What is the Cloud and how is it Architected?

Part 1 -Welcome to the Cloud! Series

This is the first in a series of blog entries about Cloud Computing.  Over the next several months, RSM will deliver a series of posts about the opportunities that the Cloud presents to business organizations like your own.

Let’s start with a question that will solicit ten different answers if asked to ten different technology providers – What is the Cloud? 

Keeping it simple, Cloud Computing provides the means to procure technology capabilities as a utility.  Equating some of the characteristics of Cloud Computing to those you would expect from purchasing a utility like electricity:

• Cloud Computing is typically paid for as a form of metered service. Like electricity, the more of it you use, the more you pay. Typical Cloud metering metrics might be as simple as how many users you have accessing a solution or as complex as how much bandwidth or compute processing power you have used over a given time period.
  
• Cloud Computing is typically provided in a highly elastic manner – meaning that it can ramp up to handle peak loads when needed.  Similar to electricity, you shouldn’t have to think about it when you require more or less at any given point in time.
  
• Access to Cloud Computing capabilities is fairly ubiquitous.  If you are on the grid, you should have electricity, and the grid is pretty much everywhere.  If you are on the Internet, you should have access to Cloud resources, and the Internet is pretty much everywhere we work now as well whether through broadband, cellular or other means.
  
• There is an aspect of management and administration which is generally provided as part of Cloud Computing solutions.  When you use the Cloud, you don’t have to worry about how it gets to you or how its reliability and availability is assured any more than you do for electricity.

You’ll hear all sorts of terminology and classification related to the Cloud.  Most of this can be boiled down along two lines.  First, what is being delivered?  Technology is generally delivered in a stack with three layers: 

• There is the infrastructural hardware, networking and facilities layer that forms the foundation.  When delivered via the Cloud this is termed Infrastructure-as-a-Service (IaaS).  The typical IaaS Cloud provider possesses, maintains and assures the availability of high volumes of computing and storage capabilities that can respond to peak demands.
  
• There is the platform layer on top of that which provides operating system level capabilities to allow software to effectively utilize the infrastructure.  When delivered via the Cloud this is termed Platform-as-a-Service (PaaS).  Much of the complication around Cloud offerings exists in this layer as the delivery platform must make the multi-tenant and highly elastic aspects of the Cloud unapparent to end users.  The purchase of PaaS capabilities typically includes foundational IaaS capabilities as well.
  
• Finally there is the software layer where the solutions we interface with reside.  In Cloud parlance, this is delivered as Software-as-a-Service (SaaS).  The typical SaaS offering includes the underlying PaaS and IaaS capabilities required to deliver the solution.  Many of us are utilizing  SaaS offerings like Gmail, Microsoft Office 365 and others without even thinking about it.

How is the Cloud solution architected? 

There are three high level architecture variations for Cloud solutions including:

• Many common SaaS offerings are delivered using a public cloud architecture.  A public cloud architecture is set up such that all of the computing and storage resources are shared in a multi-tenant, co-mingled environment.  Individual users do not possess dedicated resources.  As such, public cloud architectures typically allow for little customization and are deployed only in the provider’s environment.  Public cloud architecture examples would include Gmail, Salesforce.com, NetSuite, Microsoft Office 365 and other offerings.
  
• Private cloud architectures provide a mix of shared and dedicated resources.  A classic example would be the delivery of a corporate accounting system or e-mail system in a private cloud.  Application servers and database servers are setup for individual business use only.  Supporting infrastructure and networking capabilities are still shared and provided dynamically.  This allows an individual business to keep its data and applications separate from others which allows for customization and segregation without sacrificing the management and elasticity of the back end infrastructure resources.
  
• Finally, there are hybrid cloud architectures that combine aspects of public and private cloud setups.  Specific resources may be dedicated.  Other resources may be shared.  Some parts of the environment may be within the corporate data center while others come from outside providers.  Hybrid cloud environments are common because every application or infrastructural need within an organization may have different business or technological requirements.

So how does one go about starting to consider how they can best utilize the Cloud?  We suggest a high-level Cloud Rapid Assessment that helps you consider what opportunities and risks the Cloud presents to your business towards the development of a roadmap to the Cloud.

Looking for more details?  Stay tuned for upcoming posts discussing specific business applications and gotchas for the Cloud.  More information can also be found at RSM’s web site. Please revisit this blog in the next few weeks for the next Cloud post in our series which will discuss the virtues of Public Cloud offerings versus Private Cloud offerings. For more information on cloud computing please contact RSM’s technology consulting professionals at 800.274.3978 or email us.

Lee Voigt