Guides › Cloud Computing & DevOps
Cloud Computing Roadmap: Beginner to Advanced
Learn cloud computing from beginner to advanced with detailed explanations of cloud models, AWS, Microsoft Azure, Google Cloud, compute, storage, networking, databases, IAM, serverless, containers, cloud security, cost optimization, architecture, certifications, hands-on labs, and real-world project
Cloud Computing Roadmap: Beginner to Advanced
1. Introduction to Cloud Computing
Cloud computing is the delivery of computing resources—such as servers, storage, databases, networking, analytics, and software—over the internet. Instead of purchasing and maintaining all hardware yourself, you can use resources from a cloud provider and pay according to the pricing model.
Why Cloud Computing Matters
- Scalability: Increase or decrease resources as demand changes.
- Availability: Design applications to remain available during failures.
- Global reach: Deploy applications closer to users around the world.
- Managed services: Let providers maintain parts of the infrastructure.
- Cost flexibility: Avoid purchasing all infrastructure upfront.
- Automation: Create and manage infrastructure through APIs and code.
Cloud computing does not automatically make an application secure, inexpensive, or highly available. Those outcomes depend on architecture, configuration, monitoring, and operational practices.
2. Cloud Service Models
Infrastructure as a Service (IaaS)
IaaS provides fundamental infrastructure such as virtual machines, disks, and networks. You usually manage the operating system, installed software, application, and much of its configuration.
Examples include Amazon EC2, Azure Virtual Machines, and Google Compute Engine.
Typical use cases: Custom servers, legacy application migration, and workloads that require operating-system control.
Platform as a Service (PaaS)
PaaS provides a managed application platform. The provider handles much of the infrastructure and runtime management, allowing developers to focus more on application code.
Examples include Azure App Service, Google App Engine, and managed application platforms offered by cloud providers.
Typical use cases: Web applications, APIs, and applications that need simplified deployment.
Software as a Service (SaaS)
SaaS provides a complete application accessed through a browser or client.
Examples include hosted email, online document editors, and customer relationship management software.
Typical use cases: Business productivity, collaboration, and ready-to-use applications.
Function as a Service (FaaS)
FaaS runs individual functions in response to events. The platform manages much of the server provisioning and scaling.
Examples include AWS Lambda, Azure Functions, and Google Cloud Run functions.
Typical use cases: File-processing events, scheduled jobs, webhooks, and lightweight APIs.
3. Cloud Deployment Models
- Public cloud: Infrastructure and services are provided over a public cloud platform.
- Private cloud: Cloud-like infrastructure is dedicated to one organization.
- Hybrid cloud: Workloads or data operate across private infrastructure and public cloud.
- Multicloud: An organization uses services from multiple cloud providers.
Choose a deployment model based on requirements such as compliance, latency, cost, skills, data location, and operational complexity—not simply because one model sounds more advanced.
4. Major Cloud Providers
Amazon Web Services (AWS)
Important services to learn:
- Compute: EC2, Lambda, ECS, EKS.
- Storage: S3, EBS, EFS.
- Networking: VPC, Route 53, Elastic Load Balancing, CloudFront.
- Databases: RDS, DynamoDB.
- Identity and security: IAM, KMS, CloudTrail.
- Monitoring: CloudWatch.
- Infrastructure as code: CloudFormation and AWS CDK.
Microsoft Azure
Important services to learn:
- Compute: Azure Virtual Machines, App Service, Azure Functions.
- Storage: Blob Storage, Files.
- Networking: Virtual Network, Load Balancer, Application Gateway.
- Databases: Azure SQL Database and managed database services.
- Identity: Microsoft Entra ID and Azure RBAC.
- Monitoring: Azure Monitor.
- Infrastructure as code: Bicep and Terraform.
Google Cloud
Important services to learn:
- Compute: Compute Engine, Cloud Run, Google Kubernetes Engine.
- Storage: Cloud Storage.
- Networking: VPC, Cloud Load Balancing, Cloud DNS.
- Databases: Cloud SQL, Firestore, Bigtable.
- Identity: Cloud IAM.
- Monitoring: Cloud Monitoring and Cloud Logging.
- Data analytics: BigQuery.
Which Provider Should You Learn First?
Choose one provider for your initial hands-on learning. AWS is a common starting point because of its broad service ecosystem, but Azure may fit Microsoft-oriented workplaces, and Google Cloud may fit teams using its data and Kubernetes services. The fundamentals transfer between providers, though service names, APIs, permissions, and pricing differ.
5. Prerequisites
Learn these fundamentals before attempting complex cloud architectures.
Computer and Operating System Basics
Understand files, processes, memory, storage, permissions, environment variables, services, and logs. Practice basic Linux terminal commands.
Useful commands:
pwd
ls -la
cd /path/to/directory
mkdir cloud-lab
cp source.txt destination.txt
mv old.txt new.txt
cat file.txt
grep "ERROR" application.log
find . -name "*.log"
Learn how to read command help and manual pages. Use commands carefully, especially those that delete files or change permissions.
Networking Fundamentals
Understand:
- IP addresses and subnet masks.
- Public versus private IP addresses.
- IPv4 and basic IPv6 concepts.
- TCP versus UDP.
- DNS resolution.
- HTTP and HTTPS.
- Ports and firewalls.
- Routing, gateways, and NAT.
- VPNs and private connectivity.
- Load balancing and reverse proxies.
Common ports include 22 for SSH, 53 for DNS, 80 for HTTP, 443 for HTTPS, and 5432 for PostgreSQL. These are defaults, not universal requirements.
Git and Programming
Learn Git basics, branches, commits, pull requests, and conflict resolution. Basic scripting with Python, Bash, or PowerShell helps automate repetitive tasks.
Virtualization
Understand virtual machines, hypervisors, operating-system images, snapshots, and the difference between virtual machines and containers.
6. Compute Services
Virtual Machines
A virtual machine (VM) is a software-based computer with allocated CPU, memory, storage, and networking.
Important concepts:
- Machine or instance types.
- CPU and memory sizing.
- Operating-system images.
- Boot disks and additional disks.
- SSH keys and remote access.
- Security groups or equivalent firewall rules.
- Startup scripts and instance metadata.
- Backups and snapshots.
- Availability zones and placement.
- Auto Scaling.
Practice task: Create a small Linux VM, connect securely, install a web server, serve a test page, and remove the VM when finished.
Containers
Containers package an application and its dependencies into a portable unit. They usually share the host operating-system kernel and are lighter than full virtual machines.
Learn Docker images, containers, registries, networks, volumes, and container security. Study the separate Docker guide in this category.
Serverless Computing
Serverless platforms manage much of the underlying infrastructure. You still need to consider execution limits, concurrency, permissions, retries, logging, and cost.
Learn event triggers, function handlers, environment configuration, secrets management, and dead-letter or retry strategies where supported.
7. Cloud Storage
Object Storage
Object storage stores data as objects inside containers such as buckets. It is suitable for images, videos, backups, logs, static assets, and data archives.
Examples: Amazon S3, Azure Blob Storage, and Google Cloud Storage.
Learn:
- Buckets and object keys.
- Access policies.
- Versioning.
- Lifecycle rules.
- Encryption.
- Replication.
- Retention and deletion controls.
- Static website hosting limitations and security.
- Signed URLs or equivalent temporary access mechanisms.
Important: Never make a storage bucket public just to make an upload work. Prefer private storage and narrowly scoped access.
Block Storage
Block storage provides disk-like volumes for VMs and some applications. Learn volume attachment, filesystem creation, snapshots, encryption, performance tiers, and backup recovery.
File Storage
File storage offers shared filesystem access. It can be useful for shared application files, home directories, and legacy workloads.
Backup and Recovery
Understand the difference between replication and backup. Replication may copy accidental deletion or corruption, so it is not a complete substitute for independent backups.
Define recovery objectives:
- RPO: How much data loss, measured in time, is acceptable?
- RTO: How long can recovery take?
Test restores regularly rather than assuming that a successful backup job guarantees recoverability.
8. Cloud Networking
Virtual Private Cloud (VPC)
A VPC or equivalent virtual network provides an isolated network environment for cloud resources.
Learn:
- CIDR blocks and subnet planning.
- Public and private subnets.
- Route tables.
- Internet gateways.
- NAT gateways or NAT services.
- Network access control lists.
- Security groups or stateful firewall rules.
- VPC peering.
- Private endpoints and service endpoints.
- VPN and dedicated connectivity.
- Flow logs.
Public and Private Subnets
A subnet is not necessarily public simply because it has a public IP address. Routing and access-control configuration also matter.
A typical design places a public load balancer in public-facing subnets and application servers in private subnets. Databases are normally kept private and accessible only from approved application resources.
DNS and TLS
DNS maps names to records such as IP addresses and aliases. TLS protects data in transit and helps authenticate the server.
Learn domain records, certificates, certificate renewal, HTTPS redirects, and secure cipher configuration at a conceptual level.
Load Balancing and CDN
A load balancer distributes requests among targets. A content delivery network caches eligible content closer to users.
Consider health checks, timeouts, sticky sessions, caching rules, invalidation, and protecting origin services from direct public access.
9. Cloud Databases
Relational Databases
Relational databases store structured data in tables and support SQL.
Examples: Amazon RDS, Azure SQL Database, Cloud SQL, and other managed relational database services.
Learn schema design, indexes, backups, replication, connection pooling, high availability, patching, and access controls.
NoSQL Databases
NoSQL databases support models such as key-value, document, wide-column, or graph data.
Examples include DynamoDB and Firestore. Choose a database according to query patterns, consistency requirements, data relationships, scale, and operational constraints.
Database Reliability
- Keep databases private where possible.
- Use strong authentication and least-privilege access.
- Encrypt data at rest and in transit.
- Monitor connections, storage, latency, and query performance.
- Test backup restoration.
- Plan schema migrations and rollback.
- Avoid storing credentials in source code.
10. Identity and Access Management (IAM)
IAM controls who or what can access resources and which actions they may perform.
Learn:
- Users, groups, roles, service identities, and policies.
- Authentication versus authorization.
- Role-based access control (RBAC).
- Attribute-based access control (ABAC).
- Least privilege.
- Temporary credentials.
- Multi-factor authentication (MFA).
- Access reviews and audit logs.
- Workload identity and secret rotation.
Prefer managed identities, instance roles, or short-lived credentials over long-lived access keys where supported.
Never commit cloud credentials to Git. If a key is exposed, revoke or rotate it promptly and review logs for misuse.
11. Cloud Security
A cloud provider and customer typically share security responsibilities. The exact boundary depends on the service model.
Key practices:
- Enable MFA for privileged accounts.
- Avoid using root or owner accounts for routine tasks.
- Apply least-privilege permissions.
- Patch operating systems and application dependencies.
- Encrypt sensitive data in transit and at rest.
- Store secrets in a managed secrets service.
- Restrict inbound network access.
- Enable relevant audit and activity logs.
- Monitor for suspicious authentication and configuration changes.
- Maintain tested backups and incident-response procedures.
- Scan infrastructure and application dependencies.
- Review data retention and compliance requirements.
Use a dedicated test account or sandbox, set spending alerts, and remove temporary resources after labs.
12. High Availability and Disaster Recovery
High availability reduces downtime from expected component failures. Disaster recovery prepares for larger disruptions such as regional failures or data loss.
Learn:
- Availability zones and regions.
- Redundant application instances.
- Health checks and load balancing.
- Stateless application design.
- Database failover.
- Replication and backup restoration.
- Multi-region trade-offs.
- Failover testing.
- Recovery runbooks.
Multi-region architecture is not always necessary. It increases complexity and cost, so select it when business recovery requirements justify it.
13. Monitoring, Logging, and Observability
Monitoring helps teams detect problems and understand system behavior.
Track four useful categories:
- Metrics: Numerical measurements such as CPU utilization, memory use, request count, and latency.
- Logs: Event records that help explain what happened.
- Traces: Request journeys across distributed services.
- Alerts: Notifications triggered by meaningful conditions.
Create alerts for symptoms that matter to users, such as high error rates or elevated latency. Avoid noisy alerts that teams learn to ignore.
Use dashboards, structured logs, correlation IDs, retention policies, and access controls. Do not log passwords, tokens, or unnecessary sensitive information.
14. Infrastructure as Code (IaC)
Infrastructure as Code defines infrastructure in version-controlled files rather than relying entirely on manual console actions.
Tools include Terraform, OpenTofu, CloudFormation, Bicep, and provider-specific development kits.
Learn:
- Declarative versus imperative approaches.
- Providers and resources.
- Variables and outputs.
- State and state locking.
- Modules and reusable components.
- Plan, review, apply, and destroy workflows.
- Environment separation.
- Drift detection.
- Secret handling.
- Policy checks and automated validation.
Example Terraform workflow:
terraform init
terraform fmt -check
terraform validate
terraform plan
terraform apply
Review the plan before applying it. Use remote state with appropriate access controls and locking support. Do not publish state files containing sensitive values.
15. Cloud Cost Optimization
Cloud bills can grow unexpectedly if resources are oversized, forgotten, or transferred across networks excessively.
Best practices:
- Tag resources by project, owner, and environment.
- Set budgets and billing alerts.
- Delete unused lab resources.
- Right-size compute and databases.
- Use autoscaling where appropriate.
- Select storage tiers based on access patterns.
- Review data transfer and NAT costs.
- Evaluate commitments only after understanding steady usage.
- Track cost per user, transaction, or workload.
- Shut down non-production resources when not needed.
Budgets and alerts help detect spending; they may not automatically stop resources. Confirm the controls available in your provider.
16. Cloud Architecture and Design
When designing a cloud system, start with requirements instead of choosing services at random.
Ask:
- Who are the users, and where are they located?
- What traffic and data volume are expected?
- What availability and recovery targets are required?
- What data is sensitive?
- What latency is acceptable?
- What budget and compliance rules apply?
- Who will deploy, monitor, and support the system?
Example: A Basic Web Application
A simple architecture may include:
- DNS pointing to a public endpoint.
- HTTPS termination at a managed load balancer or edge service.
- Application instances or a managed application platform.
- A private relational database.
- Object storage for uploaded files.
- IAM roles for application access.
- Centralized logs, metrics, and alerts.
- Backups and a tested recovery procedure.
- Infrastructure defined in code.
- A CI/CD pipeline for reviewed releases.
This is a reference pattern, not a universal architecture. Smaller applications may need fewer components.
17. Hands-On Cloud Labs
Complete these labs in order. Check provider-specific instructions and pricing before creating resources.
Lab 1: Launch a Virtual Machine
- Create a small Linux VM in a test environment.
- Configure SSH access using an approved key or managed connection.
- Allow only the required administrative access.
- Install a web server.
- Verify the test page.
- Review system and application logs.
- Delete the VM and any unattached resources.
Lab 2: Host Static Content
- Create a storage bucket.
- Upload a test HTML file.
- Keep storage private by default.
- Configure a supported secure delivery method if public viewing is required.
- Test access and permissions.
- Remove the lab content and resources.
Lab 3: Build a Virtual Network
- Create a VPC and non-overlapping CIDR ranges.
- Add public and private subnets.
- Configure routes and firewall rules.
- Test allowed and denied connections.
- Review network logs and clean up.
Lab 4: Deploy a Database
- Create a small managed database in a private network.
- Restrict access to an approved application or test client.
- Create sample data.
- Test a backup and restoration procedure.
- Delete the database after the lab.
Lab 5: Deploy an Application
- Build a small API.
- Store its source code in Git.
- Configure environment variables without committing secrets.
- Deploy to a managed service or VM.
- Add logs and health checks.
- Document rollback and cleanup steps.
18. Real-World Portfolio Projects
Project 1: Static Portfolio Website
Use object storage and a CDN or supported static hosting platform. Configure a custom domain and HTTPS. Document access controls, caching, and deployment steps.
Project 2: Three-Tier Web Application
Build a frontend, backend API, and relational database. Place services in suitable network layers and restrict database access.
Project 3: Serverless File Processor
Upload a file to object storage and trigger a function that validates or processes it. Include retry handling, logs, permissions, and error reporting.
Project 4: Cloud Monitoring Dashboard
Collect application and infrastructure metrics. Create dashboards and alerts for errors, latency, and resource utilization.
Project 5: Infrastructure as Code
Create a reusable, parameterized infrastructure configuration. Include validation, reviewed plans, environment separation, and a cleanup procedure.
Project 6: Backup and Recovery Demonstration
Back up sample application data, simulate a recoverable failure in a safe test environment, restore the data, and record the measured recovery time.
For every project, include an architecture diagram, setup instructions, security decisions, screenshots where useful, estimated cost, known limitations, and cleanup steps.
19. Certifications and Career Preparation
Certification names, exam codes, and availability can change. Always check the provider's official certification page before booking an exam.
Possible learning paths:
- Cloud fundamentals: Learn shared responsibility, service models, billing, and basic architecture.
- Cloud administrator or operations: Focus on identity, networking, compute, storage, monitoring, and recovery.
- Cloud developer: Focus on APIs, managed services, serverless, deployment, and application security.
- Cloud architect: Focus on trade-offs, reliability, performance, security, governance, and cost.
- Cloud security: Focus on IAM, threat modeling, detection, data protection, and incident response.
- DevOps or platform engineering: Add Git, CI/CD, containers, IaC, automation, and observability.
A certificate can help structure learning, but it does not replace hands-on experience.
20. A 12-Week Learning Plan
| Week | Learning focus |
|---|---|
| 1 | Linux, Git, and basic scripting |
| 2 | Networking, DNS, HTTP, and TLS |
| 3 | Cloud fundamentals, regions, zones, and billing |
| 4 | Virtual machines, disks, and images |
| 5 | Object storage, file storage, and backups |
| 6 | Virtual networks, subnets, routes, and firewalls |
| 7 | IAM, secrets, and cloud security |
| 8 | Managed databases and application connectivity |
| 9 | Monitoring, logging, alerts, and recovery |
| 10 | Containers and introductory serverless |
| 11 | Infrastructure as Code and CI/CD fundamentals |
| 12 | Build, document, secure, and present a portfolio project |
Adjust the schedule to your experience and available study time. Repeat labs until you can explain the architecture and recreate it without copying every step.
21. Common Beginner Mistakes
- Learning many providers simultaneously without mastering one.
- Creating expensive resources without checking prices.
- Leaving test VMs, disks, IP addresses, or databases running.
- Giving broad permissions to make an access error disappear.
- Putting databases directly on the public internet.
- Hardcoding credentials in code.
- Confusing replication with backup.
- Ignoring logs and monitoring.
- Applying infrastructure changes without reviewing the plan.
- Following tutorials without understanding the architecture.
- Building complicated multi-region systems before learning the basics.
- Failing to document cleanup and recovery.
22. Final Learning Checklist
- Explain IaaS, PaaS, SaaS, and serverless.
- Explain regions, availability zones, and shared responsibility.
- Use a Linux terminal and Git.
- Deploy and secure a small virtual machine.
- Explain object, block, and file storage.
- Design public and private subnets.
- Explain DNS, HTTPS, load balancing, and firewalls.
- Configure least-privilege identity access.
- Connect an application to a managed database.
- Set up logs, metrics, and useful alerts.
- Explain RPO, RTO, backup, and disaster recovery.
- Define infrastructure using code.
- Estimate and monitor cloud costs.
- Deploy at least one documented project.
- Demonstrate cleanup and recovery.
Conclusion
Cloud computing is a combination of infrastructure, networking, software, security, automation, and operational discipline. Learn one provider through hands-on labs, understand the underlying concepts, and build projects that demonstrate reliability, security, and cost awareness.
More guides
- How to Use Codify: A 7-Day Start Plan — New here? Pick a roadmap, make a study plan, start the DSA sheet, revise with notes, practise for placements and test yourself — one step a day, then a daily routine.
- Infosys Preparation Guide: Online Test, Pseudocode and Interview — Roles (SE, DSE, SP), what each section of the Infosys online test checks, how to crack pseudocode and reasoning, a 4-week plan and the interview.
- Wipro Elite NLTH Preparation: Aptitude, Essay and Coding — The Wipro fresher test round by round — quant, logical and verbal, the written essay, two coding problems — plus a 3-week plan and interview tips.
- Accenture Preparation Guide: Cognitive, Technical, Coding and Communication — Every Accenture round explained — cognitive, MS Office/networking/cloud technical MCQs, pseudocode, coding and the spoken-English test — with a 4-week plan.