Facilitating the Spread of Knowledge and Innovation in Professional Software Development

Write for InfoQ


Choose your language

InfoQ Homepage Articles 2009 Trends and Directions for WebSphere

2009 Trends and Directions for WebSphere

This item in japanese


Happy 2009! Each year I outline a few technology areas that we are eagerly focusing on in WebSphere-land.
Here is my top 10 list for 2009 (in no real particular order), that I outlined on my blog, (with help of David Letterman).

  1. Business Mash-ups
  2. Business Rules
  3. Middleware-as-a-Service
  4. Rainmaker
  5. Extreme Scale
  7. Restful - Agile
  8. DataPower-lution
  9. POWERful Middleware
  10. Industry-savvy Middleware

While to some this may look like a random assembly of intriguing topics, there are a handful of overarching themes that these technologies areas are playing to. Facilitating the Alignment of Business and IT, by empowering the Business user with a set of rich, web-based business tools – is a theme dominating the SOA front in 2009. Getting the business user to electronically express their thoughts (do I dare say program?) in a repeatable, testable and visual fashion is key to topics like Business Mash-ups, Business Rules, Middleware-as-a-Service (MaaS), and Industry-savvy middleware. Agility is the second key theme that is pervasive across the list. This is especially important given the outlook of a lagging economy in 2009. Built-for-purpose middleware that radically improves time to value are the central thoughts behind our agile software endeavors and can specifically be found in WAS.NEXT, RESTful, DataPower-lution and Industry-savvy middleware. And never to be overlooked, is our perennial focus on one of WebSphere’s core values; Extreme Transaction Processing, which is the motivation behind Extreme Scale and our focus on POWER™.

Let’s look a little closer at the ten areas:

Business Mash-ups - Take the rich web experience of MySpace and Facebook and put a business twist on it and you have the essence of our Business Mashup activity. Business Mashups empower a business user to construct a custom web portal that allows them to dynamically create dashboards where they can monitor and control their essential business activities– like monitoring business processes and responding to actionable business events. Within this rich Internet application, AJAX based authoring tools allow you to model and test business processes and apply business rules to alter the behavior and flow of your processes in real time. Business Mashups utilizes the IBM Mashup platform technology that is present in products like IBM Mashup Center.

Business RulesiLog is now part of the WebSphere team… Woo-hoo! With the help of the iLog team, we will continue our integration of business rules technologies – both procedural and inference based – throughout our WebSphere product lines; including WebSphere Application Server, Business Events, Process Server and in our business authoring tools (Modeler, Monitor). In 2008, we acquired AptSoft, which gave us a collection of tools for visually constructing procedural rules that run on our event server (WBE). In 2009, we will further enhance these tools and look at integrating iLog rules technology into the mix to allow for more sophisticated inferencing to be applied to event processing and process orchestration. We will also focus on managing and governing business rules using means consistent with our SOA management story (e.g., in the WebSphere Registry)

Middleware-as-a-Service – Point your browser at a web site and you have instant middleware at your fingertips. Middleware-as-a-Service (MaaS) is focused on pre-provisioned and hosted middleware, running in a compute cloud (either on-premise or off-premise). We are initially focusing on Connectivity/Integration, Commerce and BPM in the context of a hosted/service model. We aim to take an 80/20 solution for each of these topics – meaning that we are working to provide a simplified, constrained abstraction for each of these functional areas - all driven through a rich web-based interface. Our first goal is this space is to be arms providers to creating on-premise middleware clouds in their enterprise (which is the embodiment of my term Rainmaker – more on that below). In 2009, we will also exploit public clouds, like Amazon, to run WebSphere middleware. MaaS provides advanced virtualization, automation and autonomics to reduce operational expense, “time to value” for on-premise WebSphere deployments and reduces the complexity associated with integration of services which are outside their core IT (e.g. LOB, B2B, and born-on-the-web applications/services).

Rainmaker – I’ve been using the term “Rainmaker” to encompass our activity relating to WebSphere Clouds. You bring the hardware, Rainmaker technology helps you create and manage the WebSphere Cloud. It introduces the notion of a WebSphere virtual image – which is a pre-installed, pre-configured, pre-optimized binary image of WebSphere middleware (e.g., WAS). Rainmaker allows for the cataloging and customization (e.g., adding your application and associated componentry) of WebSphere images (think CVS for virtual images). Rainmaker technology also allows for those images to be assembled into image patterns, which are groups of images that form highly available clusters that conform to 10 years of established best practices. Rainmaker can also dispense images to hyper-visors and activate them. The images and applications contained within, are continually managed, monitored and optimized by Rainmaker.  Watch this space, as we'll be making some very exciting announcements soon.

Extreme Scale – Millions of transactions per second, with predictable latency, and an effective price/performance ratio is what we aim for. Data grid technology provides the means to achieve these feats. Our WebSphere Extreme Scale (WXS) product is the focal point for Extreme Transaction Processing – recreating the attributes of CICS and TPF in a Java-centric/distributed environment. In 2008, we took another step towards infusing extreme scale throughout portfolio, by embedding WXS technology in WebSphere Business Events (WBE/XS). WBE/XS has recorded multi-million events per second in our labs. But we are not stopping here. We are looking to XS-enable our process server, ESB and Registry. In 2009, we are also looking at expanding the programming APIs that sit atop WXS; including REST APIs, simple cache APIs - inspired by the open source memcached, to the sophisticated map reduce and streaming APIs.

WAS.NEXT – At 10 years old and with 7 major versions under her belt, WebSphere Application Server (WAS) is ever moving and evolving. In 2009, the WAS team will be busy working on the next major release, which promises to be a landmark release. The days of one-size-fits-all servers have evolved towards purpose-built servers that specialize in specific types of workloads (e.g., Web). Building on the Equinox-based OSGi core, which has been present in WAS since V6.1, WAS.NEXT moves even closer to a “composable application server” platform. Profiles allow the WAS server to take on personalities; including JEE, pure Web, Messaging and Extreme Scale - while economically using only the minimum system resource to run that profile (e.g., small footprint and ultra-fast startup times.)

Restful / Agile – In 2008 we made solid progress in REST enabling our WebSphere portfolio. I’m quite proud of the team for aggressively responding to the call; including the REST enabling in CICS, WebSphere MQ, WAS, WSRR, Commerce, Portal, Process Server, and the list goes on. We have more work to do in 2009. Specifically, with all this REST work, a Common REST API is needed across these products. Complementing our work on REST is our work on Agile programming – specifically the dynamic scripting capability (PHP and Groovy) that is the hallmark of our WebSphere sMash product. In 09, we will continue to work on our web-based tooling; adding better support for forms and more “off the shelf widgets”. We also continue to focus on our business partners. In fact, I will be delivering the keynote address at SugarCon next month to illustrate how SugarCRM, the worlds most sophisticated PHP business application, runs atop sMash.

DataPower-lution DataPower-lution is about the evolution of our appliance line. Our WebSphere DataPower (DP) appliances were one of the most popular WebSphere products with customers in 2008 – and as a result, we grew the product line, adding two new (B2B and Low Latency Messaging) appliances before 2008 concluded. Given the economic outlook for 2009, our customers will be even more cost and time conscious. Our DP appliances are now proven to deliver on “time to value” and “cost of ownership/operations”; reducing a business integration task that would take weeks to a few hours. Therefore, in 2009 you can expect more appliances. DP will increasingly become a platform for new styles of appliances; similar to the way WAS became a platform for “stacked products” like Portal and Process Server. Given that, look for DP to cross over into the cloud computing and virtualization space via our Rainmaker and Middleware-as-a-Service projects. Also look for DP to focus on "WebSphere Application Firewall" functions by moving application optimizing features to the edge of the network, like fragment caching, middleware self-balancing and application content filtering. Other focus areas include improved appliance management (e.g., updating the firmware of a cluster of appliances) and scaling out our appliances via IBM BladeCenter integration.

POWERful Middleware – In 2009, we will put extra emphasis on integration of our middleware with our IBM hardware, especially our POWER™-based hardware systems. This includes taking advantage of the multi-core capabilities of our POWER6-based server, directly in our middleware; including our JVM and Just-in-time-compiler. At the simplest level, this involves more aggressive work to testing and document best practice configurations of our products on POWER hardware. For example, we can now demonstrate record transaction-processing performance when you combine the latest 64-bit POWER hardware with the latest 64-bit support in our WebSphere platform (leveraging features like 64-bit pointer compression in our JVM). As we aggressively implement our Rainmaker/virtualization Cloud strategy; our POWER-based hypervisors plays a key role in virtualizing our middleware with strong security and isolation of our virtual WebSphere environments.

Industry-savvy Middleware – Across our product line, we are focusing on how customers use our products from specific industries. In 2009, we will continue enabling our middleware to excel at providing immediate “out of the box” value for industries; like banking, insurance and financial markets. For example, in 2008 we acquired Infodyne, which gave us a huge momentum boost in the financial markets business. We quickly worked to combine the Infodyne technology, which includes highly performant market data feed handlers, with WebSphere Low Latency messaging – including it in both our WebSphere Front Office product as well as our new DataPower XM-70. In 2009, we will continue to leverage assets acquired from Webify to build out our WebSphere Business Services Fabric; which enables business users to rapidly assemble new industry aware business processes and define industry savvy vocabularies and tasks to enable the creation of new tasks from existing IT assets.

While these top ten technology focus areas will surely bring a new level of Agility and improved time to value to our WebSphere product line; there are about a dozen (or more) other interesting technology endeavors happening in WebSphere-land that I have not described here. Given this level of investment and innovation, I am certain that 2009 will lay the foundation for another solid 10 years of customer success around WebSphere.

Gennaro "Jerry" Cuomo is an IBM Fellow and one of the founding fathers of IBM WebSphere Software. He is currently the Chief Technology Officer of WebSphere, where his prime charter is to “cultivate the future of WebSphere”.

Rate this Article


Hello stranger!

You need to Register an InfoQ account or or login to post comments. But there's so much more behind being registered.

Get the most out of the InfoQ experience.

Allowed html: a,b,br,blockquote,i,li,pre,u,ul,p

Community comments

  • 2009 Trends and Directions for WebSphere-- technology & salesmanship


    Your message is awaiting moderation. Thank you for participating in the discussion.

    I fail to understand who could benefit from reading this kind of article.
    If this has any reflection on IBM standards, then good luck!

  • Re: 2009 Trends and Directions for WebSphere-- technology & salesmanshi

    by Floyd Marinescu,

    Your message is awaiting moderation. Thank you for participating in the discussion.

    I fail to understand who could benefit from reading this kind of article.
    If this has any reflection on IBM standards, then good luck!
    Hi Javaid, InfoQ covers current events in our space, so a letter by the CTO from IBM about what their plans are for Websphere is not just 'current' but forward looking. InfoQ is the only place where you can read this.

  • Decision management and WebSphere

    by James Taylor,

    Your message is awaiting moderation. Thank you for participating in the discussion.

    Given my track back has not shown up, here's a link to my response

  • It looks nice... on paper

    by Peter Veentjer,

    Your message is awaiting moderation. Thank you for participating in the discussion.

    Websphere is getting bigger and bigger every year, I guess the same goes for the problems. If it is caused by
    - inexperienced developers
    - Websphere just getting too complex for mere mortals to understand
    - nice individual technologies (like coherence) that never work well together
    But Websphere environments are never something I'm looking forward to. Imho: too much baggage without providing value.

  • Re: It looks nice... on paper

    by Colin Renouf,

    Your message is awaiting moderation. Thank you for participating in the discussion.

    The baggage is very much a thing of the past. WAS6.1 started OSGi, WAS7 is mostly built from OSGi components declaring services except where some Open Source package has been "wrapped". Individual components can be started and stopped to have appropriate builds for appropriate uses (i.e. no EJB container, etc). I now, for the first time, think it is well architected. It is also dynamic in that deployment of components causes events that cause appropriate actions and reconfiguration in the runtime. WAS.NEXT allows some of this componentisation to be more "productised".

    One other thing about this announcement - speaking as a customer - is that Jerry and his team have actually spent a great deal of time going out to the customers and architects, or working at a low level with the user groups, to find out what we wanted. A lot of these directions are as a direct result of the IBMers listening to customers, making proposals, getting feedback on the proposals, and then iterating the proposals into products.

    So, I am pleased to see the content of this letter, and look forward to Rob High (Chief Architect and IBM Fellow) going into more detail at the WebSphere User Group conference in the UK next month.....

  • Re: It looks nice... on paper

    by Alex Sorkin,

    Your message is awaiting moderation. Thank you for participating in the discussion.

    Just another soviet like propaganda article.

    Could somebody provide GOOD customer feedback on any WebsFear product?

    Most expensive, most unimproved, most buggy, most unstable, most complicated, most costly products...

    - Black Hawk down... finally...

  • Re: It looks nice... on paper

    by Robert Dean,

    Your message is awaiting moderation. Thank you for participating in the discussion.

    It's amazing that you would even bother to comment on a product with which you clearly have no recent experience.

    IBM has made steady, continuous improvement to WebSphere with each release. In fact, I would have to say that the only thing I don't like is that their new feature pack strategy makes application of maintenance more difficult and time-consuming.

    In the past several months, we've tripled the work that our WebSphere server is doing, and it's not even breaking a sweat.

Allowed html: a,b,br,blockquote,i,li,pre,u,ul,p

Allowed html: a,b,br,blockquote,i,li,pre,u,ul,p