Introduction to cloud computing

Models

Service models

IaaS (Infrastructure as a Service):

  • Raw computing resources (servers, storage, networking)

  • Maximum flexibility and control

  • Examples: AWS EC2, Google Compute Engine, Azure VMs

PaaS (Platform as a Service):

  • Development and deployment platform

  • Managed runtime environment

  • Examples: Heroku, Google App Engine, Azure App Service

SaaS (Software as a Service):

  • Ready-to-use applications

  • Minimal management needed

  • Examples: Gmail, Salesforce, Microsoft 365

FaaS (Function as a Service):

  • Serverless computing

  • Pay only for actual function execution

  • Examples: AWS Lambda, Azure Functions

Deployment models

Public Cloud:

  • Resources owned/operated by cloud providers

  • Shared infrastructure with other customers

  • Examples: AWS, Azure, Google Cloud

  • Benefits: Cost-effective, scalable, low maintenance

  • Challenges: Less control, potential security concerns

Private Cloud:

  • Dedicated to a single organization

  • Can be on-premises or provider-hosted

  • Examples: VMware, OpenStack, Azure Stack

  • Benefits: More control, better security, compliance

  • Challenges: Higher costs, requires expertise

Hybrid Cloud:

  • Combination of public and private clouds

  • Data/apps shared between environments

  • Examples: Azure Arc, AWS Outposts, Google Anthos

  • Benefits: Flexibility, best of both worlds

  • Challenges: Complex management, integration needs

Multi-Cloud:

  • Using multiple public cloud providers

  • Example: Using both AWS and Azure

  • Benefits: Avoid vendor lock-in, optimize costs

  • Challenges: Complex management, different tools

Community Cloud:

  • Shared by organizations with common interests

  • Example: Government clouds

  • Benefits: Shared costs, common compliance

  • Challenges: Limited availability, specific use cases

Components

Infrastructure

A hypervisor is a software layer between the hardware and the VMs that allows multiple VMs to run on the same physical machine.

Bare metal servers are physical servers dedicated to a single tenant, without virtualization or hypervisor layers.

Types of hypervisor:

  • Type 1 (Bare-metal)

    • Runs directly on the hardware

    • More efficient and secure

    • Used in enterprise/data centres

  • Type 2 (Hosted)

    • Runs on top of OS

    • Easier to set up

    • Used for desktop virtualization

Containers vs. virtual machines:

  • Containers share OS kernel

  • Less resource overhead

  • Less isolation

Storage and content delivery networks

Direct Attached Storage (DAS):

  • Physically connected to one server

  • Fast, simple, and low cost, but limited scalability

Block Storage:

  • Raw storage volumes presented as blocks

  • High performance and bootable

File Storage:

  • Hierarchical storage with folders

  • Easy sharing and familiar structure

Object Storage:

  • Flat structure using unique identifiers

  • Highly scalable and cost-effective, but slow

A content delivery network is a globally-distributed network of servers that delivers web content to users from the nearest geographic location, reducing latency and improving performance. It acts as a cache between users and the origin server.

Emergent trends and practices

Message brokers are middleware that enable services to communicate asynchronously by handling message queuing, routing, and delivery.

Last updated