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

When Will the Cloud Start Creating Deep Impact in the Middleware Space and What Segments Will be Affected First?

Vote 0 Votes
A month ago I asked the Forum, What are the biggest questions you have around Cloud Computing?  Miko Matsumura answered: When will the cloud start creating deep impact in the middleware space and what segments will be affected first? Enterprise-to-cloud integration? Web 2.0 Application Infrastructure? Modern Internet B2B? Supply Chain?

6 Replies

| Add a Reply
  • Middleware products should address Cloud Computing integration issues in the near future (up to two years) because some Cloud Computing segments especially SaaS are rapidly going Mainstream (for example read my post: Even SAP is offering SaaS ERP http://avirosenthal.blogspot.com/2010/04/even-sap-is-offering-saas-erp.html. However is is easier to address integration of some segments and it is a lot more difficult to address integration of other segments. My prediction is that like other technologies trends and architectures, addressing simpler integration issues will precede addressing more complex issues. Therefore Web 2.0 segment will be the segment which will be affected first. The most important and most complex segment, namely Enterprise-to-Cloud will be the last segment that will be integrated properly.

  • Now. The cloud already has a number of formidable players in the integration space, but the very nature of integration has changed completely with services being delivered from all corners of the globe and in a rapidly standardizing manner. Rather than routing transactions and performing cumbersome translation between disparate application silos, the move towards RESTful and SOAP based web service API's in the cloud has removed a large amount of the service and system integration friction that was once middleware's sweet spot. Almost all of the integration work that my team does is handled by API's, and even the orchestration part of the equation once driven by BPM tools is being supplanted by workflow and business rules native to the platform that we build on (Force.com). And for cloud to on-premise integration, we look to cloud-based solutions to provide standardized services to perform the translation from SOAP to legacy API or database integration points. Whereas we once thought of Websphere and Weblogic as the dominant solutions for getting systems to talk to each other, the new players are companies like Boomi and Denodo in the cloud-based integration space, and even API management services such as Mashery that facilitate the standardization and management of high volume web services API's.

  • Since I threw this question into the ring I thought I would specify what I meant by "deep impact"

    If you look at the impact SalesForce.com has had on CRM, they are significantly denting the sales of on premise CRM and changing the way organizations consume applications as a whole.

    I am looking for similar impact in the middleware space, and I would see this in particular impacting certain special cases of "middle". Obviously middleware grew up between the database and the user, so there's a bit of a limited view of it from that history.

    However, more recently, middleware has emerged as a platform in its own right, and I predict we will see a big emergent player in these areas:

    * Cloud to on-premise integration
    * Cloud to Cloud preintegration (integrating best of breed SaaS)
    * Cloud B2B
    * Cloud Supply Chain
    * Cloud BPM across value chain partners

    These areas seem particularly fruitful as they do not sit in the "middle" of the database and application layers, rather this new form of middleware sits between siloed organizations, between divisions, between companies and across partners.

    As such the Cloud becomes a very natural place for the evolution of such software and a new platform will emerge as a function of this.

    My 2 cents,
    Miko

  • I believe cloud deployment will have a deep and immediate impact on the ability of small to medium businesses to quickly deploy and use Web 2.0 - or Enterprise 2.0 - technology for external as well as internal collaboration and communication "at the edge" (see John Hagel, John Sealy Brown and Lang Davison's new book "The Power of Pull").

    Cloud deployment is particularly helpful for E2.0 use cases that connect external customers, clients, suppliers and partners with internal employees. Use cases include product support (particularly for complex, highly customized products), new product development, supply chain collaboration and sales / marketing communication with reseller and sales partners in private spaces as well as publicly facing marketing and communication.

    Cloud deployment of modern E2.0 technology can minimize IT support overhead while providing secure TLS/SSL encrypted access for private spaces and open Web access to other spaces on the same server.

  • This is an excellent question Miko, because to date I think there has been a generalization of Cloud as a deployment alternative for production servers, etc, when at first there are few enterprise-level, middleware-enabled business processes that can realistically survive in a Cloud without having "wires hanging out" - constraints of over-utilized systems that cannot be placed in the Cloud.

    The pre-integration Cloud you mention - well, the Dev & Test Cloud is going to be a significant area of growth over the next 5 years. It is hard to develop and deliver apps for the Cloud, when there are so many dependencies not available in the Cloud - mainframes, 3rd party apps like SaaS, systems of record and other over-utilized and pay-per-use assets that Development and Test teams simply can't budget for, since their use is highly variable. We will need to find a way to take our middleware processes, and other external constraints, and make them available for integration purposes in the Cloud.

    I've made it one of my little missions to better describe this phenomenon with a series of YouTube "chalk talks" there's a series of 4 videos - http://bit.ly/9Bc2J3

    There's a reason why current successes in the Cloud are kind of consumer-oriented, content-heavy apps, and not as much "big iron" multi-enterprise apps and middleware. Pre-Production clouds will be a necessity for Cloud Apps.

    The fears that Cloud Apps are not bulletproof enough for serious middleware and integration use are unfounded - good software, wherever it is deployed, is still up to people: good architecture, elegant development, responsible security approaches and early testing. If we give them the means to do their jobs better in Cloud pre-production environments the success will follow.

  • Cloud infrastructure for applications does need to have middleware to go along. To that extent, messaging oriented middleware and service bus are already seeing movement. Business Process Management (BPM) oriented middleware on the other hand is considerably complex and does not gel well from a fitment / benefit perspective in cloud. To a large extent, the impact to this segment will be dictated by application infrastructure like server/container and how this gets defined in cloud. Middleware will have to complement these pieces and that will define the order of impact on segments.

Add a Reply

Recently Commented On

Monthly Archives

Blogs

ADVERTISEMENT