1. Ensure you have an up-to-date command-line interface tool installed: the AWS CLI, Azure CLI or gcloud.
2. Confirm you have adequate permissions to spin up compute instances and create functions or containers.
3. Review your provider’s free tier and billing alerts to avoid unexpected costs.

• Hardware to virtual machines (IaaS)
• Runtimes to managed platforms (PaaS)
• Applications delivered over the web (SaaS)
• Containers managed at scale (CaaS)
• Event-driven code execution (FaaS)

• Migrate existing applications into the cloud without major rewrites
• Run custom virtual machines with full operating system control
• Build test and development environments quickly

• Pros: Maximum flexibility, full control over OS and middleware
• Cons: You must manage patching, scaling and security yourself

• AWS EC2
• Google Compute Engine
• Azure Virtual Machines

# Example: Launch a t3.micro EC2 instance
aws ec2 run-instances \
  --image-id ami-0abcdef1234567890 \
  --instance-type t3.micro \
  --key-name mykeypair

• Deploy web applications without provisioning servers
• Focus on code and configuration rather than OS maintenance
• Rapidly scale applications during traffic spikes with minimal effort

• Pros: Fast time to market, built-in load balancing, scaling, monitoring and health checks
• Cons: Less control over the underlying environment, and potential provider lock-in

• Heroku
• Google App Engine
• Azure App Service

• Use Enterprise software without installation or updates
• Outsource non-core functions like CRM, collaboration or accounting
• Start using new features in minutes

• Pros: No infrastructure or platform to manage; automatic updates
• Cons: Limited customization; subscription costs can add up

• GitHub for code hosting
• Slack for team communication
• Salesforce for customer relationship management

• Run microservices with simplified container scheduling
• Standardize development and production environments
• Scale container workloads elastically

• Pros: Portable workloads, built-in orchestration features
• Cons: You still manage the containers and their configuration

• AWS EKS (Elastic Kubernetes Service)
• Google GKE (Google Kubernetes Engine)
• Azure AKS (Azure Kubernetes Service)

• Execute code in response to events without servers. (i.e. events/triggers such as HTTP calls, file uploads or schedules.)
• Build lightweight microservices and background tasks, without provisioning servers
• Execute short-lived tasks with instant fine-grained scaling, whilst paying only for execution time

• Pros: Zero server management; fine-grained scaling (pay only for execution time)
• Cons: Latency on cold starts and time limits on execution

• AWS Lambda
• Google Cloud Functions
• Azure Functions

# Example: AWS Lambda function event source
Events:
  FileUpload:
    Type: S3
    Properties:
      Bucket: my-logs
      Events: s3:ObjectCreated:*

• DBaaS (Database as a Service): Managed SQL or NoSQL databases
• BaaS (Backend as a Service): Prebuilt backend APIs for mobile and web apps
• DaaS (Desktop as a Service): Virtual desktop environments in the cloud
• MLaaS (Machine Learning as a Service): Pretrained models and AI APIs

These niche services fit specific workloads and reduce the need to build and maintain complex systems in-house.

To decide, ask yourself:

1. How much control do I need? IaaS or CaaS for deep customizations.
2. How quickly do I need to ship? PaaS and SaaS let you move faster.
3. What is my budget? FaaS (Serverless) can be cost effective for spiky workloads.

• Security: Apply the principle of least privilege at every layer.
• Monitoring: Use centralized logging and metrics to track cost and performance.
• Avoid lock-in: Plan/Define clear exit strategies, especially for PaaS and SaaS.
• Start small: Prototype in a sandbox before extending into production.

• Try spinning up a free tier IaaS instance today
• Deploy your first serverless function with AWS Lambda
• Explore a PaaS offering and compare deployment times

Dive in, experiment, and let as-a-Service accelerate your Modern TechOps.