Cloud Computing

Slide 1 / 14

Define Cloud Computing

Cloud computing is a specific delivery model for computing services—not a synonym for the internet. This deck covers: one-sentence definition, 5 essential characteristics, delivery model vs 'using a browser', and practical scenarios plus common pitfalls.

Cloud computing

On-demand delivery of computing services over a network using provider infrastructure. Services include servers, storage, databases, networking, and software. Delivered over a network (often the internet). Provider runs infrastructure; you consume services. Pay-as-you-go is common, but not defining.

Provisioning (create and make ready)

Provisioning plus scaling without owning hardware is a strong cloud signal. Provision means create and make ready to use. Create via portal, Application Programming Interface (API), or automation. Scale without procurement or hardware ownership. Browser access does not equal delivery model.

5 essential characteristics (National Institute of Standards and Technology (NIST))

If a service fits these five traits, it matches the core cloud model: on-demand self-service, broad network access, resource pooling, rapid elasticity, and measured service.

On-demand self-service

You can provision what you need, when you need it—without provider staff. Provision without provider staff interaction. Create resources in minutes. Interfaces include portal, Command-Line Interface (CLI), and Application Programming Interface (API). Self-service does not equal auto-scaling.

Broad network access

Services are reachable over the network from many client types. Available over the network. Works from laptops, phones, thin clients. Standard access mechanisms. Access method does not equal cloud definition.

Resource pooling (multi-tenant)

Shared provider capacity with logical separation between customers. Shared pool of compute, storage, networking. Multi-tenant: many customers, one provider pool. Logical separation between customers. Pooling does not mean 'everyone shares the same data'.

Rapid elasticity

Capacity can scale out and scale in quickly to match demand. Scale out for spikes, scale in when demand drops. Fast changes in capacity (short lead times). Can be automatic. Elasticity does not equal self-service.

Measured service

Usage is metered for transparency, monitoring, and reporting. Metered usage (monitor, control, report). Enables consumption-based billing. Helps you see what you used. Measured does not mean 'cheap' or 'fixed price'.

Cloud is about speed and flexibility

Agility is the core shift; cost can be a benefit. Provision in minutes, not weeks. Faster experimentation. Adjust capacity as needs change. Don't define cloud by cost alone.

Shared responsibility + IaaS/PaaS/SaaS

Responsibility is shared, and the split changes by service model. Shared responsibility: provider vs customer. Infrastructure as a Service (IaaS): customer manages more. Platform as a Service (PaaS): provider manages more platform. Software as a Service (SaaS): customer manages least (config/users/data governance).

Practical: browser vs 'need a VM today'

Use the characteristics to interpret what a scenario is really saying. Browser use is not a cloud definition. Look for provisioning plus scaling without owning infrastructure. 'Need a Virtual Machine (VM) today' indicates on-demand self-service. Expect metering indicates measured service.

Practical recap: spikes + IaaS

Elasticity handles changing demand; IaaS doesn't remove customer responsibilities. Traffic spike then normal indicates rapid elasticity. Scale out/in without long lead times. Infrastructure as a Service (IaaS) does not mean 'provider manages everything'. Customer still manages apps, access, and data.

Common pitfalls + practice prompts

Most confusion comes from mixing up the internet, characteristics, and service models. Internet use does not equal cloud definition. Cloud can be public, private, or hybrid. Self-service does not equal rapid elasticity. Resource pooling does not mean no isolation. Measured service does not mean cheap or fixed. Practice: define, rewrite, and map scenarios to characteristics.