Cloud Concepts, Service Models and Deployment Models · Lesson 1

What Is Cloud Computing? Essential Characteristics and Practical Value

8 min read

Lesson objectives

Explain cloud computing through its essential characteristics rather than reducing it to a remote or internet-connected server.
Distinguish the five essential characteristics and explain how each changes resource consumption.
Evaluate a simple situation to decide whether cloud elasticity and measured use create practical value or whether requirements point elsewhere.

Start with the behavior you need from a resource, not the location of the server

Cloud computing is not simply “a server in someone else’s building.” In the NIST definition, it is a model for convenient, broad, on-demand network access to a shared pool of configurable computing resources such as networks, servers, storage, applications and services. Those resources can be provisioned and released rapidly with limited management effort or provider interaction. The important point is the behavior of the service: how capacity is requested, changed and measured, not merely where the hardware sits. A remotely hosted server can therefore lack important cloud characteristics, while a cloud service may hide the server layer completely from the consumer.

Characteristic	Practical meaning	Quick check
On-demand self-service	A consumer can provision computing capability when needed through the offered mechanism without a human provider action for every request.	Can I request or change the resource through the service mechanism?
Broad network access	Capabilities are available over the network through standard mechanisms suitable for relevant client platforms.	Is network access an inherent part of the service design?
Resource pooling	The provider pools resources for multiple consumers and dynamically assigns or reassigns them according to demand.	Are resources managed as a pool rather than as one permanently dedicated machine?
Rapid elasticity	Capabilities can expand or contract rapidly and may appear extensive compared with traditional procurement.	Can capacity grow and shrink when demand changes?
Measured service	Cloud systems meter resource use and provide suitable transparency about consumption.	Can consumption be observed and associated with the resource being used?

Why the five characteristics belong together

The five characteristics prevent us from defining cloud by one signal. Internet access alone is insufficient: a website can run on a fixed server that changes only after a manual request. Virtualization alone is also insufficient: an organization might divide an on-premises server into virtual machines while keeping capacity fixed and provisioning manual. A monthly bill does not by itself prove measured service in the cloud sense. When on-demand service, network access, resource pooling, elasticity and measurement work together, the consumption model changes. The consumer can request capacity, alter it and observe usage rather than beginning every change with hardware purchase and installation.

Practical value is not a promise that cloud is “always cheaper” or “always more secure.” Elasticity can reduce the need to own peak capacity permanently when demand fluctuates, but a poorly designed environment can still cost more. Measured service can improve consumption visibility without making every cost predictable. Fast access to resources can accelerate experimentation, but it can also accelerate uncontrolled resource creation when governance is weak. NIST guidance for evaluating cloud services therefore treats opportunities, risks, performance, reliability and security requirements as matters to assess before adoption rather than assuming one model fits every workload.

Situation	What we observe	Disciplined conclusion
Seasonal demand rises tenfold	Capacity must grow and later shrink quickly	Elasticity may be valuable if the selected service and architecture actually support it.
Very stable workload with strict latency needs	Demand is predictable but performance requirements are demanding	Compare options; do not assume cloud or non-cloud is automatically superior.
Teams create resources without controls	Self-service exists but governance is weak	Speed is useful, but identity, policy, budget and guardrails are still needed.
Remote fixed server on an annual contract	Hardware is remote but provisioning and expansion are slow and manual	Remote hosting alone is not sufficient evidence of cloud characteristics.

When assessing an offering, write four observations before focusing on the provider name: How do I request a resource? How do I reach it? How can capacity change? How is consumption measured? Then add a fifth question about constraints: What performance, reliability, security and data requirements could make some options unsuitable? These questions do not select a product for you. They turn the word “cloud” from a label into verifiable behavior. In the next lessons, service models, deployment models and shared responsibility will add another layer: not only whether a service behaves like cloud, but also what capability is delivered and who manages each layer.

Exercise

Exercise 1 — Classify the evidence. A service hosts an application on a remote virtual server. Internet access is available, but additional capacity requires emailing the provider, capacity stays fixed for the contract term, and no detailed usage data is provided. List the cloud characteristics that the description proves and those it does not prove. Do not add unstated assumptions.

Check your answer

Guided solution: the description proves network access, and it may indicate virtualization, but virtualization is not itself one of the five essential characteristics. The information is insufficient to prove on-demand self-service, rapid elasticity, measured service or resource pooling. The disciplined conclusion is not “definitely not cloud”; it is “the description does not demonstrate the full NIST model, so we need more evidence about provisioning, pooling, elasticity and measurement.” This separates evidence from inference.

Exercise

Exercise 2 — Make a preliminary decision statement. Write four sentences for a fictional service with stable demand most of the year and a large two-week peak. State the demand pattern, one potential value of elasticity, one risk or constraint to evaluate, and one measure you would observe after deployment to judge whether the decision worked.

Check your answer

Model answer: “Demand is normally stable but rises sharply during registration. We want capacity to increase during the peak and fall afterward rather than keep maximum capacity all year. Before choosing, we will examine performance, security, the cost of scaling and what data leaves our current environment. After deployment, we will compare peak resource use and peak-period cost with an appropriate baseline.” The answer is strong because it links a characteristic to context, names a constraint and a measure, and avoids declaring cloud superior before comparison.

Summary: cloud is a resource-consumption and management model, not merely a remote location for hardware. The five essential characteristics give you a practical test: on-demand self-service, broad network access, resource pooling, rapid elasticity and measured service. Value appears when those characteristics address a real need. A sound decision still weighs opportunity against performance, reliability, security, cost and other constraints instead of treating the word “cloud” as the decision itself.

Lesson summary

Cloud is defined by service behavior and essential characteristics, not simply by server location. Use the five characteristics to test a description, then evaluate value and constraints in context.