People often ask me, “What is cloud computing?” My response is, “What isn’t it?” As a sign of the times, a quick Google search reveals over 28,000,000 search results returned for the term “cloud computing.” Facebook returns 170 separate pages dedicated to cloud computing, while YouTube returns 5,320 videos covering the topic.
Obviously, this is deep, deep water we are diving into. Evaluating and researching this topic quickly becomes a tsunami of information in which you can get lost. So let’s raise our heads above the water line, take the conversation back up a notch, and discuss it at a higher level. Let’s explore the uses and examples of cloud computing and not get lost in the tsunami waves of consulting lingo and technology-speak on this immensely deep topic.
Much as the earth’s oceans can simply be described as “a major body of saline water covering 71 percent of the earth’s surface,” cloud computing can be highly summarized as “a general term for anything that involves delivering hosted services over the Internet.” Just as the tip of the iceberg sticking out of the water does not begin to reveal the iceberg’s true size and mass, describing cloud computing in these terms does not expose the size of the subject matter, as it increases the deeper you go below the surface.
Generally speaking, there are three major oceans on the planet: the Atlantic, Pacific and the Indian. (Yes, I know you can subdivide and add in the Southern Ocean and the Arctic, but they are often consumed by either the Pacific or the Atlantic in broader terms and discussions.) For cloud computing, I will take a similar approach and break it into three major categories: Infrastructure as a Service, Platform as a Service, and Software as a Service. Since we’re talking technology here, we are obligated by technology law and Dilbert comic strips to add in their associated acronyms—IaaS, PaaS, and SaaS, respectively. We will touch on each one to give you a solid foundation on cloud computing. But first, let’s figure out where these clouds came from, and the two distinct manners in which they are formed in the first place.
Where Do Clouds Come From, Anyway?
Back in the day, when technology geeks (myself included) were creating diagrams to show visual representations of technical elements (servers, switches, routers) and connection points (buildings, cities, offices), the common representation for “the network” was to draw a “cloud” symbol. This cloud symbol represented more often than not the telephone network provided by AT&T or Verizon to connect the offices, buildings and technology to each other via T1 (phone) lines or other cables. The cloud often included the “magic” (such as the software and services) that happened via the major carriers (AT&T, Verizon, etc.) to deliver data and voice from one office to another. The “magic” and the “cloud” have stuck ever since. After all, it’s always an easy out for the geek in all of us to just shrug off the deep technical questions and say, “Hey, it just happens in the cloud.”
Two Distinct Cloud Formations
There are two basic cloud formations that occur: public and private.
- Public Clouds. Public clouds are formed to sell/provide services to you on the Internet. Major public cloud providers include Amazon’s Elastic Compute Cloud (EC2) model, Google’s App Engine and Microsoft’s Azure Services. Your company may buy services from these public cloud providers to provide a “technical playground” for developers to create custom applications for your business, without needing to invest upfront capital dollars in hardware and infrastructure. Public clouds usually sell to the marketplace. Often, public clouds offer free services, but typically that is with the belief that the end user will ultimately move up the value chain and purchase services from the supplier to generate revenue from these cloud services.
- Private Clouds. Private clouds are formed predominately by corporations or trading groups. For example, a large company like Caterpillar may form a private cloud in which application developers across the globe can access computing servers and/or storage space to build, test and deploy customized software. Or within the private cloud, voice-over-IP telephone calls may be routed across the globe to eliminate long distance and international toll charges. Private clouds are formed for private use and management. They help different offices and locations share the corporate resources (computing servers, disk space and technical infrastructure). These resources are shared on demand without human intervention. Private clouds are built to supply cloud services to a defined population, like a company.
From these two cloud formations, three distinct computing models emerge in defining cloud computing: Infrastructure as a Service, Platform as a Service and Software as a Service.
- Infrastructure as a Service (IaaS). The IaaS computing model is often referred to as “utility computing.” It enables an entity, such as a corporation, to only pay for the computing infrastructure that it consumes. Much like the water or electrical bills you pay for in your own home, in an IaaS cloud, only what is consumed is purchased. The “infrastructure” typically involves computing servers that are virtualized and accessed by the buyer thru an “API” (Application Programming Interface) layer. The API layer allows the buyer to perform actions in the same manner as if that infrastructure was sitting in the room next to them. APIs allow the buyer to access virtual servers (create/delete/change), start and stop processes, allocate storage and ensure unique IP addresses. IaaS is truly a virtual cloud where we technology geeks can spin up or shut down servers while the buyer only pays for what is used. Amazon is a leader in this space with its EC2 platform (aws.amazon.com/ec2).
- Platform as a Service (PaaS). The PaaS computing model combines hardware and software together in one package that is accessed by the buyer. Typically, the buyer will ultimately be an application developer who accesses a suite of tools (servers and software) to create an application or a unique business tool. The best examples of this in the marketplace today are Google’s App Engine (code.google.com/appengine) and Force.com (from Salesforce.com). These environments provide the suite of tools that developers use in this cloud environment to create and develop products. The PaaS cloud allows this development over the Internet utilizing an API layer, web portals, or at times, locally-installed “cloud gateways.”
- Software as a Service (Saas). The SaaS computing model combines software and hardware together as well. However, instead of developing code on the platform like PaaS, the end user simply interacts with the software via the web and via a portal. The SaaS market space is very wide. It can be simply an email host provider (such as Google Mail, hosted Microsoft Exchange or Yahoo) or a very specific tool set such as Salesforce.com (a CRM system) or TurboTax online. There are a couple key points to keep in mind when discussing the SaaS model: data storage and accessibility. In SaaS, the end user interacts with the software tool and the data is stored in the SaaS cloud. SaaS services can be accessed anywhere, as both the tools and the data reside in the cloud.
Are all these combinations of different cloud environments just consultant-speak and new marketing spins to try to get my money? Far from it. Cloud computing really is a new model of consumption. With cloud computing technologies, businesses achieve the following:
- Save capital dollars in onsite technology infrastructure.
- Consumption model—only pay for technology that is consumed.
- Leverage technology resources and skill sets from one business location to another.
- Save project dollars—reduce licensing, implementation and consulting costs substantially by taking advantage of SaaS environments.
- Business velocity—achieve significant agility in responding to business needs (i.e., the ability to add a new server for the sales website in less than an hour versus a week or more).
- Generate true ROI on technology investments. (For example, on one of my projects, the ROI was achieved once nine new servers were implemented; we stopped tracking the ROI after we hit more than 70 servers on that investment!)
My intention has been to break down the information tsunami of “cloud computing” and make it clear and simple by discussing the two types of cloud formations and the three major categories. I hope this article has broken down cloud computing from the proverbial analogy of “drinking from a fire hose” to more of a “tall cool glass of water” scenario for you.
Is the water cool—and does it taste good? Or do you think I’m all wet? Let me know either way at twitter.com/GregJohnsonCTO. iBi
Greg Johnson is chief technology officer of Pearl Companies.