Optimizing Cloud - Cirrus, Cumulus, Stratus and Nimbus
January 08, 2012

Dennis Drogseth

Share this

Each year since Q1 2010, EMA has done a major research effort on how cloud-related technologies were being adopted by IT and service provider organizations. In 2010, we looked at basic patterns of cloud adoption from a management perspective — with a balance between systems management and service management concerns. Last year at this time we reviewed how cloud technologies were becoming assimilated into top-down service management strategies.

This year the goal was to establish a more current vision of how cloud-related technologies — from converged infrastructure, to autoscaling, to VDI, to virtualized applications, to cloud bursting — were combining with critical management capabilities ranging from application performance and User Experience Management to cross-domain configuration automation, CMDB/CMS, service desks, chargeback and accounting, etc. across what turned out to be more twenty potential management or cloud technology choices and many multiples more of potential combinations.

I’m just finishing up the analysis — I had more than a thousand pages of data to review — and will be giving a Webinar on the research results on February 2nd. (You can register at www.emausa.com)

But for this blog I thought I’d present a few highlights that reflect a definite progression to show how and why cloud IS in fact becoming assimilated by IT. You, by the way, are the very first to see some of these results thanks to the way APMdigest deadlines fell in the process of preparing the analysis.

First a few table stakes.

We had 160 global respondents with the preponderance in North America but a significant population in Europe/ UK. All respondents were actively involved in cloud-related initiatives (we had tricks for weeding out those who weren’t).

About 55% of the respondents were executives or managers, and the rest professionals — the great majority of whom had direct “hands-on” involvement in cloud. Most of the companies were enterprise — including 25% with more than 20,000 employees — but we also had a significant population of between 250 and 1,000 employees.

Nearly half, 48%, said that cloud was adopted and was an “essential part of their business.” This was up from 33% last year—which was up only marginally from 2010. Only 12% said cloud was adopted by “only supplemental” to other types of computing, a drop downward from 29% the year before. The rest felt cloud was “important.”

So for those of you who remain skeptics — cloud seems to be here to stay. But just what that means only opens the door to a whole new set of complicated questions. We parsed on internal and external cloud services with more granularity – but the clear trend was to show that smaller companies had more external cloud, while larger companies had more internal cloud with an overall increase in hybrid (internal and external) deployments.

Oh and for other year-to-year trends — IT budgets continue to grow more than decline. Last year 60% of budgets increased and only 17% declined. The rest stayed the same. That looked pretty good then. But this year shows that 69% grew and only 8% declined, while growth at higher percentages was definitely on the upswing.

This year we looked a lot at skill sets, technologies and how they clustered with a number of factors: success rates, company size, and overall flavor of cloud.

And I’m not ready yet to develop a definitive taxonomy, that’s why the headline features cumulus, stratus, cirrus and nimbus as options.

The fact is — just like cloud formations — cloud adoption patterns cluster in various forms for various reasons. Cloud formations are assigned genus — e.g. cumulus, nimbus and then combinations such as cumulonimbus, a tall, vertical rain cloud that looks a little like a wannabe cyclone or nuclear explosion that never touches the ground.

I’m sure that cloud adoptions in IT will have patterns just as complex if not more so, once variables are combined such as company size, business service, skill sets, organizational role, business vertical and model, management technologies, and internal and external cloud technologies adopted.

The top five skill groups involved in cloud in current research were: Infrastructure Services, Application Performance Management, Infrastructure Performance Management, Infrastructure Architecture and Design, and Capacity Planning and Optimization in that order.

The most highly ranked management or cloud related technology was User Experience Management much to my pleasant surprise – which narrowly edged out virtualized configuration and patch management. The lowest ranked in the list of 22 — though still seen as relevant and important — was “Service Catalog.” But these rankings could be deceptive depending on which breadcrumb trail you followed in seeking to define the equivalent of a cumulonimbus cloud formation.

And this is the area I’m still pursuing — and probably will continue to do so well into the year even after the Webinar.

But I’ll give you one example of what I’m talking about by picking one of my favorites — the CMDB/CMS. CMDB/CMS was in the lower quarter for both technologies deployed in support of cloud and “critical importance” rankings. But was four times more likely to get a “critical” ranking by IT organizations who felt their cloud deployments were “Very Successful” versus “somewhat successful” or lower.

CMDB/CMS deployments were more likely to involve C-level execs, be deployed in support of cloud in mid-tier enterprises (versus small businesses or very large enterprises with more than 20,000). The CMDB/CMS group was also yet more likely to see its revenue increase, dramatically more able to support more business services via cloud, substantially more progressed in adopting a wider variety of cloud services, more likely to have a wider array of other advanced management technologies deployed, and more likely to have internal and external SLAs and OLAs in place, etc.

I’ve looked at the similar patterns for SLM/UEM, Chargeback and Usage-based Accounting, and Capacity Planning and Optimization Analytics so far, and while all show areas of strength, none have come close to CMDB/CMS in clustering such broad advantages.

Now, I can’t say that CMDB/CMS is entirely the cause here. It could also be partly the effect as it signals a more mature, more committed, and more executively engaged IT organization. But I should also stress that these patterns aren’t linear. They favor different qualitative values by and large, as well. So that, for instance, Usage-based Accounting deployments typically signal a significantly stronger skill and purpose focus on asset management and financial planning with direct ties to capacity planning. Maybe it will become the stratus cloud technology in the overall taxonomy.

I’ll end on one point that has stood out in this as well as other surveys, but was exceedingly dramatic here. The absolutely most advanced company size group was 5,000 to 20,000 employees. Below, the focus gradated downwards towards more external services and less internal sophistication on average. And above, while internal became yet more important, the technological and organizational maturity dramatically degraded.

And while there are very visible exceptions, this is common in my experience. Why, you may ask, when these are the biggest, richest, most powerful IT organizations overall? The answer as always is, sad to say, “politics.” In this case, the politics of entrenched fiefdoms — combined with a massive and often ungovernable array of investments.

Dennis Drogseth is VP at Enterprise Management Associates (EMA).

Related Links:


Dennis Drogseth is VP at Enterprise Management Associates (EMA)
Share this