* Managing app performance - On the verge of a new breakthrough or still stuck in the mud?
EMA is about to embark on research targeting technology choices, organizational dimensions, satisfaction levels and issues involving managing application performance over the network. Once again – as with CMDB adoption – I am entering this research with a private wager, or if you prefer, a private agenda.
My agenda is actually twofold: The optimistic version is that technology advances and new levels of attention to best practices are leading to a breakthrough in how applications services are managed over the network. The skeptic’s version of this same position is that while six or seven years ago, the network management team woke up to the word “application” – as something it would have to contend with, not much has changed beyond, to coin one of my favorite stolen phrases, “more intelligent finger pointing.”
In other words, while the seeds for progress have been sown in terms of advances in tool set technologies, awareness of application issues and the rumblings of relevant improvements in process and organizational dynamics – those seeds have so far borne fruit in advancing individual performers and tactical initiatives, rather than furthering strategic initiatives and organizational transformation. Most simply put, we are empowering individuals primarily, rather than empowering organizations.
You might well wonder, what’s wrong with that? If network engineers, for instance – probably the most common group addressed by many of the advances here – are more productive, effective and are more able to correlate network and application performance, isn’t that enough? Or if new tools in configuration management are allowing network managers to achieve better operational efficiencies and reduce configuration errors that impact application performance, what more could you ask for?
My answer is that from a technology perspective, these are excellent first steps. I’m all for making individual performers more productive and introducing advances in adapting, for instance, flow-based insights into troubleshooting and capacity planning. I’m all for providing individual performers with better quality of experience metrics to assess the efficiencies of their own services and the services of outsourced providers, or leveraging more advanced analytics to support anomaly detection and intelligent baselining.
My emphasis on organizational vs. individual empowerment has to do with two assertions. The first is that managing applications across the network is accelerating in importance to a degree unimagined in the past because of the increasingly distributed nature of working, the growth in application complexity, the converging IP services – most conspicuously VoIP, and new technology areas such as wireless and Web services and service-oriented architectures. As this last paradigm becomes more and more prevalent, and I know this will take time, the network will have to become much more resilient and dynamic. This is not only in a utility sense but even more so as an intelligent source of monitoring application health and instrumenting and actualizing complex policies. And it will have to do this in as close to absolute real-time as possible.
The second “fact” is that given the above, managing application delivery over the network will require tightly knit team dynamics supported by management technologies that allow application and network specialists to see the world consistently. Here, initiatives such as ITIL and the CMDB will join with advanced analytics and flow-based insights that can provide common ground to support cohesive dynamic decision-making. The notion that managing application delivery over the network can be done effectively by a group of enclaved individuals with siloed tools – no matter how good those tools are – simply won’t work when so much is at stake and when the levels of interdependency between network performance and application performance are so high.
There is, of course, another approach – one that doesn’t exclude this direction but which, you might say, complements it. And that’s the likely rise in outsourced services to address all this complexity. EMA’s research will look at this requirement, as well.
