Back in the “good” or “bad” old days, life used to be simpler for those of us focused on BSM. Application Management, though strategic, was domain-centric and BSM was where everything came together.
But as technologies evolved, and a growing focus on real-time currency, in part due to cloud, caused a combination of panic and realignment in many IT organizations, APM became more cross-domain in nature and BSM became more of a watchword for strategies that get postponed for blue skies and quieter times.
As an analyst covering BSM who works closely -- virtually hand-in-hand with our applications management lead, Julie Craig -- this has been an interesting year for both of us. We’ve had to sort through how we cover what we do and found, I think, good answers based not on market definitions and their associated acronyms, but on problem sets, points of view, and IT constituencies.
It is interesting to me how much acronyms, technologies, and trends in “high-tech chic” come and go, but how rooted actual problem sets are and how grudgingly they actually change, or in reality progress.
So, for what it’s worth, I’ll provide some EMA insights on APM and BSM which are best viewed as overlapping circles rather than discrete boxes, or as I like to say: “we’re often looking at the same mountain, albeit from different sides.”
This is a big debate in the industry -- in part thanks to ITIL -- but it’s one on which Julie and I didn’t waste much time. I take the simple-minded approach that an IT “business service” is anything that meaningfully touches the daily work life of a non-IT user, whether it’s a single, simple application lodged on a desktop for productivity, a communications-centric e-mail service, a complex ERP business ecosystem, VoIP, video streaming or whatever. If IT owns and delivers it, and if a significant population of consumers (internal or external) depend on it to get through their day, then it’s a business service.
In other words, it’s a business service if IT is to blame when there’s a marked rise in chain smoking, drug addiction, or serial violence due to service disruptions due to poor performance, design, stability, or simply lack of choice.
For a few reasons – I took the lead on this one. Partly because I’d pioneered UEM research at EMA and partly because UEM, if fully explored, goes a long way to support business impact values, governance, SLM and other BSM requirements.
Seven years ago, I did a few webinars stating that UEM (well, we called it QoE back then) was going to be the new SLM. Meaning that if you really cared about meaningful business metrics for governance, all the vagaries of UEM were the right place to start.
UEM can also shed light on HOW individuals use applications as they relate to business efficiencies, business outcomes and business relevance, and so can be an instrumental part of portfolio planning -- this, sadly, still much more of a future potential than one actualized today in most IT shops.
EMA has done a lot of research in organization and process, and the truth is that IT organizations are evolving differently. To some degree, I’ve worked most closely with Operations concerns on the one hand, and VP and C-level executives on the other. Julie has worked most closely with application support and application owners from the applications area but of course carries over into operations as well. There are also business owners for applications that touch both of us.
In general, though, I’d say APM today tends to live in between lower layer BSM concerns centered in operations and cross-domain infrastructure, and higher layer BSM concerns centered in business impact and governance. Needless to say, this “Oreo Cookie” approach can cause confusion and disruption unless you take a collaborative approach inclusive of both APM and BSM, as EMA does.
Needless to say, most of this is Julie’s. However the area of strategic dialog between IT and business planners for higher level portfolio priorities is the domain of BSM (and by coincidence is literally defined as such in ITIL v7). This “strategic” discussion of value, relevance, business impact and IT optimization is still, sad to say, in something between a desert wasteland and an actual abyss—where culture, language, values and mindset are still largely separate and disconnected between the two groups. But there are improvements coming for a variety of reasons -- and perhaps somewhat perversely, cloud is one of them.
This is shared across APM and BSM -- especially as some new entrants in ADDM are targeting real-time, transaction-centric performance interdependencies versus a more configuration-centric approach to application supportive infrastructure.
Along with service modeling, ADDM is also becoming the new spine for a CMS as it extends beyond a CMDB with benefits ranging from asset management, governance and compliance, to change impact analysis, data center consolidation, and now actual performance management. It’s worth noting that a number of virtualization initiatives tracked by EMA have benefited strongly from ADDM.
CMDB/CMS remains a foundational piece to BSM, although also supportive to APM, especially since so many application performance issues are caused by disruptive planned and unplanned changes.
EMA will be placing a strong focus on analytics in 2012. And in some respects, APM-related analytics are beginning to mirror some of the broader, cross-domain assimilative benefits of BSM solutions.
As per my prior blogs, analytics is a rich and varied area and can pertain to financial planning and optimization, provisioning, change and capacity planning, performance management, governance, security, and business planning. The fact that analytics as a whole is getting increased attention by both vendors and IT is good for both APM and BSM.
Automation and analytics go hand-in-hand and this is more the province of BSM -- especially truly cross-domain automation procedures in which processes can be streamlined, codified, and hardened to support complex IT processes.
EMA’s simple sketchpad for automation groups includes people-to-people, people-to-machine, and machine-to-machine. BSM automation takes data center automation up several levels beyond traditional “systems management” into a more holistic, service-centric perspective.
This is admittedly far from a complete list of either APM or BSM concerns. However, I thought I’d share it as a touchstone of how EMA is tackling the broader discussion.
EMA also recognizes that just as APM is shifting in definition, BSM as a term, may fall out of fashion and may need to redefine itself in one or multiple new flavors. However, in my opinion this would be a shame, because the core of BSM is its cohesive look at end-to-end service management with a focus on domain neutrality, IT optimization and business impact -- a cohesiveness that could all too easily be lost in more fragmented but also more trendy buckets.