Warning: session_start(): open(/var/lib/php/session/sess_0at3fr332jndfckd8mmm6qijk0, O_RDWR) failed: No space left on device (28) in /var/www/blogs/wp-content/plugins/zemanta/zemanta.php on line 633
Next generation IT management « EMA Blog Community
posted by    | February 9, 2012 | 0 Comments

crystal_ball2-217x300

Here is my yearly prediction column, a bit delayed. Topic of the day is next generation IT management. (Not next generation IT; plenty of folks are covering that!)

The tools market for IT service management is glutted. My friend Jan van Bon has been curating a list.ly list of “ITIL tools” that is now up around 350. While there is great diversity in that list – not all the tools do the same thing – there is also tremendous redundancy. We have a surplus of “flexible workflow-based Pink-verified ITSM tools that can run either in the Cloud or on premise.” (The standard pitch on the Fusion exhibition floor.)

Clearly this is an unstable situation. We see the obvious market shakeout response in BMC’s acquisition of Numara. This is a form of horizontal integration. But what about vertical integration?  Now that we have multiple IT service management products all more or less doing the same thing, surely some will start to look for innovation opportunities outside their traditional scope.

What direction will this innovation take? Well, first, let’s look at some underlying dynamics. I believe that the Agile movement and the increasing influence of Lean are great sources of innovation in IT management. And just beneath those influences we have the insights of complexity theory and modern neuroscience. I believe these will be the fundamental drivers of IT management innovation for the next few years.

DevOps is an obvious manifestation of these deeper dynamics. It is an important development, and no-one should dismiss it. The reality is that mature, highly-scaled IT organizations like Amazon ARE making production changes with ever increasing velocity, and this is NOT happening at the expense of daily stability. DevOps is silo-busting, a bottom-up challenge to traditional IT boundaries that have been recognized since the time of Alan Turing. Now, that’s innovation.

I examined last year what DevOps means for the IT management market. Since then, we’ve also seen Application Lifecycle Management vendor Serena expand into the IT service management market. In general, it appears inevitable that project management, ALM, and ITSM will continue to converge. They all are channels of IT demand and execution, and too many people in the modern IT shop wind up handling multiple queues in part due to the existence of siloed tools. A Lean approach calls for simplifying and consolidating these queues, so look for market evolution along those lines. The majority of discussions I’m having lately with both vendors and end users confirms this.

Kanban for software development is a major factor, already a key part of tools from ALM vendors like Rally Software, and it’s only a short leap from there to kanban for operations (and then what happens to the traditional ITIL-based service desk?). More generally, part of the attraction of  kanban is that it represents new tactile and visual approaches that may be more effective for human understanding, not to mention better controlling multitasking and encouraging a sense of flow.

In terms of new visual approaches, Serena’s David Hurwitz calls for a unified operations calendar (among other things in that excellent article). In my own experience combining project, release, and change data into one overall calendar view was a highly compelling visualization that stimulated intense leadership interest. Yet how many ITSM vendors are taking full advantage of the powerful data visualization capabilities now available to software developers? Far too few, in my view.

It’s no secret that my background is as a practicing enterprise architect, and in that role I encountered many EA tools vendors. To the extent that they were just modeling tools, I never found them that compelling. Boxes and lines backed by metamodels; you can (speaking as someone who once wallpapered his cube with the UML spec) spend your life there and the rest of the world will see it as an ivory tower.

But modern EA vendors like alfabet and Troux have been moving into IT portfolio and even financial management territory for some time. They are not just modeling sketchpads, these are production systems. It’s hard to have a credible opinion about an application portfolio if you don’t know the costs. How can you have a full current/target/gap analysis if you don’t understand the financial consequences? It’s also hard to have a credible opinion about an IT budget if you don’t understand the functional business capabilities and value streams the IT infrastructure is supporting.

So, my question/prediction: when will we see the combination of an EA vendor with a full fledged IT finance vendor? That would be powerful indeed. Along the same lines is the idea of the IT data warehouse, and we also see the beginnings of a market there.

Software asset management remains a troubled area. In at least one conversation lately, the pain of managing traditional software licenses was a primary factor in driving someone to cloud services. Licensing transparency seems unfortunately to be diametrically opposed to vendors’ profit motives in too many cases, and I’m not sure we’ll see any breakthroughs soon. (Note: I”ll be speaking at the 2012 SAM Summit in Chicago.)

On the process front, look for Adaptive Case Management influences over IT management tools. Hornbill already allows for some variability in IT process execution, based on recent briefings. Process mining is an important related development, although it’s not clear how that market might evolve. Will end user organizations purchase their own process mining tools, or will they be acquired more by consultancies? Perhaps managed service providers seeking to add value will offer process mining as a service.

Further out on a limb would be the combination of process mining with ontology mining. We have so many a priori opinions about what people are doing and how they describe it. Does the evidence support those opinions? What if we threw out all the tools and told people to only work with email for a month, logged all the interactions, then mined the results. What would we see? Then, we engineer the system. On a related front, social media and Watson are already stimulating interest in unstructured text analytics for IT management. I expect to see more on that front this year.

The potential of SaaS delivery, finally, may be a key enabler for many of these dynamics. I’ve discussed elsewhere the power of such multi-tenant services, and hear increasing discussion around them.  IT financial management or the intricacies of software entitlements got you stumped? Need specialized analytics on your application portfolio? Can’t afford to do your own process mining? Get just enough expertise on demand from your IaaS or SaaS provider -  it’s happening today, and promises to increase.

Thoughts?

Credit to Aale Roos (@aalem) and James Finister (@jimbofin) for interesting ongoing background discussion on Service Desk 2.0, and also to @theitskeptic whose Future of IT Management post inspired me to get off my duff & get this written.

Enhanced by Zemanta


Posted in Uncategorized



Leave a Reply

Your email address will not be published. Required fields are marked *

*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>



Warning: Unknown: open(/var/lib/php/session/sess_0at3fr332jndfckd8mmm6qijk0, O_RDWR) failed: No space left on device (28) in Unknown on line 0

Warning: Unknown: Failed to write session data (files). Please verify that the current setting of session.save_path is correct (/var/lib/php/session) in Unknown on line 0