Did you know there were 882 SOA-related patents last year — way more than were filed at the top of SOA’s hype cycle?
Neither did I. But it turns out, that’s a huge jump from 2006-2009, when SOA seemed to be at the top of its hype cycle and on the lips of every vendor and trade journalist (guilty).
Philip Stander recently revisited SOA, pointing out it’s being reborn in the wave of cloud computing and SaaS. He also looked at Google search hits for service-oriented architecture as compared to other mainstream ICTs (information and communication technologies) and found a sharp increase in sites publishing SOA-related material since 2008.
His conclusion: Service-oriented architecture is more relevant now than ever.
“So, looking beyond hype and ‘opinion pieces’, significant research and development has been extended during the past decade with early adopters probably embarking in Research and Development projects early in the first decade of the millennium,” he writes. “This empirical analysis provides a more reliable measure of the perceived value of SOA by the broad ICT innovation community.”
SOA has even settled in at academia, with schools promoting and publishing papers about its value. Stander points out a search for “SOA University course” brought back 800,000 results, and that the University of Queensland and Carnegie Mellon University both offer courses in SOA.
There are a number of reasons for SOA’s success, but earlier failures may have been because too many organizations tried to push top-down SOA, Stander writes. He and his colleagues at GlobeTOM, a South Africa-based systems integrator and product development company, contend that SOA must evolve from the bottom-up:
SOA is about gradually transforming your present IT assets into a Service Repository asset. With this approach, your assets gradually become your service repository and its underlying business processes that may leverage other finer-grained or atomic business services. It is our view that the disillusionment that culminated in statements that SOA is dead originated from large enterprises that wrongfully believed SOA to be the silver bullet that will magically rectify highly fragmented and broken IT infrastructures with a major top-down approach.
SOA adoption is one of the areas GlobeTOM specializes in — along with, oddly enough, golf booking technology. That’s not to downplay what he’s saying about SOA, much of which I agree with, but it’s worth noting.
Of course, he also notes other reasons for SOA’s newfound success, including its ability to deliver on agility, using business process management (BPM) as a foundation for SOA, and SOA’s essential role in cloud computing and SaaS.
You might also want to explore the potential of using SOA with an API strategy. The New York Times shared its own success with this approach several years ago, but as more companies adopt an API strategy, it’s worth revisiting the topic now.
Stander reminds us that SOA still has value, and has actually delivered on much of its promise:
- Reuse and agility, resulting in a quicker time-to-market for new services and solutions
- Consolidation of silos and better business alignment for IT
- Reducing costs while increasing the competitive advantage for the business
Now that we’re years into SOA and it’s matured, it’s worth revisiting — even if you believe SOA failed your organization — to see what’s worked and what hasn’t, because, as the evidence shows, service-oriented architecture isn’t going away anytime soon.