YouTube. Netflix. Uber. Spotify. TikTok. You name it. You sign up and get your own account. Once you set it up however you want, you can access it from any internet-enabled device, including smartphones and smartwatches.
In case your device breaks, is lost, or you switch to a new one, you can still access the account, as well as the settings and information it contained, from another device without having to recreate everything from scratch. It’s like your account’s information was stored safely on a hard drive somewhere other than your physical device.
Without the cloud, none of these things would exist. Heck, you wouldn’t have Google Drive or Dropbox to store those unfinished novels or songs you hope to release someday. And that’s just one example.
But, let’s not get ahead of ourselves here. We promise plain English, and we’ll start from the beginning.
If you’d rather listen to what the cloud is and how it started, “Cloud Atlas: How The Cloud Reshaped Human Life”, is a podcast telling the story of the cloud, from prehistoric times (i.e., the 1990s) to the present day. Find the full podcast series here.
Table Of Contents
- What Is The Cloud In Simple Terms?
- Why Is It Called “The Cloud”?
- How Did The Cloud Start?
- Why Use The Cloud?
- Cloud Delivery Models: What Are The Four Cloud Computing Services?
- Cloud Deployments: What Are The Four Types Of Cloud Computing?
- What Are The Top Cloud Service Providers?
- How To Understand, Control, And Optimize Costs In The Cloud
What Is The Cloud In Simple Terms?
The Cloud is a worldwide network of servers, plus support infrastructure, that runs on the internet, not locally on a device.
Servers are powerful physical or virtual computers at the heart of a network that are designed to “serve” specific functions, such as hosting application software, running a database, or delivering online storage to multiple users at the same time.
All you have to do is connect your device to the internet and sign in to the specific service/application you want. You log in via a mobile app or web browser like Chrome.
Why Is It Called “The Cloud”?
The cloud is basically a combination of three computing aspects that multiple users share at the same time; online storage, database, and processing power (compute). Picture this:
Credit: How servers in the cloud work – CloudFlare
The cloud is not a physical thing, but the servers in it are, and are housed in data centers all over the world. And, unlike personal computers, these data centers are owned and managed by cloud computing specialists called Cloud Service Providers (CSPs).
CSPs design, build, and maintain the technology, services, and infrastructure their data centers require to serve their customers’ application requirements.
Renting this infrastructure for your cloud computing needs is a smart way to save money and time, whether you’re an individual or a company. We’ll share why in a few moments.
So, where did this thing start?
How Did The Cloud Start?
The technology that supports cloud computing has its roots in the 1960s, when IBM virtualized operating systems, which enabled users to share the same resource simultaneously.
However, the cloud as we know it today was born in the early 2000s when engineers at Amazon, then a burgeoning e-commerce service, sought to build a highly scalable platform for its rapidly expanding needs.
To do that, they had to engineer new ways of extending the growing online service without crashing it. The problem: to build new pieces of functionality required engineers to start coding them from scratch every time.
(The audio clips below are from the podcast, “Cloud Atlas: How The Cloud Reshaped Human Life”. Find the full podcast here.)
Software engineers needed to spend a lot of money buying equipment, a lot of time putting it together, and a lot of skills and hands on deck to prepare a data center on their premises or closeby for their upcoming project.
Michael Skok faced this very problem when he was CEO of AlphaBlocks, a software company he founded in 1996.
Now, imagine if engineers could simply rent ready-made capabilities whenever they need them instead of constructing everything themselves.
Wouldn’t it be great if you could get the online storage, database, and computing power (compute) you need without having to build the infrastructure from scratch every time? You could save a lot of time and money. Think about it:
Amazon’s CEO, Andy Jassy, said that this is what his team came up with internally.
Amazon also needed speed and scalability, so it created the Amazon Web Services (AWS) platform, starting with Simple Storage Service (Amazon S3).
AWS now offers more than 200 cloud services, including Amazon S3 (storage), Amazon RDS (database), and Amazon EC2 (compute). Since then, other companies have emerged as worthy competitors, such as Microsoft Azure, Alibaba Cloud, and the Google Cloud Platform (GCP).
Why Use The Cloud?
Companies use the cloud for various benefits.
- Low capital investment – Instead of purchasing, installing, and maintaining physical servers in physical data centers, your company can access technology services, including compute (CPU, RAM, and networking), storage, and databases, whenever you need them from a cloud provider like Amazon Web Services (AWS).
- Cloud computing – You can perform computation work from anywhere at any time, on any device connected to the internet. Switching devices and locations will not affect your ability to carry on working exactly where you left off.
- Pay-as-you-go pricing – With this model, you only pay for the resources you use (compute, storage, and database). There are no upfront payments or long-term commitments required.
- Faster time to market – By renting cloud computing infrastructure on-demand, you can launch and run a project more quickly, reducing the time between idea and finished product.
- Agility for business transitions – The traditional process for pivoting or switching a business that no longer works would involve selling old equipment (at a highly depreciated price), buying new infrastructure components tailored to the new line of business, and basically starting over in the IT department. In the cloud, you don’t have to put up with long, punitive yard sales to get anything new that you need.
- Real-time collaborations – Cloud technologies enables you to work on a project with a colleague or entire team at the same time, over the internet, even if you’re hundreds of miles apart.
- Remote working – If you have internet access, you can access your work files and databases anytime, anywhere.
- Service persistent and high availability – Cloud servers are always online from different parts of the world, and you can tap them whenever you need to.
- Backup and restore services – Your data is copied to multiple data centers around the globe to ensure it is not lost in case of a disaster in one region.
- Near-infinite scaling – The cloud provides enormous computing capacity at your fingertips. You are also free to increase or decrease the resources you use at any time to meet your application needs. To boost performance, scale up/increase usage; to save money, scale down/decrease usage.
- Managed services – A cloud provider builds, operates, and updates the infrastructure, saving you the time and energy to focus on improving your specific product.
Next, let’s explore how you can use the cloud as an organization/company/business.
Cloud Delivery Models: What Are The Four Cloud Computing Services?
There are now four ways to deliver cloud computing infrastructure; Infrastructure-as-a-Service (IaaS), Platform-as-a-service (PaaS), Software-as-a-Service (SaaS), and Functions-as-a-Service (FaaS or Serverless).
Here’s a short and sweet rundown of each.
With this model, you rent servers, networking, and storage capacity from a cloud provider on a pay-as-you-go model, such as from AWS, Azure, or Alibaba. You can increase or decrease these resources as you require to meet your changing computing needs.
The concept of IaaS is similar to leasing a plot of land on which you can build whatever you want. You just need to come with your own building materials and equipment.
Here, the cloud provider offers everything over the internet – servers, networks, unlimited storage, the operating system software, middleware, and databases – at their data center. Rather than paying for hosting your applications on a platform, you pick and pay for the specific things you need to build, test, deploy, maintain, update, and scale the application.
PaaS is similar to renting the tools and equipment necessary to build a house, rather than renting the house itself.
You access a SaaS application through the Internet instead of installing it on your devices. You pay a monthly or annual subscription to use it, and the provider builds, secures, and updates the app.
SaaS applications are also referred to as cloud applications or cloud-based software.
SaaS is by far the most common cloud delivery method for its convenience and straightforward service model. SaaS is like renting a house: the landlord builds and maintains the house, but you can use it as if you owned it once you’ve paid rent.
With this newest model, you can leave server management to the cloud provider, including provisioning, scaling, patching, and scheduling (hence the term “serverless”).
As with all cloud computing models, serverless applications still run on servers. You just aren’t managing them yourself. Also, serverless apps do not run on dedicated machines.
Think of Serverless as taking up an Airbnb as and when you need a room, apartment, or entire condo. The majority of housekeeping is handled by someone else. You pay based on the length of time you use the place.
A subset of Serverless computing, this allows you to segment cloud applications into even smaller components that run only as needed.
It’s like renting a house bit by bit: You only pay for the dining room to cover dinner time, the bedroom for the duration you sleep, and the living room for the precise time you are in there. If you are not using it, you won’t pay rent on it.
Moreover, serverless functions scale up as the number of users increases. Think of that as your rented dining space and table expanding on-demand when a few more guests arrive for dinner — you only have to pay for the exact amount of space each of you takes up. As soon as a guest leaves, the space shrinks automatically and you stop paying for it.
Cloud Deployments: What Are The Four Types Of Cloud Computing?
Away from cloud models, there are four ways to deploy your applications on the cloud:
In a public cloud setup, the cloud service provider provides you with a range of different resources for building your cloud applications and supporting infrastructure across multiple data centers.
It’s called a public cloud because multiple organizations share these resources (multi tenancy) at the same time to achieve economies of scale (cost savings).
In a private cloud, a server, data center, or distributed network is fully dedicated to a single organization. No sharing.
Often, companies use a private network to protect confidential data, including business secrets, and to back up data.
An hybrid cloud deployment combines one or more public clouds with one or more private clouds. You may even link the servers in your local data center (on-premises deployment) .
For instance, you may use their private cloud as a backup for workloads you run on a public cloud.
A Multi-cloud deployment involve using computing resources from two or more public clouds simultaneously, as a single environment.
These resources could be from two or more cloud service providers, and can include SaaS, PaaS, IaaS, and Serverless. Imagine leasing several adjacent plots of land from different landlords and using them fluidly as one plot. While each lot handles a specific business arm or function, they all belong and work together for the same company at the same time.
As you can imagine, managing this setup can be a challenge. In exchange, though, you’ll avoid vendor lock-in, will have a greater choice of services, and will have access to more innovation from different vendors.
Multi-cloud deployments can also be hybrid cloud, and vice versa.
Next, who do you go to for cloud services?
What Are The Top Cloud Service Providers?
The number of cloud providers have grown exponentially over the last couple of years, from large providers for enterprise-grade organizations to more specialized providers for startups and specific industries.
Credit: Cloud providers market share Q4, 2022 (without SaaS market share) by Synergy Research
Amazon Web Services (AWS)
Amazon Web Services dominates the entire cloud services market with 34% of the market share. It provides over 200 services for a variety of purposes, primarily Amazon Elastic Compute Cloud (EC2) for computing, Amazon Relational Database Service (RDS) for databases, and Amazon Simple Storage Service (S3) for cloud storage.
Even though AWS is primarily a public cloud, it also supports usage on-premises, at the edge, in a private cloud, in a hybrid cloud, and in a multi-cloud environment.
With 23% of the market share, Azure is Microsoft’s answer to Amazon Web Services. Besides private and on-premises deployments, it also supports hybrid/multi-cloud resources and edge computing. See our AWS vs Azure pricing guide for some comparisons. The Azure platform is open to everyone, but it is especially useful for organizations that use Windows applications often.
Google Cloud Platform (GCP)
Google’s cloud platform offers a wide range of features. Known for its ease of use and big data/machine learning capabilities, it is the third-largest provider. Moreover, Google pioneered Kubernetes, the open-source, highly scalable, and self-healing container management platform that it now offers through its Google Kubernetes Engine (GKE).
Alibaba Cloud is the largest cloud services provider in the Asia-Pacific region, and fourth globally. It also provides services like DDOS protection, big data processing, and more. Alibaba Cloud’s competitive advantage is its Content Delivery Network (CDN).
It offers the same cloud services as others here, although it markets to enterprises. Oracle claims to offer 2X better compute price-performance than AWS. The Oracle Cloud Infrastructure is optimized to run Oracle-related workloads just like Azure is for Microsoft workloads.
IBM Cloud, DigitalOcean Cloud, Huawei Cloud, Dell Technologies, and Tencent Cloud are other prominent cloud providers.
What Next: How To Understand, Control, And Optimize Cloud Costs
The cloud is great for business. But there’s one major problem. Cloud resources are so easy to access and use that it is easy to overconsume them without even realizing it.
According to annual surveys, including our own report here, the average organization loses about 30% of its cloud budget to wastage, which can range from idle compute resources to resources that do not stop running (and billing you) when no longer needed.
The majority of cloud cost management tools only display the total or average costs you incur. An advanced cloud cost intelligence platform llike CloudZero reveals exactly how much you are spending on each customer, product, feature, environment, project, team, and more to inform your next step.
CloudZero is capable of this for multi-cloud environments (AWS, Azure, and GCP), across platforms (Kubernetes, Snowflake, Databricks, New Relic, MongoDB, etc), and even for multi-tenant environments.
And, get this, without requiring perfect cost allocation tags.
But, reading about CloudZero is one thing. Seeing CloudZero in action is another. Get a free tour of CloudZero here.