Infrastructure as a Service (IaaS) is a cloud computing model that delivers virtualized computing resources over the internet — including virtual machines, storage, networks, and operating systems — without purchasing or managing physical hardware. Major providers (AWS, Microsoft Azure, Google Cloud Platform) offer extensive global infrastructure spanning multiple regions and availability zones, billed on a pay-as-you-go basis. The shared responsibility model defines the security boundary: providers secure the underlying physical infrastructure and hypervisor, while customers manage their operating systems, applications, data, and access controls. In 2026, IaaS has expanded well beyond basic VM hosting to encompass AI-optimized GPU compute (Gartner projects AI-optimized IaaS spending at $37.5 billion in 2026), confidential computing, sovereign cloud, and FinOps governance — making infrastructure decisions a strategic business differentiator.
What This Cheat Sheet Covers
This topic spans 21 focused tables and 164 indexed concepts, 159 flashcards. Below is a complete table-by-table outline of this topic, spanning foundational concepts through advanced details.
A jump-to index of every table row in this cheat sheet.
An interactive map of every table and concept in this topic.
Table 1: Core IaaS Concepts
These eight concepts form the foundation every IaaS practitioner must internalize before anything else. Understanding the difference between elasticity and provisioning, or why multi-tenancy is safe despite shared hardware, prevents architectural mistakes that are expensive to undo.
| Concept | Example | Description | |
|---|---|---|---|
aws ec2 run-instances --image-id ami-12345 --instance-type t3.medium | • Software emulation of a physical computer running an OS and applications • the foundational compute resource in IaaS | ||
Type 1: VMware ESXi, Hyper-VType 2: VirtualBox | • Software layer enabling multiple VMs on a single physical server • Type 1 runs directly on hardware; Type 2 runs on a host OS. | ||
$0.0416/hour for t3.medium= $30.37/month continuous | On-demand billing where you pay only for resources consumed — no upfront commitments or long-term contracts. | ||
Scale from 2 to 20 instances during traffic spikeReturn to 2 when load drops | Ability to automatically scale resources up or down based on demand — the key differentiator from traditional infrastructure. | ||
terraform applyaws ec2 run-instances | Automated allocation and configuration of resources, typically completed in minutes vs. weeks for physical hardware. | ||
Multiple customers' VMs on same physical serverIsolated via hypervisor | Architecture where multiple customers share the same physical infrastructure while maintaining logical isolation enforced by the hypervisor. | ||
99.99% uptime = 4.38 min downtime/month99.95% = 21.9 min/month | • Contractual guarantee of service availability • providers issue credits if uptime falls below the committed percentage | ||
aws ec2 describe-instancesaz vm list | All resources manageable via programmatic interfaces, enabling automation, Infrastructure as Code, and CI/CD integration. |