We use cookies and other similar technologies (Cookies) to enhance your experience and to provide you with relevant content and ads. By using our website, you are agreeing to the use of Cookies. You can change your settings at any time. Cookie Policy.
Start a Discussion
Cloud Computing
user-pic

Are We Making the Same Mistakes With Cloud Computing That We Made With SOA?

Vote 0 Votes

Are we repeating the same old mistakes that we made with SOA, or have we learned from our mistakes?

14 Replies

| Add a Reply
  • We have learned a lot from SOA in recent years, especially at it relates to governance and service lifecycle management. But the most important lesson is the way the new approach needs to be sold to the business -- that is, it needs to be offered as a solution for a specific business pain point. SOA proponents are well aware that no one in the business is going to sign on to "SOA," but will sign on to "better streamlining of purchase orders," or "better customer service." Likewise, the business isn't going to adopt "cloud computing" just because it's cloud computing.

  • Most "new" technologies that are strongly hyped are often fraught with some amount of unbridled awe as well as a decent spate of legitimate claims. SOA was never meant to be a specific implementation but rather an architectural framework. But, I can’t tell you the number of people who wanted to "buy a SOA solution". Similarly with Cloud Computing, everyone now wants to "get in the cloud". While the cloud isn’t anything new, the opportunities to "offload computing resources" have exploded – and there is good reason to explore those options if you need a dynamically scalable environment and don’t want to manage all the resources and support yourself.

  • The "cloud" has some potential and does SOA. However, the real enabler is the underlying technology. With SOA it was http (and still is), that was what CORBA never had.

    With the cloud the underlying technology is virtualization and I believe that is very powerful and will change thing. So I propose the question be "Are we making the same mistakes with virtualization that we made with http?" My answer would be no, while http was very strongly hyped, virtualization is rather quiet.

  • SOA made plenty of sense theoretically, but it lacked a compelling business case because it often entailed extended projects, significant costs and uncertain benefits. Cloud computing is not just another overhyped term because it is already enabling users to accelerate their projects, save money and create new revenue opportunities.

  • My view is that we need SOA (or something very similar) to make cloud computing work There isn’t just one Cloud but a number of different sorts: private Clouds and public Clouds, which themselves will divide into general-purpose and specialized ones, and all of those Clouds will be full of pre-defined and readily available services. Therefore, SOA in this regard represents a very different opportunity; it becomes a mechanism whereby a user can put together an “application? based around normal working patterns, using readily available services.
    Could be the saviour of SOA as we imagined it.

  • Yes, as with SOA, suppliers are totally defining cloud computing for marketing advantage so that there is no possible way to measure its success or failure. Is it just a synonym for the SaaS method of delivering software? Is it really just another way of describing the multiple technologies of virtualization (desktop, server, storage, memory, application)?

    Isn't it really just Sun's "the network is the computer" 20 years later? No, actually it is Multics 45 years later (see multicians.org).

    -- Dennis Byron

  • user-pic

    In my view SOA is able to provide most of what it promises but it asks for a lot in return from the business. We the architects/developers/marketers failed to stress on the latter part and that's why I believe SOA has mainly been successful in greenfield projects.
    It's easier for businesses to wrap their head around utility computing. It is fairly obvious to business what they are gaining and what they will be losing (e.g. control over hardware). And in cloud computing it is easier for businesses to have a hybrid solution in place so the transformation plan can be straightforward.

  • We are trying to avoid the mistakes made in SOA standardization by bringing together most of the groups working on Cloud Computing interoperabilty to discuss future activities. See the Agenda for the March 23 Cloud Computing Interoperabilty Workshop in Virginia "http://tinyurl.com/c87pty"

  • There are many successful implementations of SOA. A company would be reluctant in general to advertise its success with SOA because it is their competitive advantage. The business case is compelling but it is difficult to explain it clearly to convince to those who have control over funding.

  • user-pic

    SOA "failed" because it was to damn hard to get going. By the time you built a prototype with a real enterprise process a few months went by and you lost your executive buy in. Why I like Cloud Computing, it's because you can decide to outsource a business process overnight (well, almost!) to a third party with little preperation. Now that's the language executives like!

  • user-pic

    I agree with Samy. You don't see benefits of SOA till your services start getting reused and their is no immediate ROI. On the other hand moving to cloud provides immediate economic benefits.

  • user-pic

    As like any other technology SOA in itself is neither goog nor bad; ultimately the value that business get out of its implementation drives its success measurement.
    Many vendors and companies jumped on SOA bandwagon without first understanding the business domain and problem the technology was supposed to address; SOA implementation without understanding the business process which its suppose to represent just adds another layer of complexity to existing application. SOA is a response to integration and collaboration issues; to make it a success; process and organization governance issues must be considered as well.
    I personally think Cloud computing and its manifestation (SaaS, IaaS, PaaS) has a much better chance to succeed, as vendor adopting it, should have a good understanding and knowledge of the business domain which they are addressing , otherwise market forces will defeat them, unlike internal IT CC vendors compete with each other and that drives the innovation and push cost down, in this economic climate we can expect more vendors adopting CC and specifically SaaS and IaaS as it allows the business to avoid huge upfront IT commitment and adopt a pay as you go model; on the other hand some vendors still have not abandon their "lock in" approach even after entering SaaS market; to see the full benefit we need to see more market entrant and most notably from open community.

  • Cloud computing has avoided the drag of being technology-driven and instead started as a business-driven way to deliver content and functions to a broad audience with relatively low cost – think SaaS. In my experience many grass roots efforts with services and SOA were quickly dragged down by complex SOA “initiatives.?

    We are moving into a new phase of SOA with the appearance of operating systems from traditional OS vendors moving into the cloud and I expect this will open up the market to wide spread adoption of cloud platforms by ISVs and organizations.

  • user-pic

    Much to the chagrin of technology marketeers, no one has yet come up with a generally recognizable TLA (three letter acronym) for cloud computing so that they could take it on the hype roadshow. The reality is any technology innovation, be it a concept, architecture style, or a real product, risks not living up to its promise because of the hyped-up positioning that vendors and other pundits take in order to further their own cause. SOA fell in that trap and so will cloud computing if the IT industry doesn't stop treating every new wave of thinking and solution as a hammer looking for a nail. As has been stated already in several responses, proper articulation in terms of business pain or competitive advantage is the ultimate success factor.

    Having said that, I do agree with the notion that SOA is one of the keys to making cloud computing pay off, the same way as BPM was a killer app for SOA.

Add a Reply

Recently Commented On

Monthly Archives

Blogs

ADVERTISEMENT