Unit
- IaaS definition: Infrastructure as a Service (IaaS) delivers virtualized computing resources over the internet. Users rent virtual machines, storage, and networking capacity on a pay-as-you-go model, eliminating the need for physical hardware ownership.
- Cloud service model layers: IaaS is the lowest of the three primary cloud service models. It provides raw infrastructure building blocks, while PaaS adds managed runtime environments, and SaaS delivers complete applications to end users.
- Core characteristics: On-demand self-service, broad network access, resource pooling, rapid elasticity, and measured service. These five essential traits, defined by NIST, distinguish IaaS from traditional hosting.
- Hypervisor role: The hypervisor software abstracts physical hardware and allocates CPU, memory, and storage to multiple virtual machines. Type 1 hypervisors run directly on hardware, while Type 2 hypervisors run on top of an operating system.
- Resource abstraction: IaaS providers virtualize compute, storage, and networking so that each tenant sees isolated resources. Virtual private cloud environments extend this isolation with customizable network topologies.
- Shared responsibility model: The provider secures the physical infrastructure, hypervisor, and networking fabric. The customer secures operating systems, applications, data, network configurations, and identity management.
- Economic benefits: Capital expenditure shifts to operational expenditure. Organizations avoid upfront hardware costs and can scale resources up or down based on demand, reducing idle capacity waste.
- Common IaaS workloads: Web application hosting, data backup and disaster recovery, high-performance computing, testing and development environments, and big data analytics all benefit from the elasticity of IaaS.
- Provider abstraction: The IaaS interface hides the complexity of underlying physical hardware. Customers interact through a control plane (web console or API) to manage their virtual infrastructure.
- Virtual machine types: IaaS providers offer a wide range of predefined instance families optimized for different workloads such as compute intensive tasks, memory heavy databases, storage optimized applications, or GPU accelerated machine learning. Users choose instance types based on CPU, RAM, storage, and network performance requirements. Many platforms also allow custom machine configurations to fine tune resources.
- Storage options: IaaS platforms deliver multiple storage services including block storage volumes attached to virtual machines for persistent data, object storage for scalable unstructured data like backups and media files, and file storage for shared access across instances. Each storage class has distinct performance tiers, durability guarantees, and pricing models that align with different application demands.
- Virtual networking: IaaS enables creation of isolated virtual private cloud environments with customizable IP address ranges, subnets, route tables, and internet gateways. Users implement network security via access control lists and security groups that act as virtual firewalls at the subnet and instance level. Advanced features include VPN connections, direct peering, and traffic mirroring.
- Auto scaling and load balancing: Auto scaling groups automatically adjust the number of virtual machine instances in response to metrics such as CPU utilization, memory usage, or custom application performance indicators. Load balancers distribute incoming traffic across healthy instances, improving fault tolerance and ensuring high availability. This combination allows applications to handle variable workloads efficiently.
- Disaster recovery and backup: IaaS providers offer snapshot capabilities for virtual machines and storage volumes, enabling point in time backups. Cross region replication allows copying data to geographically distant data centers, supporting robust disaster recovery plans. In the event of a regional outage, users can failover to replicated infrastructure, minimizing downtime and data loss.
- Hybrid and multi cloud strategies: Many enterprises adopt hybrid cloud architectures by connecting on premises data centers to IaaS environments via dedicated direct connections or encrypted VPN tunnels. Multi cloud approaches involve using infrastructure from different providers to avoid vendor lock in and optimize costs. These strategies require careful network design and consistent management tooling.
Code Example
Python
import cloudsdk
# Initialize the cloud compute client
compute = cloudsdk.ComputeClient(region='us-east')
# List all virtual machines in the account
instances = compute.list_instances()
for instance in instances:
print(instance.name, instance.state)This snippet demonstrates listing IaaS virtual machines using a fictional SDK. The compute.list_instances() call interacts with the provider's API to retrieve a list of all VM instances, showing how IaaS exposes resource management through code.
Quick Revision
- IaaS delivers virtualized compute, storage, and networking on a pay-as-you-go model.
- Five NIST characteristics define IaaS: on-demand, broad access, resource pooling, elasticity, measured service.
- The hypervisor enables resource abstraction, running multiple VMs on physical hardware.
- Shared responsibility model places OS and above in the customer's domain.
- Economic benefits include capex-to-opex shift and elastic scaling.
System Architecture
Pattern: IaaS Layered Stack
Components:
- Physical Layer: Data centers with servers, storage arrays, and network switches. This layer is owned and managed completely by the provider.
- Virtualization Layer: Hypervisors (e.g., KVM, VMware ESXi) that abstract physical resources into pools of virtual CPU, memory, storage, and network capacity.
- Control Plane: API endpoints, web console, and CLI tools that allow customers to provision, configure, and monitor their virtual infrastructure.
- Customer Virtual Infrastructure: Virtual machines, virtual disks, virtual networks, subnets, security groups, and load balancers that customers manage.