol | + FQDN, URL filtering, threat intel |
| IDPS | No | Yes (Premium SKU) |
| TLS inspection | No | Yes (Premium SKU) |
| Managed by | Customer | Managed service (Azure) |
| Cost | Free (NSG) | ~$1.25/hr + data processing |
| Logs | NSG flow logs (Log Analytics) | Azure Monitor / Event Hub |
Use both: NSGs for subnet-level filtering, Azure Firewall for central egress control.
23. What is Azure Load Balancer vs Application Gateway vs Front Door?
| Feature | Load Balancer | Application Gateway | Azure Front Door |
|---|---|---|---|
| Layer | L4 (TCP/UDP) | L7 (HTTP/HTTPS) | L7 (global HTTP) |
| Scope | Regional | Regional | Global |
| SSL termination | No | Yes | Yes |
| WAF | No | Yes (WAF SKU) | Yes (WAF policy) |
| Path routing | No | Yes | Yes |
| Health probes | TCP/HTTP | HTTP | HTTP |
| Backend | VMs, VMSS, IP | VMs, VMSS, App Service | Any public endpoint |
| Anycast | No | No | Yes (PoP network) |
| Best for | Internal L4 LB | App-aware regional LB | Global HTTP acceleration |
24. What is VNet Peering?
VNet Peering connects two VNets so traffic flows over Azure backbone (private, low latency, high bandwidth):
- Local peering — same Region
- Global peering — different Regions
- Non-transitive by default (A↔B, B↔C does NOT mean A↔C)
- Use Azure Firewall or NVA for hub-spoke transitive routing
- No bandwidth limits, charged per GB transferred
25. What is Azure Private Link / Private Endpoint?
Private Endpoint = a private IP in your VNet for a PaaS service (Blob, SQL, Key Vault, etc.):
- Traffic stays on Azure backbone, never traverses public internet
- Disables public access to the service
- Works across tenants and subscriptions
- Required for compliance (PCI-DSS, HIPAA)
az network private-endpoint create \
--resource-group rg-prod \
--name pe-storage \
--vnet-name vnet-prod \
--subnet snet-private \
--private-connection-resource-id /subscriptions/.../storageAccounts/stprod \
--connection-name psc-storage \
--group-id blob
Databases
26. What is Azure SQL Database vs Azure SQL Managed Instance vs SQL Server on VM?
| Feature | Azure SQL Database | SQL Managed Instance | SQL Server on VM |
|---|---|---|---|
| Type | Fully managed PaaS | Managed instance | IaaS (full control) |
| SQL Server compatibility | Partial (max 99%) | Near 100% | 100% |
| Instance-scoped features | No (DB-scoped only) | Yes (jobs, linked servers, CLR) | Yes |
| Licensing | Included / Hybrid Benefit | Included / Hybrid Benefit | BYOL or included |
| Maintenance | Fully automated | Automated + maintenance window | Manual |
| Cost | Lower | Medium | Higher |
| Best for | Modern cloud-native apps | Lift-and-shift from on-prem | Full SQL Server features |
27. What is Azure Cosmos DB?
Cosmos DB is a globally distributed, multi-model NoSQL database:
| Feature | Description |
|---|---|
| APIs | NoSQL (native), MongoDB, Cassandra, Gremlin (graph), Table |
| Distribution | Turn-key global replication (active-active) |
| Consistency | 5 levels: Strong / Bounded Staleness / Session / Consistent Prefix / Eventual |
| SLAs | 99.999% availability, <10 ms reads/writes at p99 |
| Partitioning | Horizontal auto-partitioning by partition key |
| Indexing | Automatic on all fields by default |
| Pricing | RU/s (Request Units) — abstracted throughput |
| Serverless | Consumption-based, no provisioning |
The 5 consistency levels are unique to Cosmos DB — you trade consistency for latency/throughput.
28. What is the difference between Azure Cache for Redis and Azure Cosmos DB?
| Aspect | Azure Cache for Redis | Cosmos DB |
|---|---|---|
| Type | In-memory cache / message broker | Multi-model distributed database |
| Persistence | Optional (RDB/AOF) | Always persistent |
| Data size | Typically GBs (memory-bound) | PBs |
| Latency | Sub-millisecond | Single-digit milliseconds |
| Best for | Session cache, rate limiting, leaderboards, pub/sub | Operational database, global distribution |
29. What is Azure Synapse Analytics?
Synapse is a unified analytics platform combining:
| Component | Description |
|---|---|
| Dedicated SQL pool | Former Azure SQL Data Warehouse (MPP engine) |
| Serverless SQL pool | On-demand T-SQL queries over Data Lake |
| Apache Spark pool | Managed Spark clusters |
| Synapse Link | Near-real-time analytics over Cosmos DB, Dataverse |
| Synapse Pipelines | ETL/ELT pipelines (ADF-compatible) |
| Power BI integration | Embedded datasets |
Use Synapse for large-scale analytics and data warehousing; use Azure SQL Database for OLTP.
Azure DevOps & Pipelines
30. What is Azure DevOps?
Azure DevOps is a set of developer services:
| Service | Purpose |
|---|---|
| Azure Boards | Agile work tracking (Kanban, Scrum, backlogs) |
| Azure Repos | Git (or TFVC) source control |
| Azure Pipelines | CI/CD for any language, platform, cloud |
| Azure Test Plans | Manual and exploratory testing |
| Azure Artifacts | Package management (NuGet, npm, Maven, PyPI) |
Integrates with GitHub, Jira, Slack, ServiceNow. Free for public projects, 5 free users for private.
31. What is an Azure Pipeline?
An Azure Pipeline is a YAML-defined CI/CD workflow:
trigger:
branches:
include: [main]
pool:
vmImage: ubuntu-latest
stages:
- stage: Build
jobs:
- job: BuildApp
steps:
- task: NodeTool@0
inputs:
versionSpec: '20.x'
- script: npm ci && npm run build
- task: PublishBuildArtifacts@1
inputs:
pathToPublish: dist
- stage: Deploy
dependsOn: Build
jobs:
- deployment: DeployProd
environment: production
strategy:
runOnce:
deploy:
steps:
- task: AzureWebApp@1
inputs:
appType: webApp
appName: app-prod
Key concepts: stages (major phases), jobs (parallel units), steps (tasks/scripts), environments (approval gates), service connections (auth to Azure).
32. What is the difference between Azure DevOps and GitHub Actions?
| Feature | Azure DevOps Pipelines | GitHub Actions |
|---|---|---|
| Source control | Azure Repos or GitHub | GitHub |
| Syntax | YAML (azure-pipelines.yml) |
YAML (.github/workflows/) |
| Marketplace | Extensions (~1,000) | Actions (~20,000) |
| Environments | Approval gates, deployment | Environments with protection rules |
| Self-hosted agents | Yes | Yes (runners) |
| Free tier | 1,800 min/month | 2,000 min/month |
| Best for | Enterprise .NET/Microsoft ecosystem | Open source, GitHub-native |
Monitoring & Management
33. What is Azure Monitor?
Azure Monitor is the unified monitoring platform for Azure:
| Component | Purpose |
|---|---|
| Metrics | Time-series numeric data (CPU %, request rate) |
| Logs | Log Analytics Workspace (Kusto Query Language) |
| Alerts | Rules triggering actions (email, webhook, ITSM) |
| Application Insights | APM for web apps (traces, dependencies, exceptions) |
| Container Insights | AKS and container monitoring |
| VM Insights | VM health, processes, network |
| Network Watcher | Network monitoring, packet capture |
| Workbooks | Interactive dashboards |
34. What is Application Insights?
Application Insights is an APM (Application Performance Monitoring) service:
- Auto-collects: request rates, response times, failure rates, dependency calls
- Distributed tracing with correlation IDs across services
- Live Metrics — real-time stream
- Availability tests — ping tests from global PoPs
- Smart detection — anomaly detection (spike in failures, unusual latency)
- SDKs for .NET, Java, Node.js, Python, JavaScript
from applicationinsights import TelemetryClient
tc = TelemetryClient('<instrumentation_key>')
tc.track_event('UserSignup', {'plan': 'pro'})
tc.track_metric('OrderValue', 99.99)
tc.flush()
35. What is Azure Policy?
Azure Policy enforces organizational standards and compliance at scale:
| Feature | Description |
|---|---|
| Policy definition | JSON rule (e.g., "VMs must use Premium SSD") |
| Initiative | Group of policies (e.g., "PCI-DSS compliance") |
| Assignment | Policy/initiative applied at scope |
| Effect | Deny / Audit / Append / Modify / DeployIfNotExists |
| Compliance | Dashboard showing non-compliant resources |
| Remediation | Fix existing non-compliant resources |
Built-in initiatives: CIS Azure Benchmark, NIST SP 800-53, PCI-DSS, Azure Security Benchmark.
AI & Machine Learning
36. What is Azure OpenAI Service?
Azure OpenAI provides access to OpenAI models (GPT-4o, GPT-4, o1, DALL-E, Whisper, text-embedding) via Azure infrastructure:
| Azure OpenAI | OpenAI API |
|---|---|
| Data stays in Azure region | Data sent to OpenAI US |
| Azure AD authentication | API key |
| Private Endpoint support | No |
| Azure compliance (HIPAA, PCI) | Limited |
| Content filtering | Configurable |
| Fine-tuning | Yes (GPT-3.5-Turbo, GPT-4) |
| Committed throughput (PTU) | Yes |
from openai import AzureOpenAI
client = AzureOpenAI(
azure_endpoint="https://my-resource.openai.azure.com",
api_key="<azure-api-key>",
api_version="2024-10-21"
)
response = client.chat.completions.create(
model="gpt-4o", # deployment name
messages=[{"role": "user", "content": "Explain Azure RBAC"}]
)
37. What is Azure Machine Learning?
Azure ML is an enterprise-grade ML platform:
| Component | Description |
|---|---|
| Studio (web UI) | Visual interface for all ML tasks |
| Compute clusters | Training compute (auto-scales to 0) |
| Compute instances | Dev VMs (Jupyter) |
| Datasets | Versioned data assets |
| Experiments | Training runs with metrics logging |
| Models | Registered model versions |
| Endpoints | Online (real-time) / Batch inference |
| Pipelines | Reusable ML workflows |
| MLflow integration | Experiment tracking and model registry |
| Responsible AI | Explainability, fairness, error analysis dashboards |
Architecture & Best Practices
38. What is the Azure Well-Architected Framework?
Five pillars for cloud-native design:
| Pillar | Key practices |
|---|---|
| Reliability | Multi-AZ/region, chaos testing, retry policies, health endpoints |
| Security | Zero trust, least privilege, defence in depth, encryption |
| Cost Optimization | Right-sizing, Reserved Instances, auto-scaling, tagging |
| Operational Excellence | IaC, GitOps, observability, runbooks, blameless post-mortems |
| Performance Efficiency | Autoscaling, caching, CDN, async messaging, profiling |
39. What is a Hub-Spoke network topology in Azure?
Hub-Spoke is the most common Azure network architecture:
┌──────────────────┐
│ Hub VNet │
│ Azure Firewall │
│ VPN/ExpressRoute│
│ Shared Services │
└────────┬─────────┘
│ VNet Peering (non-transitive)
┌────────────┼─────────────┐
│ │ │
┌─────▼────┐ ┌─────▼────┐ ┌─────▼────┐
│ Spoke 1 │ │ Spoke 2 │ │ Spoke 3 │
│ (Dev) │ │ (Staging)│ │ (Prod) │
└──────────┘ └──────────┘ └──────────┘
Hub contains shared services (firewall, DNS, VPN gateway). Spokes route internet egress through the hub. Traffic between spokes goes via hub firewall (hairpinning).
Azure Virtual WAN automates this pattern at scale.
40. What is Azure Landing Zone?
A Landing Zone is a pre-configured Azure environment that implements enterprise best practices:
- Management Group hierarchy
- Policy assignments (Azure Policy Initiatives)
- RBAC roles
- Hub network (connectivity subscription)
- Log Analytics workspace
- Azure Defender enabled
Deployed via Enterprise-Scale Landing Zone (Bicep/Terraform templates). Platform team manages hub; application teams manage spoke subscriptions.
Additional Questions
41. What is Azure ExpressRoute?
ExpressRoute is a private, dedicated circuit between on-premises and Azure:
| Feature | ExpressRoute | VPN Gateway |
|---|---|---|
| Connection | Private (via carrier) | Public internet (encrypted) |
| Bandwidth | 50 Mbps–100 Gbps | Up to 10 Gbps (UltraPerformance) |
| Latency | Predictable, low | Variable (internet-dependent) |
| SLA | 99.95% | 99.9% |
| Cost | High (carrier + Azure) | Lower |
| Best for | Hybrid cloud, compliance, high bandwidth | Branch offices, backup connectivity |
ExpressRoute doesn't traverse the public internet — required for financial services and healthcare.
42. What is Azure Service Bus vs Azure Event Hub vs Azure Event Grid?
| Feature | Service Bus | Event Hub | Event Grid |
|---|---|---|---|
| Pattern | Message queue / pub-sub | Event streaming | Event routing |
| Protocol | AMQP, HTTP | AMQP, Kafka, HTTP | HTTP/webhooks |
| Ordering | Yes (sessions) | Per partition | No guarantee |
| Replay | Dead-letter queue | Retention (up to 90 days) | No |
| Max message | 256 KB (Standard), 100 MB (Premium) | 1 MB | 1 MB |
| Throughput | Medium | Very high (millions/sec) | Medium |
| Best for | Reliable messaging, workflows, RPC | Telemetry, log streaming, Kafka | Azure events, serverless triggers |
43. What is Azure Logic Apps?
Logic Apps is a visual workflow orchestration service:
- 400+ connectors: Office 365, Salesforce, SAP, Oracle, Twitter, SQL, Blob
- Built-in connectors: HTTP, schedule, condition, loops, variables, scope
- Stateful workflows with checkpointing (Standard plan)
- Integration Service Environment (ISE) — VNet-injected for secure integration
- Trigger types: HTTP request, schedule, connector events (new email, new row)
Pricing: Consumption (per action) or Standard (per app, vCores).
44. What is Azure Bicep?
Bicep is a domain-specific language (DSL) that transpiles to ARM JSON templates:
param location string = resourceGroup().location
param storageAccountName string
resource storage 'Microsoft.Storage/storageAccounts@2023-01-01' = {
name: storageAccountName
location: location
sku: {
name: 'Standard_LRS'
}
kind: 'StorageV2'
properties: {
accessTier: 'Hot'
}
}
output storageId string = storage.id
Bicep vs Terraform:
- Bicep: Azure-only, first-class ARM support, free, no state file
- Terraform: Multi-cloud, large ecosystem, state management required
45. What is Azure Active Directory B2C?
Azure AD B2C is a customer identity solution for external-facing apps:
- Supports email/password, social login (Google, Facebook, Apple, GitHub)
- Custom user flows (sign-up, sign-in, password reset, profile edit)
- Custom policies (Identity Experience Framework) for advanced scenarios
- Scales to hundreds of millions of users
- Separate tenant from corporate Azure AD
- Integration: OIDC, SAML, OAuth 2.0
- Free tier: 50,000 MAU/month
46. What is Azure Cost Management?
Azure Cost Management + Billing provides:
| Feature | Description |
|---|---|
| Cost analysis | Breakdowns by resource, tag, subscription, service |
| Budgets | Alerts when spending reaches thresholds |
| Advisor recommendations | Right-sizing, reserved instance opportunities |
| Exports | Automated exports to Storage Account |
| Invoices | Download and manage billing |
Cost optimisation strategies:
- Reserved Instances — 1–3 year commitment, up to 72% savings
- Azure Hybrid Benefit — reuse Windows Server / SQL Server licenses
- Azure Spot VMs — up to 90% savings for interruptible workloads
- Dev/test subscriptions — discounted rates for non-production
- Auto-shutdown — stop VMs outside business hours
47. What is Azure Site Recovery (ASR)?
ASR is a disaster recovery and migration service:
- Replicates VMs (Azure, VMware, Hyper-V, physical) to a target Azure Region
- RPO: typically 30 seconds for Azure VMs
- RTO: typically <2 hours
- Test failover: non-disruptive DR drills in isolated VNet
- Failover + failback supported
- Used for both DR and lift-and-shift migration
48. What is Azure Arc?
Azure Arc extends Azure management to non-Azure environments:
| Resource Type | What Arc enables |
|---|---|
| Servers (Windows/Linux) | Azure Policy, Azure Monitor, Defender, RBAC on on-prem VMs |
| Kubernetes clusters | GitOps, Azure Policy, Azure Monitor, any K8s cluster |
| SQL Server | Azure security, inventory, best practices on on-prem SQL |
| Data services | Azure SQL MI and PostgreSQL running on-prem/other clouds |
| App Service | App Service and Functions on any K8s cluster |
49. What is the difference between SLA, SLO, and RPO/RTO in Azure context?
| Term | Definition | Azure example |
|---|---|---|
| SLA | Microsoft's uptime guarantee | Azure SQL Database: 99.99% |
| SLO | Your internal target (stricter than SLA) | Your target: 99.95% |
| RPO | Recovery Point Objective — max data loss | Cosmos DB multi-region: ~0 |
| RTO | Recovery Time Objective — max downtime | AKS with replicas: minutes |
Composite SLA: if App Service = 99.95% AND SQL = 99.99%, combined = 99.95% × 99.99% ≈ 99.94%.
50. What is the Azure Shared Responsibility Model for different service types?
| Responsibility | On-prem | IaaS (VM) | PaaS (App Service) | SaaS (Office 365) |
|---|---|---|---|---|
| Physical security | Customer | Azure | Azure | Azure |
| Network controls | Customer | Customer | Azure | Azure |
| OS patching | Customer | Customer | Azure | Azure |
| Middleware | Customer | Customer | Azure | Azure |
| Runtime | Customer | Customer | Azure | Azure |
| Application | Customer | Customer | Customer | Azure |
| Identity | Customer | Customer | Customer | Shared |
| Data | Customer | Customer | Customer | Customer |
The higher you move up the stack (IaaS → PaaS → SaaS), the less you manage.
Common mistakes
| Mistake | Problem | Fix |
|---|---|---|
| Putting all resources in one subscription | Blast radius, quota issues | Separate dev/staging/prod subscriptions |
| Using Resource Group as billing unit | Resource Groups don't map to invoices | Use tags + Cost Management |
| Ignoring NSG flow logs | No network visibility | Enable flow logs → Log Analytics |
| Owner role for service principals | Over-privileged | Use Contributor or custom least-privilege role |
| Not using Managed Identity | Credentials in code | Use System-assigned Managed Identity |
| Single-region deployment | No DR | Multi-AZ minimum, multi-region for critical apps |
| Ignoring Azure Advisor | Missed savings/security | Review weekly |
| Using Azure AD B2C for internal users | Wrong product | Use Azure AD for employees |
Azure vs AWS vs GCP
| Feature | Azure | AWS | GCP |
|---|---|---|---|
| Market share | ~25% | ~31% | ~11% |
| Compute | VMs, App Service, AKS, Functions | EC2, ECS, EKS, Lambda | GCE, GKE, Cloud Run, Cloud Functions |
| Managed K8s | AKS | EKS | GKE |
| Serverless | Azure Functions, Container Apps | Lambda, Fargate | Cloud Run, Cloud Functions |
| Identity | Azure AD (Entra ID) | IAM | Cloud IAM |
| Managed DB | Azure SQL, Cosmos DB | RDS, DynamoDB | Cloud SQL, Bigtable, Spanner |
| Object storage | Blob Storage | S3 | Cloud Storage |
| CDN | Azure Front Door | CloudFront | Cloud CDN |
| AI/ML | Azure OpenAI, Azure ML | SageMaker, Bedrock | Vertex AI |
| Hybrid | Azure Arc, ExpressRoute | Outposts, Direct Connect | Anthos, Interconnect |
| Best for | Microsoft ecosystem, enterprise | Broadest services, start-ups | Data/ML, Kubernetes |
6 FAQ
Q: What Azure certification should I study for first? A: AZ-900 (Azure Fundamentals) for non-technical roles; AZ-104 (Azure Administrator) or AZ-204 (Azure Developer) for technical roles. Solutions architects: AZ-305.
Q: What is the difference between Azure AD and on-premises Active Directory? A: Azure AD is cloud-native (OAuth 2.0, OIDC, SAML — no LDAP/Kerberos). On-prem AD uses Kerberos/LDAP. Azure AD Connect syncs on-prem identities to Azure AD. Azure AD DS provides managed domain services (LDAP, Kerberos) in the cloud.
Q: Is Cosmos DB free? A: There's a free tier (1,000 RU/s + 25 GB per account), and a serverless mode where you pay only for RUs consumed. No continuous free tier beyond the free account limits.
Q: When should I use Azure Functions vs App Service? A: Use Azure Functions for event-driven, short-lived tasks (HTTP trigger, queue consumer, timer). Use App Service for long-running web apps and APIs that need persistent connections, WebSockets, or predictable latency.
Q: What is the difference between Azure DevOps and GitHub? A: GitHub is the developer platform (code, PRs, Issues, Actions CI/CD). Azure DevOps is a broader suite (Boards, Repos, Pipelines, Artifacts, Test Plans). Both are Microsoft products and integrate with each other. New projects often choose GitHub; large enterprises with existing Azure DevOps stay there.
Q: How do Azure Reserved Instances work? A: You commit to using a specific VM family and region for 1 or 3 years, paying upfront or monthly. You save up to 72% vs pay-as-you-go. Reservations are flexible — they apply to matching VM sizes in the same family. Best for stable workloads running >80% of the time.