Part I:  A Little Background

Saas – Software as a Service
Paas – Platform as a Service
Iaas – Infrastructure as a Service

If you are already familiar with these three acronyms, skip to Part II…

Software as a Service is the most common form of interaction with the cloud.  As previously mentioned, Salesforce.com, a pioneer in this category, has an API that enables you to write programs that extend the features of its service.  In other words, you could write a program that automatically pulls your account leads from Salesforce.com, scans the status of each account and conducts a relevant news search (using other APIs) for idle accounts.  Reports can be generated based on the aggregated data.  The news search service API could be google,  reuters, blogs, rss feeds etc… This type of API-enabled service is widely available across many websites.

Once you have a program that accesses services using an API, then you may have a problem because you want the program to run all of the time.  You can’t run it on your machine, because you don’t want your machine to be running the program all day.  You can’t run it on a web server because it will impact the performance of your website (same with the app server etc…).  You could however write the program and run it in the cloud on a virtualized platform.  This type of the cloud is known as Platform as a Service.  You have essentially chosen a provider that runs the platform you develop on and are comfortable with.  You therefore lease some computing space to execute your program.

Now that you have a report being generated by a program being run in the cloud, what about the distribution of that report?  If you want to make the report available to the very edge of the network across the world as quickly as technically possible, then you could leverage a provider that distributes and caches the report file on servers throughout the world.  This is an example of leveraging Infrastructure as a Service.  Services like Amazon Cloud Front maintain distributed servers for just that purpose.

Part II:  Opportunity

Another way to view the different cloud concepts is to recognize that they are mainly about how systems interact with other systems.  Why is this important to me?

Opportunity

• You should know your options
• Demand for real-time data is growing
• Tech-savvy departments can have more freedom to roam in the cloud, without IT constraints
• Diversification- resell your own data processing services in the cloud.
• Make your services more widely available
• Integration- once an ecosystem of services exists, complete with service adapters between standardized formats, integration efforts can be reduced.

Potential Cost Savings

• Pro-rated licensing charges on a per use basis
• Scale on demand- provision new servers as needed
• Consolidation of resources to manage
• Start-ups may use the cloud to keep costs down and to focus on core competencies, rather than buying equipment upfront
• Larger IT departments with a data center can get more breadth of range using the cloud without hiring a system admin for a new/different platform

Source: Gary Larson’s The Far Side

There are many factors to consider when looking at the cloud.  Greg Shipley, CTO of Neohapsis, a risk management firm, reminds me that choosing a provider is not simple from a risk management perspective. “The potential for cascading failures increases as cloud providers construct technologies and services on top of other cloud providers.”  In other words, if you use a service that culls demographic information for consumers in China, but you don’t realize that the service relies on Facebook APIs, then you could be inconvenienced if China decides to block Facebook.

Not every problem has a solution in the cloud.  Even if there is a viable solution for your business, there may be a fundamental hesitation.  “As the importance of a service to a business grows, there is the perception that the business is at the mercy of their provider”, Griffin Caprio – Founder & President, 1530 Technologies, Inc. What if the service goes down?  Perhaps the provider is sold to another company and the new owner starts increasing your costs.  Maybe your service provider stops supporting the service you are using for lack of customers.  Whatever the situation, you should be prepared.

So how do you avoid these pitfalls?  Planning and asking the right questions.  These are all manageable issues that are can be properly addressed when you consider each step in our Nine Steps to Cloud Nine.

Part III:  Nine Steps to Cloud Nine

The following steps should help you avoid costly mistakes as you inspect opportunities to leverage or build services in the cloud:

1.  Goals defined
2.  Roadmap
3.  IT involvement / governance / SLA
4.  Choose platform/provider

• Security (user admin access, network, storage, encryption)
• Integration (seamlessly integrated to your systems)
• Portability (migrate a virtual instance between private VPN)
• Vendor lock-in / flexibility
• Marketplace longevity/stability

5.  Lightweight prototype of a core feature
6.  Test thoroughly
7.  Measure/extrapolate against goals (Until success – repeat step 4-7)
8.  Architecture defined
9.  Build the application

The next step is the fun part- determining which strategic goals may have a solution in the cloud.  Look for future posts that discuss some example strategies in more detail.