Part 1: Core Concepts (2 Marks)
- Cloud Computing (NIST SP 800-145 Definition): A model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources (networks, servers, storage, applications) that can be rapidly provisioned and released with minimal management effort or service provider interaction.
- NIST 5 Essential Characteristics:
- On-demand self-service
- Broad network access
- Resource pooling (Multi-tenancy)
- Rapid elasticity
- Measured service (Pay-as-you-go)
- Deployment Models: Public Cloud (AWS, Azure), Private Cloud (On-premise OpenStack), Community Cloud, Hybrid Cloud.
Part 2: Structural Frameworks (10 Marks)
Cloud Service Delivery Models (SPI Model)
Software as a Service (SaaS): Gmail, Office 365, Salesforce
- User manages: ONLY user application data & configurations
- v
- Platform as a Service (PaaS): Heroku, AWS Elastic Beanstalk
- User manages: Applications and Code Logic
- v
- Infrastructure as a Service (IaaS): AWS EC2, Azure VMs, S3
- User manages: OS, Middleware, Runtime, Apps, Networking
- v
- Cloud Provider manages: Physical Servers, Hypervisor, SAN
Part 3: Distinction Tables
| Basis of Comparison | IaaS | PaaS | SaaS |
|---|---|---|---|
| Control Level | Highest; full control over OS and network stack. | Medium; control over application code and dependencies. | Lowest; user interacts only with pre-built application. |
| Vendor Responsibility | Physical hardware and hypervisor only. | Hardware, OS, runtime, and server patching. | Entire software lifecycle end-to-end. |
| Target User | Systems engineers, DevOps architects. | Software application developers. | End users, enterprise staff. |
Part 4: Worked Problems & Exam Traps
Evaluating Total Cost of Ownership (TCO) in Cloud Migration
CapEx vs OpEx:
- Traditional On-Premise: High Capital Expenditure (CapEx - buying physical server racks, SAN storage, backup generators).
- Cloud Computing: Zero CapEx; pure Operating Expenditure (OpEx - monthly utility billing matching actual server usage).
Multi-tenancy means physical hardware is shared with other customers. Improper hypervisor isolation can theoretically expose virtual machines to side-channel cache attacks (e.g. Spectre/Meltdown).