* Networking for application delivery
In 2000 and 2003, EMA embarked on some research that we called NFAD – not actually a reference to a Middle East terrorist group, but an acronym meaning “networking for application delivery.” Those reports looked at a wide variety of technologies and disciplines used by both the data center and the NOC to address the monitoring and active improvement of application performance over the networked infrastructure.
Well, needless to say, with the growth of VoIP and IP convergence, the nodding awareness that wireless may become something more than a confused mud puddle when it comes to management, and perhaps most significantly – the advent of Web services and service-oriented architectures – improving networked application performance is key. This is all the more true as work habits become increasingly distributed, markets have become more and more global, and interdependencies in our “flat world” from a macroeconomic level have placed the network at the center, rather than the periphery, of end-user application experience.
Some things are not likely to change between 2006 and 2003. For instance in both reports, security and troubleshooting – as general categories – were the two top salient priorities and I don’t expect to see a difference now. However, I am anticipating some new shading in some of the following areas:
* Organization and process – a heightened attention to best practices and the grudging awareness that IT cultural transformation is necessary.
* A dramatic rise in the importance of configuration management.
* The importance of active management in terms of QoS, application acceleration, server load balancing, and other areas.
* The rising importance of transaction-driven analysis and a more serious concern for end-user quality of experience.
* Application modularity as a more transformative issue, as in the case of Web services with its unpredictable “machine-to-machine” communication.
* The rising importance of integrating pre- and post-deployment processes and technologies.
* The growing importance of flow-based management and real-time monitoring vs. traditional, delay-implicit polling.
* A more sophisticated approach to infrastructure optimization.
* A rising awareness of the importance of cost analysis, including capturing service utilization and demand.
On the other hand, one thing I do expect will remain consistent above all: confusion in the face of simply too many technologies, tool sets, and residual “armed camps.”
What do you think? Let me know.
