10 Must-Know Topics For Software Architects In 2009

In the last year or so, after quite a lull, the software architecture business has gotten rather exciting again. We're finally seeing major new topics emerging into the early mainstream that are potential game-changers, while at the same time a few innovations that have been hovering in the margins of the industry are starting to break out in a big way.

The big changes: The hegemony of traditional3 and 4-tier application models, heavyweight run-time platforms, and classical service-oriented architecture that has dominated for about a decade is now literally being torn asunder by a raft of new approaches for designing and architecting applications.

These might sound like incautious words but major changes are in the air and architects are reaching out for new solutions as they encounter novel new challenges in the field. As a consequence, these new advances either address increasingly well-understood shortcomings of existing approaches or add new capabilities that we haven't generally focused on before but are becoming increasingly important. A few examples of the latter include creating reusable platforms out of applications from the outset (theopen API story) or cost-effectively creating architectures that can instantly support global distribution, hundreds of terabytes of data, and tens of millions of users. There are others that we'll explore throughout this post.

These innovations are hallmarks particularly of the largest systems being built today (which are running into unique challenges due to scale, performance, or feature set) though these software advances are also moving across the spectrum of software from everyday corporate systems and Internet applications to new mobile devices and beyond, such as the emerging space of social networking applications.

Mainstays of application architecture such as the relational database model, monolithic run-times, and even deterministic behavior are being challenged by,, and new

Let's also not forget about Web 2.0 approaches and design patterns which are becoming ever more established in online applications and enterprise architecture both.,    

To try to get a handle on what's happening and to explore these emerging topics, I've been doing conference talks lately about the transformation of software architecture that we're beginning to see in so many quarters these days and generally finding consensus that the exciting days of architecture are back, if they ever left. Now it's up to us to begin the lengthy process of taking many of these ideas into our organizations and integrating them into our thought processes and architectural frameworks and bringing them to bear to solve problems and provide value. As one software architect came up and asked me recently, "How do I get my organization to understand what's happening out there?" This is an attempt at addressing that question.

Here's a list of the most important new areas that software architects should be conversant in and looking at in 2009:

10 Must-Know Topics for Software Architects in 2009

1.Cloud Computing.Amazon EC2. While the term doesn't have an exact definition, it covers the gamut of utility hosting toPlatform-as-a-service (PaaS). I'vecovered cloud computing on ZDNet in detail beforebroken down the vendor space  

2.Non-relational databases.Is The Relational Database Doomed?" While it's far too soon to declare the demise of the workhorse relational database that's the bedrock of so many application stacks, there a large number of promising alternatives emerging. Why get rid of the traditional relational database? Certain application designs can greatly benefit from the advantages of document or resource-centric storage approaches. Performance in particular can be much higher with non-relational databases; there are often surprisingly low ceilings to the scale of relational databases, even with clustering and grid computing. And then there isabstraction impedance, which not only can create a lot more overhead when programming but also hurts run-time performance by maintaining several different representations of the data at one time during a service request. Promising non-relational solutions includeCouchDB, which I'm starting to see in more and more products, as well asAmazon SimpleDB,DrizzleMongo, andScalaris. While many applications will continue to get along just fine with relational databases and object-relational mapping, this is the first time that mainstream database alternatives are readily available for those that are increasingly in need of them.  

3.Next-generation distributed computing.excellent story today in the New York TimesHadoopvideo interview with Mark Risher and Jay Pujaraenables them to slice through billions of messages to isolate patterns and identify spammers. They can now create new queries and get results within minutes, for problems that took hours or were considered impossible with their previous approach." While Hadoop has considerable momentum, other similar offerings include the commercialGridGainDisco  

4.Web-Oriented Architecture (WOA).I've discussed Web-Oriented Architecture on this blog now for several years and my mostcomplete write-up is here. In short, the premise is that RESTful architectures (and the architecture stack above and around it including data representation, security, integration, composition, and distribution) are a more natural, productive, and effective way to build increasingly open and federated network-based applications. TheWOA debategood overview I did a short while back  

5.Mashups.wondered today in Infoworldmashupsnext major new application development modelsEnterprise Web 2.0 Predictions for 2009Mashup Patterns  

6.Open Supply Chains via APIs., which an increasingbody of evidenceopen supply chains." Opening up your business in a scalable, cost-effective manner as a platform for partners to build up on is one of the most powerful business models of the 21st century. However, there seems to be a large divide between native-Web DNA companies and traditional organizations in understanding how important this is (it's increasingly mandatory in order to compete online). All evidence so far points to this as one of the most important, though potentially difficult, things to get right in your architecture. Security, governance, scalability, and ease-of-consumption are all major subject areas and our enterprise architetures and SOAs must be ready for this business strategy as more and more organizations open up. Here's my recent"state of the union"  

7.Dynamic Languages.JavaErlangdetailed comparisonRails,CakePHP, andGrails  

8.Social computing.social computingReed's Lawof my view of Web 2.0 in the enterpriseThe Elements of Social Architectureand I've identified some of the core patterns for this in my Enterprise 2.0 mnemonic,FLATNESSES. Fnding a high-value place for social computing in our enterprise architectures will be essential for modern software efforts.  

9.Crowdsourcing and peer production architectures.open business models50 Essential Strategies for Creating a Successful Web 2.0 Product, these architectures of participation create most of the value in the software systems that employ them. If you're not sure this is a software architecture issue, just look at Amazon'sMechanical TurkCrowdSound, that latter which is a widget that allows even end-users to dynamically include crowdsourcing into their applications. You can also read John Tropea'snew exploration of this topic  

10.New Application Models.OpenSocial applications