Sunday, October 01, 2006

SOA the new legacy

SOA is probably the most hyped topic at this point in time. Its probably right up there with utility computing and virtualisation. I suppose this means identity management has taken a bit of a step down until the next terrorist attack.

The move to service orientated archtecture seems a popular adoption of most critical system business still running on their stable legacy systems. The reason for this mostly being that its in a modular in stucture and easier to manage. Financial institutions with their most arcane technologies would find this highly desirable as it is then possible to stage out the development stages and control cost and scope with the proper management.

Its funny how the currect hype topics seem to link so closely with one another and seem to compliment each other rather than contradict each other as much. My belief is that the reason why its so out there is to hide the limitations of the inherant problems with SOA. Scalability and performance. N recently mentioned performance issues of java and how incredibily memory hungry it was. My prompt reply was, its still dependant on the designer and its hardware scalable.

Its true that java isn't the most efficient language on the block, however the limitations can be minimised by throwing more hardware at it with the gamble that the development costs will still be cheaper than sleek C code.

So there's it. Another corporate conspiry similar to selling sugar separately from lemonade.

No comments:

dead pi

Well, I guess it has to happen at some point. the home automation raspberry pi has died. Much to do with the stupid Strontium mini SD card. ...