consultaglobal

“Cloud computing evolved rapidly during 2008, with infrastructure-based computing services and broader cloud application development platforms emerging to fill specific positions in a solidifying market structure. That structure encompasses the provision and management of rapidly scalable, remote, virtual computing resources charged according to usage and the additional tools needed to manage those resources and develop applications to run on them.”

Read Danny Dicks’s  “Interoperability: A Key Challenge for Cloud Computing” on Light Reading. 

As shared in yesterday’s post, Cloud Slam 2009 is a weeklong conference packed with back to back presentations, all of them delivered from “the cloud.” This morning we saw Cisco’s David Bernstein presentation jointly with his video feed. His talk focused on infrastructure challenges and what it takes to simplify cloud building. By the way, Cisco signed the Open Cloud Manifesto and David claimed that addressing interoperability happens to be an unavoidable matter.

Cloud Slam will eventually make presentations and videos available on their website. In the meantime, if you were in need for a quick crash course on the very basics around cloud computing, here is a video discussing the subject “in plain English:”

Business press covering ICT, information and communication technologies, reports that this sector appears to be better prepared to weather the economic downturn than most. Some point to the lessons learned in the dot.com burst which have made ICT vendors be more agile. There have also been numerous references to cloud computing and the growth of mobile broadband.

Focusing on cloud computing, this service is seen as a cost effective way of enabling both consumer and enterprise applications become far more affordable than ever before. This reminds me about he discussion on business model innovation and disruptive technologies which I blogged about this past weekend, where killer applications where defined in terms of affordability and coverage.

Business wise, cloud computing is positioned as a way to do more with less. For instance, in yesterday’s session we heard statements such as “$7 of every $10 of IT spending is related to maintenance.” Moreover, enterprise IT utilization was shown as being “sub 10% and slow to react on average.” Cloud computing services deliver a different cost structure. An alternative way of handling enterprise IT which is expected to be significantly more cost efficient. My recollection is that Mark Werrel mentioned the example of a leading financial enterprise  incurring just $0.56 (cents) application / hour with full IT support.

This brings up the subject of business models. Much of what’s discussed is based on “metering,” a utility model. The metrics thrown around are: CPU usage, network traffic such as GB and TB delivered, storage, requests per second, seats (headcount in terms of users), etc. Generally speaking, what get’s most people’s attention are business arrangements that require no up front commitment and pay as you grow (elastic) contracts. Long story short: scale up or down as needed and on the spot. DC stands for data center in the below chart:

The fact is that one size does not fit all and, it so happens, there are different implementation models:

• Public cloud: access over the Internet available to all users, shared infrastructure.
• Private cloud: behind the corporate firewall for use by people in the enterprise and partners.
• On premise: customer owned and outsourced services.
• Off premise: collocation (multi-client data center) and public clouds.
• Hybrid: enterprises leveraging some combination of the above, the so-called mashup of clouds.

Providers in the cloud computing business are deploying the following service models, where each subsequent  level can imply a higher level of abstraction as follows:

• IaaS, Infrastructure as a Service: the delivery of infrastructure by means of a platform virtualization environment where server, network, storage, resources become available on demand. Examples: Amazon EC2 and S3, Mosso, Terremark, etc. Reliable and granular enough reporting on what’s utilized is critical: it enables billing for these services.
• PaaS, Platform as a Service: adds a layer of services on top of IaaS. Application developers can take advantage of APIs, application programmable interfaces, to easily invoke specific assets or resources. Add to that the availability of forms, drop down menus, drag and drop icons part of visual programming languages designed to speed up the creation of applications (workflow, user interface, security being some examples.) This means the delivery of a computing platform equipped with a solution stack as an on demand service. In some cases, end users themselves can create applications that meet the needs of a given business process without having to do any coding or calling IT for support.

• SaaS, Software as a Service: actual on demand applications for end users such as Expedia and Salesforce, which would otherwise be more costly (2.5% up for on premise applications was the percentage shared by one of the speakers.) Most typically, SaaS  seats on top of IaaS and PaaS with services licensed out directly to end users.

Note that PaaS might integrate an app store. This means that you can make your application become available to end users and to other developers who might like to use it, tweak something to customize it, or to build a more complex one upon it. That is often subject to revenue sharing agreements between the owner of the source application and the PaaS provider.

Last but not least:

• XaaS, Anythying as a Service: needless to say that this is a very dynamic industry, so we are likely to see other models emerge in the next few years. Note that SaaS, the same acronym as above, is claimed by some storage vendors as Storage as a Service instead.

In any case the main benchmark is the cost/benefit ratio:

• Cloud computing providers make a point about faster application deployment and, therefore, the kind of speed to market which is harder to achieve when dealing with one’s systems.
• The “the cloud” becomes immediately available and is scalable on a need basis.

What’s most interesting is the fact that this kind of infrastructure and services become available to small and medium size enterprises. This is another source of innovation as more entrepreneurs can get things done quicker. At the time I’m posting this, Intuit is giving a good presentation on that subject.

J. de Francisco blogging from Chicago on April 21

consultaglobal.org  :::  disclaimer & disclosure

Other posts of interest:

• Cloud Slam 2009 (1); (2)
• eComm 2009 (1); (2); (3)
• 2.0 companies leverage 2.0 technologies
• open cloud manifesto