SOA is not differentiating, Cloud Computing is

By Mark Skilton, Capgemini

Warning: I confess at the start of this blog that I chose a deliberately evocative title to try to get your attention and guess I did if are reading this now. Having written a couple of blogs to date with what I believed were finely honed words on current lessons learnt and futures of technology had created little reaction, so I thought I’d try the more direct approach and head directly towards a pressing matter of architectural and strategic concern.

Service Oriented Architecture (SOA) is now commonplace across all software development lifecycles and has entered the standard language of information technology design. We hear “service oriented” and “service enabled” as standard phrases handed out as common terms of reference. The point is that the processes and practices of SOA are industrial and are not differentiating, as everyone is doing these either from a design standpoint or as a business systems service approach. They enable standardization and abstraction of services in the design and build stages to align with key business and technology strategy goals, and enable technology to be developed or utilized that meets specific technical or business service requirements.

SOA practices are prerequisites to good design practice. SOA is a foundation of Service Management ITIL processes and is to be found in diverse software engineering methods from Business Process Management Systems (BPMS) to rapid Model Driven Architecture design techniques that build compose web-enabled services. SOA is seen as a key method along the journey to industrialization supporting consolidation and rationalization, as well as lean engineering techniques to optimize business and systems landscape. SOA provides good development practice in defining user requirements that provide what the user wants, and in translating these into understanding how best to build agile, decoupled and flexible architectural solutions.

My point is that these methods are now mainstream, and merely putting SOA into your proposal or as a stated capability is no longer going to be a “deal clincher” or a key “business differentiator”. The counterview I hear practitioners in SOA will say is that SOA is not just the standardized service practices but is also how the services can be identified that are differentiating. But that’s the rub. If SOA treats every requirement or design as a service problem, where is the difference?

A possible answer is in how SOA will be used. In the future and today it will be a business differentiator in the way the SOA method is used. But not all SOA methods are equal, and what will be necessary to highlight SOA method differentiation for business benefit?

Enter Cloud Computing, its origins in utility computing and the ubiquitous web services and Internet. The definitions of what is Cloud Computing, much like the early days of Service Orientation, is still evolving in understanding where the boundary and types of services it encompasses. But the big disruptive step change has been the new business model the Cloud Computing mode has introduced.

Cloud Computing has introduced automatic provisioning, self-service, automatic load balancing and scaling of resources in technology. Building on virtualization principles, it has extended into on-demand metering and billing consumption models, large-scale computing resource data centers, and large-scale distributed businesses on the web using the power of the Internet to reach and run new business models. I can hear industry observers say this is just a consequence of the timely convergence of pervasive technology network standards, the rapid falling costs per compute and storage costs and the massive “hockey stick” movement of bandwidth, smart devices and wide-scale adoption of web-based services.

But this is a step change movement from a simple realization that it’s just “another technology phase”.

Put another way: It has brought the back office computing resources and the on-demand Software as a Service Models into a dynamic new business model that changes the way business and IT work. It has “merged” physical and logical services into a new marketplace on-demand model that hitherto was “good practice“ to design as separate consumer and provider services. All that’s changed.

But does SOA fully realize these aspects of a Cloud Computing Architecture? Answer these three simple questions:

• Does the logical service contracts define how multi-tenant environments need to work to support many concurrent services users?
• Does SOA enable automating balancing and scaling to be considered if the initial set of declarative conditions in the service contract don’t “fit” the new operating conditions that need scaling up or down?
• Does SOA recognize the wider marketplace and ecosystem dynamics that may result in evolving consumer/producer patterns that are dynamic and not static, driving new sourcing behaviors and usage patterns that may involve using services through a portal with no contract?

For sure, ecosystem principles are axiomatic in that they will drive standards for containers, protocols and semantics which SOA standards are perfect to adopt as boundary conditions for service contracts in a Service Portfolio. But my illustrations here are to broaden the debate as to how to engage SOA as a differentiator when it meets a “new kid on the block” like Cloud, which is rapidly morphing into new models “as we speak” extending into social networks, mobile services and location aware integration.

My real intention is to raise awareness and interest in the subjects and the activities that The Open Group is engaged in to address such topics. I sincerely hope you can follow these up as further reading and investigation with The Open Group; and of course, do feel free to comment and contact me J

Cloud Computing and SOA are key topics of discussion at The Open Group Conference, London, May 9-13, which is underway. 

Mark Skilton, Director, Capgemini, is the Co-Chair of The Open Group Cloud Computing Work Group. He has been involved in advising clients and developing of strategic portfolio services in Cloud Computing and business transformation. His recent contributions include the publication of Return on Investment models on Cloud Computing widely syndicated that achieved 50,000 hits on CIO.com and in the British Computer Society 2010 Annual Review. His current activities include development of a new Cloud Computing Model standards and best practices on the subject of Cloud Computing impact on Outsourcing and Off-shoring models and contributed to the second edition of the Handbook of Global Outsourcing and Off-shoring published through his involvement with Warwick Business School UK Specialist Masters Degree Program in Information Systems Management.