Skip to content

Beginner · What cloud computing is · topic 1 of 50 · 25 min

What the cloud is, and why companies use it

Explanation

Cloud computing means renting computing power, storage and software over the internet from a provider, and paying only for what you use. You get a server in minutes instead of weeks, and you can grow or shrink with demand. The trade-off: bills can surprise you, you depend on the provider, and outages still happen.

From buying to renting

A server is a powerful computer that runs programs for other computers, such as a website or a database. Before the cloud, a company that wanted a website bought servers, put them in a room with power backup and air conditioning, hired people to run them, and waited weeks for delivery. This is called on-premises (often shortened to "on-prem"): the hardware sits on the company's own premises or in a rented rack.

Cloud computing turns this around. A cloud provider, such as Amazon Web Services (AWS), Microsoft Azure or Google Cloud, owns huge data centres (buildings full of servers). You rent a slice of them over the internet, start it from a web page or a command, and pay by the second, hour or gigabyte. When you stop using it, you stop paying.

Think of it like this: Owning servers is like buying a car: a big payment up front, then fuel, insurance and repairs, even on days it stays parked. The cloud is like booking a cab: you pay per trip, someone else maintains the vehicle, and on a wedding day you can book five cabs at once without owning five cars.

The NIST definition: five essential characteristics

NIST, the US National Institute of Standards and Technology, wrote the definition most textbooks and exams use (Special Publication 800-145). A service is "cloud" when it has these five characteristics:

CharacteristicWhat it meansEveryday example
On-demand self-serviceYou get resources yourself, without phoning or emailing anyoneClicking "Launch instance" and getting a server in a minute
Broad network accessYou reach it over the network from laptops, phones and scriptsManaging your servers from the AWS app on your phone
Resource poolingThe provider shares one big pool of hardware among many customers, kept separate from each otherThousands of companies run on the same data centre, each seeing only its own resources
Rapid elasticityCapacity grows and shrinks quickly, often automaticallyGoing from 2 servers to 20 for a sale, then back to 2
Measured serviceUsage is metered and billed, like electricityA bill that says 730 server-hours and 50 GB of storage

CapEx versus OpEx

CapEx (capital expenditure) is money spent up front on things you own for years, like servers. OpEx (operational expenditure) is money spent regularly on running costs, like rent or a phone bill. On-premises is mostly CapEx: you pay lakhs of rupees before the first customer arrives. The cloud is mostly OpEx: you pay monthly for what you used. A startup with little cash likes OpEx because it can start small and grow with its revenue.

The benefits

  • Agility: a developer can try an idea in an afternoon instead of waiting weeks for hardware.
  • Elasticity: you handle a traffic spike by adding servers for a few hours, then remove them.
  • Global reach: you can run your app in Mumbai, Frankfurt and Virginia without opening offices there.
  • Economies of scale: providers buy hardware and power in enormous volumes, so their cost per server is lower than yours would be.
  • No undifferentiated heavy lifting: you stop spending time on cables, cooling and disk replacements, and spend it on your product.

The honest downsides

  • Cost surprises: a forgotten server, a large database or heavy data downloads can produce a bill far bigger than expected. You need budgets and alerts (module 2).
  • Lock-in: the more provider-specific services you use, the harder it is to move to another provider.
  • Skills: cloud work needs new skills in networking, security and cost control. A badly configured cloud is less safe than a well-run server room.
  • Outages still happen: providers have whole-region problems from time to time. For example, a large AWS problem in the US East (N. Virginia) region in October 2025 disrupted many well-known apps for hours. You design for failure (module 10).
  • Steady workloads are not always cheaper: if you run the same 10 servers flat out for five years, buying can cost less. The cloud shines when demand changes.
Careful: Common mistake: saying "the cloud is always cheaper". It is cheaper for changing, uncertain or small workloads, and for avoiding idle hardware. For a steady, predictable load, compare honestly, as you will do in the task below. Another mistake is thinking "it is in the cloud, so it is backed up and secure". Backups and security settings are still your job (see the shared responsibility topic).
Tip: In interviews, define cloud computing in one line ("on-demand computing resources over the internet with pay-as-you-go pricing"), then name two or three of the NIST characteristics with an example. That answer is short and shows you know the standard definition.

Need help?

Stuck on a word or an idea? Search this course, or every course, and jump straight to the explanation or a worked question.

Try:

Watch the lesson

What the cloud is, and why companies use it · 1/7

What is the cloud?

Welcome to cloud computing. In this lesson you will learn what the cloud really is, and why so many companies use it.

Prefer a video by a teacher? Find free videos on “what is cloud computing for beginners NIST five characteristics” on YouTube ↗

In real life: A sale day at an Indian e-commerce company

An online shop normally needs 2 servers. During its yearly festive sale, in the style of Flipkart's Big Billion Days, traffic jumps about five times for three days, and at the stroke of midnight it can be even higher. On-premises, it would have to buy 10 servers that sit mostly idle for the other 362 days. In the cloud it runs 2 servers all year and adds 8 more for the 72 sale hours, then removes them. The same pattern shows up on a railway booking site at Tatkal hour (10 am for AC classes), when lakhs of people try to book in the same few minutes.

python
base, peak, sale_hours = 2, 10, 72
rate = 10   # rupees per server-hour (made-up number)

always_on = base * 365 * 24 * rate
sale_extra = (peak - base) * sale_hours * rate
print("Rent for the year:", always_on + sale_extra)   # 180960
Rent the base all year, and the extra servers only during the sale.

Practice (0/3 done)

Task 1 · Code · Python

Write rent_cost(servers, hours, hourly_rate) that returns what it costs to rent that many servers for that many hours at hourly_rate rupees per server-hour. Example: rent_cost(2, 8760, 10) returns 175200 (2 servers for a full year of 8,760 hours).

Task 2 · Code · Python

Compare buying with renting for one year. If you BUY, you must buy enough servers for the peak: cost = peak * (server_price + running_cost), where running_cost is power and upkeep per server per year. If you RENT, you run base servers for all 8,760 hours of the year, plus (peak - base) extra servers only during sale_hours, at hourly_rate. Write yearly_costs(base, peak, sale_hours, server_price, running_cost, hourly_rate) that returns a dict {"buy": ..., "rent": ..., "cheaper": "buy" or "rent"}. If they cost the same, say "buy".

Task 3 · Question

A college buys 20 servers for ₹30 lakh this year. Which kind of spending is that?

Quick quiz

Question 1

1.Which of these is one of the NIST five essential characteristics of cloud computing?

Question 2

2.Your bill shows 730 server-hours and 40 GB of storage. Which characteristic does that show?

Question 3

3.Which workload gains the MOST from the cloud?

Question 4

4.Which statement about the cloud is TRUE?

Read more (free)

Open a question to see the answer and why. The first ones are this topic's own quiz, so try the quiz above before you peek.

Which of these is one of the NIST five essential characteristics of cloud computing?

Answer: Rapid elasticity

NIST lists on-demand self-service, broad network access, resource pooling, rapid elasticity and measured service.

Your bill shows 730 server-hours and 40 GB of storage. Which characteristic does that show?

Answer: Measured service

Measured service means usage is metered and billed, like an electricity meter.

Which workload gains the MOST from the cloud?

Answer: A ticket site with a huge spike for 15 minutes each morning

Spiky, changing demand is where elasticity saves the most money.

Which statement about the cloud is TRUE?

Answer: You can start small and pay as you go

Pay as you go is real. Outages happen, security is shared, and steady loads can be cheaper on-premises.

What is the best first step to avoid a surprise cloud bill?Related

Answer: Set a budget alert on day one

A budget alert warns you early. Free tiers and prices change, so check them regularly.

From Deploying to the cloud

You stop (not terminate) an EBS-backed instance. What happens?Related

Answer: The root volume is kept and you still pay for it

Stopping ends the compute charge but the EBS volumes remain, and storage is still billed.

From EC2: virtual servers in the cloud

Who empanels cloud service offerings for Indian government departments?Related

Answer: MeitY

The Ministry of Electronics and Information Technology (MeitY) runs the empanelment.

From Public, private, hybrid and multi-cloud

Which statement about multi-cloud is most honest?Related

Answer: It adds identity, networking, billing and skills overhead, so it needs a clear reason

Multi-cloud can be justified, but every extra cloud adds work and data transfer cost.

From AWS vs Azure vs Google Cloud: the service map

Pass the quiz (60% or more) or mark this topic as done when you are ready.

Your progress is saved on this device only.

Next: IaaS, PaaS, SaaS and serverless →

Next steps

  1. 4 more topics in “What cloud computing is” before its exam: IaaS, PaaS, SaaS and serverless, Public, private, hybrid and multi-cloud, Regions, availability zones and edge locations, The shared responsibility model.
  2. Module 1 exam: What cloud computing is · 13 questions · 20 min, without help. 5 topics of this module still to finish first.
  3. Then the next module, AWS account, IAM and billing. 50 topics left in the course before the final exam.