Cloud is a technology buzzword with many meanings. Dropbox is cloud. Microsoft Office 365 is cloud. Salesforce is cloud. Any software that you access via the Web is technically cloud. ‘Cloud’ is so prominent in today’s technology industry lexicon, that it is hard to remember a time without it.

The standardization of ‘cloud’ is very recent. Consider that a 2004 review of the Beta release of Amazon Web Services never mentions the word ‘cloud’. And ‘cloud’ does not show any prominence in Amazon’s blogs until 2006!

So by the time the National Institute of Standards and Technology (NIST) published a definition of Cloud in 2011, the NIST researchers had had at least seven years to observe the rise and evolution of cloud computing to guide their definition:

“Cloud Computing is a model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction.”

The essential characteristics of cloud

There are five key traits that cloud infrastructure has:

  1. On-demand self-service – the self in self-service can be an individual person or an individual corporation
  2. Broad network access
  3. Resource pooling
  4. Rapid elasticity
  5. Measured service

Cloud services can generally be broken down into three service models: Infrastructure as a Service (IaaS), Platform as a Service (PaaS) and Software as a Service (SaaS). Each of these has a similar component stack. The difference is which parts of the stack are managed by the provider and which are handled by the consumer.

Cloud Computing Service Model Responsibility Key

By understanding these traits, you’ll be able to understand the differences between modern cloud infrastructure and traditional solutions. in general, traditional solutions have the consumer manage the entire stack in the leftmost column. Cloud is primarily identified by its ability to share resources and scale up and down according to demand.

Different types of cloud

The Deployment Model for Cloud is delineated less along the lines of ‘where’ the Cloud is deployed and more along the lines of ‘with whom’ the Cloud is shared. There are a variety of models that can be used to deploy cloud services in any given business.

  • Private Cloud: for use by one entity/organization
  • Public Cloud: open use by general public
  • Community Cloud: for a use by an affiliation of organizations. The most famous Community Cloud is the AWS GovCloud (US) for U.S. Government-related workloads
  • Hybrid Cloud: for NIST, this means a common cloud software across multiple cloud providers so workloads can port and burst between providers
  • Multi-Cloud – Using two or more clouds to gain either advantage from a unique offering or to protect from Regional or Global cloud provider failure

The paradox of choice

Clouds are not only hardware. Clouds have multiple attributes and can be categorized by each of the following attributes:

  1. The cloud’s Physical Hosting Location whether on the customer’s location (On-Premises) or a vendor’s location (Off-Premises).
  2. The Procedures and Human Operators operating the cloud whether customer-controlled or vendor-controlled.
  3. The Tenancy of the Cloud, whether for one tenant or multiple tenants.
  4. Hardware and software that comprise the cloud’s services.

The following matrix gives an idea of what kind of cloud is created by different combinations of attributes.

The interworkings of cloud location, tenancy and operator responsibility.

Cloud is NOT the resurrection of mainframe

You might be familiar with the term ‘mainframe.’ Up until recently, the mainframe was the primary way for businesses to process large amounts of data reliably. As the data processing speed increased, and storage costs decreased, the technology shifted to be more nimble. . Today we have an abundance of choice, which can lead to a high anxiety of making the wrong choice.

Rackspace’s article promoting the launch of the Rackspace Cloud captures the key difference between Cloud and Mainframe:

“As this landscape has evolved, some have dismissed cloud computing as just a return to the mainframe era. This view is fundamentally wrong. Mainframes were available to only the smartest employees at the richest companies. The cloud is accessible to all, and usable by anyone, at low cost. Its ubiquity is the source of its power.”

The Cloud is accessible to “two guys in a garage” where mainframes were the bastion of big banks, Wall Street Firms and the world’s largest airlines. Cloud democratizes computing and puts power into the hands of anyone who is willing to challenge industry incumbents for market share.

Extra credit: hyperscale vs private cloud

Captured your interest? Good, read on! Bored to tears? Fair enough, stop here! The concepts in the final section should help a next-level understanding of how cloud companies structure and sell their services.

The largest Public Cloud vendors are known as Hyperscale Public Cloud. You’re likely familiar with the industry leaders in this space: Amazon Web Services, Microsoft Azure, and Google Cloud.

When Public Cloud providers started offering private network spaces, the Public Cloud became more accessible to the enterprise. This allowed a private layer on top of the Public Cloud, also known as the virtual private cloud.

It’s virtual because the full capacity of the device is potentially shared across customers. The opposite of virtual is dedicated, where the device is only used by a single user.

With the virtual private cloud, enterprises can assign their own range of private IP addresses, create their own subnets and connect back to their physical data centers with VPN. Even though ‘private’ is contained in the name of Virtual Private Cloud, it is still considered Public Cloud.

A fully Private Cloud can be deployed on on-premise cloud hardware. This allows customers to have dedicated cloud hardware in their data centers. Enterprises can piece together network, storage and compute hardware for their private cloud, or they can work with a vendor for an integrated software and hardware suite installed on- or off-premise.

Private cloud provides compute, network and storage in an on-demand fashion and is almost always virtualized. Private cloud can be managed by corporate staff or be managed by a vendor.
The major differentiator between public cloud and private cloud is that private cloud is single tenant whereas public cloud has multiple tenants.

The cloud is everywhere! Hopefully this article helped you understand it a little bit more than before.