Andrew Ferrier

Chris Tomkins and I both work on the WebSphere ESB team, and have been blogging about it and related IBM SOA products for some months. We’ve now decided to join forces and launch a new blog called SOA Tips ‘n’ Tricks. This will contain technical tips on ESB and other products as well as wider issues – we don’t know exactly how it will evolve so please feel free to leave us some feedback on the things you read – what you like, and what you don’t. I plan to discontinue writing about ESB or SOA specifically here on my personal blog, although I’ll continue to discuss wider IBM issues.

As Adrian has pointed out, WebSphere ESB 6.0.2 has just been announced. This will be available around the end of the year (together with corresponding new versions of WebSphere Process Server and Websphere Integration Developer). There are a whole host of new features which increase ESB’s capability, as well as other improvements. See Adrian’s post for more information.

WebSphere MQ link allows you to connect WebSphere Application Server (or any WAS-based product, such as WebSphere ESB) to a WebSphere MQ server. From the perspective of MQ, WAS/ESB’s messaging engine appears to be just another MQ server (and, accordingly, you connect them together with sender & receiver channels). From the perspective of WAS/ESB, MQ appears to be a foreign bus. Thus, ‘foreign destinations’ (WAS/ESB) and ‘remote queues’ (MQ) can be used as appropriate to exchange messages across the link. Both point-to-point and pub/sub style messaging can be used, and the MQ link maps corresponding message features as closely as possible.

This is particularly useful for ESB, as it allows it to mediate messages from and to an MQ queue using JMS bindings. You can find more information on how to set up an MQ link in the Infocenter.

I am now on the official list of IBM external bloggers. You can find plenty of my colleagues there also, blogging on a variety of IBM-related and other topics.

If you’re using WebSphere ESB, Fixpack 3 for version 6.0.1 has just been released, which fixes a variety of problems. You can find out more here.

One of the troubles of working in the IT industry is that no-one ever agrees on what the next big thing means. This is partly because software terms tend to run out from between your fingers when you try to grab them, just like soft putty – want to try defining ‘blogging‘? One example I’m thinking about a lot at the moment is SOA (Service-oriented architecture). The problem isn’t just that people don’t know what SOA means, but that there isn’t a single definition – every software vendor has a subtly different message.

ZDNet Asia recently discussed SOA. I noticed that they mentioned that SOA was built on Web Services. This is often true, but the IBM product I work on, WebSphere ESB, allows one to build an SOA entirely without Web Services if desired – for example, you could use it to build one based on JMS messaging. A wide and generous definition of SOA is probably the only one that makes any sense. Some of IBM’s recent SOA products are built on SCA, which allows for a clearly-defined component model for SOA, but this doesn’t have to be used either: you could build an SOA with Visual Basic, a printer, and a minimum-wage employee to enter data, although your response time might suffer somewhat. The only requirement is that the architecture has to be oriented around services (hence the name).

Some might argue that this model is so simple as to be useless, and that might be true if organisations stuck to the widest definition, but if you’re able to adopt a particular type of technology (for example, going the WebSphere/SCA route), there is the possibility to begin to reap the benefits of reuse, adaptability, etc.

So be generous with the definition, but specific in the implementation.

Business Week recently pointed out, in an article linked from Sandy Carter’s SOA blog, that IBM now makes more profit from software than services, which had been our mainstay for a number of years now. I’m not 100% clear on the details (it’s not often I get to look at the big book o’ accounts), but it’s good to know that we in IBM Software Group are doing good for the company’s bottom line.

This is significant as IBM gets ready for our big SOA announcement on 3rd October. Whether you believe the hype or are cynical about SOA being the next big thing, you can find out more details on the IBM website here and register for the webcast.

Joe McKendrick discusses SOA and reuse in a recent blog entry, essentially drawing on some comments from David Chappell that reuse didn’t do as well as predicted in the era of object-orientation, and that SOA isn’t faring well in this department either. Dave Linthicum, in his latest podcast, also discusses this topic.

I’m not sure I can comment that widely on the state of current SOA projects, and I would agree that SOA may suffer from similar management problems to that of object-orientation: if developers of SOA systems aren’t rewarded for saving time with a reuse strategy, they won’t be enthused to do so. This is an important part of any software project, and encouraging reuse is a best practice that shouldn’t be restricted to object-orientation or SOA.

However, whilst I agree that SOA has other benefits apart from encouraging reuse, I have a fairly high opinion of its potential in that respect. It’s important to understand what we mean by reuse. Reuse rarely means using an object or service as it is. There is often a mismatch between the interface offered by the service (object) being consumed, and the service (object) that needs to call this interface. Expecting anything else is unrealistic (even if future reuse plans are made). This is often solved using something like a façade pattern in object-oriented languages, and some form of mediation with services (such as that offered by WebSphere ESB). The latter is often easier, because there is a lower degree of coupling than inside a single programming language, and because programming code is not often needed, and this is why I believe SOA reuse is simpler – if done well. Of course, some work is still required, but this greater ease of reuse makes it a realistic strategy for more scenarios.

I would agree, however, that, as is often the case, the project management problems here are the greatest ones.

I’ve just received my IBM special edition of Service Oriented Architecture for Dummies. It’s shorter than the regular book, but is free – so you can use this link and order one yourself. It only covers the basics of SOA, and doesn’t dive into depth on ESBs, SCA, or any particular product, but despite being so general, it’s nevertheless worth a read if you’re looking for a good introduction to what SOA is all about.

Of course IT fads come and go, and many people have been saying that SOA is another one of these. In fact, many of these fads do last – they just lose the buzzword feel (e-business being a striking example). SOA certainly has some of the hype associated with a fad, but the book does highlight one of the key points that could play in its favour and allow it to persist – that keeping old IT infrastructure is A Good Thing, and that SOA is iterative and about reuse. This is a more realistic business IT model than throwing everything away and starting again, and thus it may see more practical success than some models that have been consigned to the dustbin because they demanded too much up-front.

Of course SOA still has some way to go, particularly with some of the -ies: scalability, security, usability, reliability, and so on. This special edition doesn’t really delve into those topics. But it’s a good start. I’d encourage anyone interested in SOA to read this book – it concludes with an promising example of how Delaware Electric are using WebSphere ESB to build an SOA.

Update 2006-09-22: I’ve been informed that we’ve run out of copies of the IBM special edition of this book – sorry! You can still pre-order a copy of the full book on Amazon by using the first link above, however.

Andre Tost has written a short, but very helpful, article that clarifies a few of the more confusing issues surrounding Web Services. None of the concepts are particularly new, but Andre provides a good summary of some potential pitfalls. It’s well worth reading if you’re interested in SOA or ESBs in general, as well as Web Services specifically.

As an aside, I found this article as part of the WebSphere ESB support RSS feed. For those that aren’t aware, there are support RSS feeds for several IBM software products, which provide new information on documentation, fixes, and other issues. Find the product you’re interested in on IBM’s website, and click on the ‘Support’ link. Then look for the orange RSS logo for the feed.