“The Network is your computer,” goes one of the slogans.

The techno buzz around the office and online has been about “cloud computing.” So when the email appeared in my box about Sun Microsystems offering a webinar called “Introduction to Cloud Computing…for Enterprise Users,” I signed up—only to see that opportunity get overshadowed by other commitments. Then, they sent a follow-up email offering the archived webinar online. Perfect.

The Next Big Thing or a Fad?

Dr. Lew Tucker presented on cloud computing and suggested that this term integrates a range of factors. There on a digital cloud in blue space were the various terms: “software as a service,” “grid computing,” “application hosting,” “virtualization,” and “database as a service.” While web companies have been transitioning to so-called “cloud computing,” enterprise information technology (IT) units in companies have been much slower on the uptake. They have remained with their traditional data centers. Tucker showed a slide that speculated that cloud computing may be the IT method of choice for various enterprises in the next few decades.

The official definition Tucker used for “cloud computing” was from NIST (or the National Institute of Standards and Technology at http://www.nist.gov/index.html). Their definition, as cited by Tucker, reads: “a model for on-demand network access to a shared pool of configurable computing resources that can be rapidly provisioned and released with minimal management effort or service provider interaction.” The NIST site on cloud computing hedges by asserting that this is an evolving paradigm (http://csrc.nist.gov/groups/SNS/cloud-computing/).

Some Attributes of Cloud Computing

Essentially, on ground, this is about a shift from dedicated in-house technological infrastructures for data management to moving some IT functions online…into the “cloud” (something like distributed computer services). Here, entities may pay per use instead of having to pay 24/7/365 even when resources are not being used. These services may be scalable depending on whether an organization has to ramp their uses up or down. They would be accessing virtualized physical resources, probably through automated operations (with the expensive human intervention removed). The idea of cloud computing isn’t only about applications but permeates all levels of the IT infrastructure, including the physical one…with the virtualization of servers. A number of applications may run off the same server. In addition, typical functions that are usually housed within an organization may be virtualized—such as the storage of data, the functions of servers, and networking.

The “business case” for cloud computing includes lower costs by using the pay-as-you-go model. Developers may rapidly create IT functionalities through “self provisioning” and mashing up multiple functions through API enablements. Lastly, cloud computing may offer flexibility for on-demand, elastic, and scalable services.

Tucker suggests that applications themselves will likely soon have to monitor themselves in terms of load and be able to automatically request more computing resources to reduce loads…but also be able to scale down depending on computing demand, to save moneys. He offered a slide with a number of logos of various companies—some big-name, some apparently fresh—who have come on board with cloud computing.

Cloud Computing Models

The presenter suggested a number of cloud computing models: “software as a service (applications on demand), platform as a service (developer platform for creating applications), and infrastructure as a service (storage and compute capabilities offered as a service)”.

He cited Google Mail and Google Docs as examples of software as a service (SaaS). There is nothing to buy or download or manage. It all happens on the Web. These spaces are multi-tenant ones. He also cited Microsoft Exchange Online, Yahoo Mail, TurboTax Online, Cisco WebEx Weboffice, and Salesforce.com as some other examples of SaaS.

Platform as a Service (PaaS) offer ways for on-demand development and deployment of unique applications. These support unique programming by developers for specific contexts. These spaces serve as both platforms and a channel to consumers. He cites Facebook as a platform as a service; this connects people and offers many tools for them to self-express, share information, receive messages, and communicate. The Google AppEngine, Force.com, Netsuite Business OS, ad Heroku are other examples of Platform as a Service.

He discussed storage of digital files and data and computation as part of an infrastructure delivered as a service or Infrastructure as a Service (IaaS). Consumers may access an on-demand virtual infrastructure that is usually built at the lowest level of an IT infrastructure—with the storage of information and the applications delivery. He cited Amazon AWS, Microsoft Azure, RackSpace Cloud, Savis, Terremark, and Joyent as examples of “Infrastructure as a Service.”

An Evolving Web

He asserted a natural evolution of the Web, progressing from websites to applications to developer platforms to computing and storage on the World Wide Web. While websites and applications are broadly used, the move to developer platforms and computing and storage on the Web have not gained as much traction.

The financial models for such services currently are ad-based, with free services to users—for many of the most popular applications. Others are subscription-based or use-based.

He said that various standardization organizations are working to strive for “cloud APIs” (application programming interfaces) for interoperability between these various cloud systems—for easier mash-ups. He also highlighted the imaging of virtual machines that are complete and pre-configured for various operating system and machine setups, for easy use and download. He showed Amazon’s AWS with its numerous machine “images” for different stacks.

Trust a Core Issue

Tucker said that data governance and application security are critical issues. How well does a company or organization trust a particular service provider to protect information and information access in a “shared, multi-tenant environment.” Can the providers fulfill legal requirements to produce created information for regulatory compliance or in response to a legal request for discovery? Is the cloud computing reliable?

Later in the presentation, the speaker described some ways that cloud computing may be enhanced more in terms of safety—such as by encrypting data before uploading into the cloud and then decrypting after download. He also suggested that virtual machines may be “hardened” for security. He described ways that HIPAA compliance may be achieved even in a cloud computing environment.

Other Functionalities

Cloud computing may also enable large-scale analytics of various datasets from different data streams, to surface “hidden” information. So-called “new analytics” may offer more agile information collection and vetting.

Public and Private Clouds

Organizations may choose to put some of their contents in public clouds, some in private clouds, and some in a hybrid setup. The benefits of a public cloud relate to the wide access to “unlimited resources” and the flexibility of paying as one goes. The on-the-premises private cloud is built on a company’s own datacenter and on its own dedicated and directly owned IT infrastructure and machines. A hybrid mixes both public and private clouds in a strategic way. Even a private cloud may provide efficiencies with virtualization and data center automation. He cited vendors who support the building of private clouds built on a shared pool of virtual resources.

Government Leadership regarding Cloud Computing

He expressed surprise at apps.gov, which is a federal government endeavor at cloud computing efficiencies (https://apps.gov/cloud/advantage/main/start_page.do). This site promises responsive deployment of new applications and major cost savings.

With so much buzz about this concept, it was helpful to have a webinar walk-through. If this does take, this will have an impact on instructional design and how we access software, platforms, and infrastructures. The scenario goes like this: We can get on our Netbooks and with connectivity, we can do whatever we usually do on our robust (ripped) desktops and laptops. We won’t have the thousands of dollars worth of software on our computers, but we’ll be able to build using Web interfaces that connect to various machines elsewhere. That is a pretty massive paradigm shift but one worth considering.