This is the fifth and, I promise, last column on cloud (at least for a while). In this column, I’m going to pretend that you’re already dealing with some cloud initiatives — whether you had anything to do with them or not — and trying to take some control over chaos . As some of you may remember, I had proposed appointing a “Cloud Tsar” to document and evaluate who was doing what inside and outside of IT — whether software-as-a-service (SaaS), infrastructure-as-a-service (IaaS), or platform-as-a-service (PaaS).
But, in this column, I’m going to draw on some December data from a global research initiative EMA did called Operationalizing Cloud to give you some pointers about what to do next. Just for the record, we surveyed 155 respondents globally across verticals for companies of 500 and above employees. Of the respondents, 26% were C-level and 57% were director level and above. This was purposeful. We were seeking the “top-down” view on cloud deployments.
Starting over
Perhaps what is most striking is 70% of all respondents had to “redo” or “rethink” their initial cloud initiatives! The dominant change they wanted to make was process re-definitions so that different individuals and groups could work together more effectively to support the dynamic, cross-domain requirements cloud engenders.
Changes in management technology adoption for provisioning services and monitoring and assuring services for performance were near the top of the list, as well, along with changes in application development and design.
It might be also worth noting that more than 25% of respondents also had to re-assign ownership for organizational leadership. And, as IT executives, you may well want to know just what those organizational changes are likely to be. To answer this, I’ll combine a little data with a little observed insight from working with IT organizations and hope the combination shapes up for you.
Cloud teams
As a single organization, Cloud or Virtualization Support edged out Data Center/Server Ops as the primary owner for cloud. This represents a radical jump upward (doubling in percentage) over a year ago. This is a self-styled, specifically focused team that may or may not have strong cross-domain capabilities, depending on the maturity of the organization and of the effective maturity of the cloud deployments — the two go hand in hand. Perhaps more significantly, the third entrant here, architecture/infrastructure services, when combined with cross-domain service management, became the single, most pervasive organizational force in assimilating cloud and also the most successful one.
When we did the analysis, we saw that this group led in accelerating cloud deployments, and was also more advanced in deploying critical service management technologies that in turn brought strong benefits based on our data. We tracked twelve technologies, but a few examples are CMDB/CMS, application dependency mapping, service catalogs, integrated service desks, and IT process automation, or runbook.
We also asked respondents if they had a “cross-domain services management group” as a general statement of maturity. Now, I know that this could mean many things and go by many different names but the data that came back was nevertheless astonishingly consistent: If you have a cross-domain services management group not only are you categorically ahead of IT organizations that do not, you are significantly more likely to see your IT budget increase!
In terms of cloud, you are advantaged in both the extent to which cloud services can be assimilated, and the effectiveness with which they can bring value. The value we tracked included opex and capex savings, provisioning new and existing services, improved service performance, and even enhancing business models and supporting revenue creation, among others.
This group, by the way, is more often than not led by… you! Better than 50% of the respondents indicated C-level oversight, and 70% were VP level and above.
If your cloud initiatives are sputtering and you don’t have a cross-domain services management team (call it what you will including “architecture and infrastructure services”) maybe it’s time to create one. After all, the demands of cloud computing aren’t limited to systems management. One could argue that virtual machines are fundamentally a networked phenomenon with disruptive impacts on the whole infrastructure, including storage, let alone applications that are themselves modular, dynamic and often unpredictable (e.g., Web 2.0) in terms of components.
Reorg for the cloud?
So should cloud computing make you rethink your organizational structure? Maybe so. Once again, it’s the tail wagging the dog, I know. But cloud computing has apparently gotten quite a lot of heft behind its tail.
And it’s still worth doing. 57% of our respondents were “somewhat satisfied” and 34% were “completely satisfied” with their cloud deployments. That’s up only a tad from last year.
To distill an awful a lot of words (five columns worth) into three key points I’d say:
- Cloud is a confusing admixture of technologies — it’s important to prioritize where and how you want to adopt them; ·
- Cloud is not an end in itself, but provides a means to deliver services more flexibly, resiliently, cost-effectively and responsively if well planned, managed and integrated with broader IT service directions; and ·
- Cloud is a fundamentally cross-domain vs. “systems-only” or even “systems-centric” phenomenon once understood — and this has profound implications on process and organizational structures, as well as management technologies for assuring, optimizing and provisioning IT services. ·
So what’s next for you with cloud? I’d love to hear from you on that score.
Dennis Drogseth is VP of Boulder, Colo.-based Enterprise Management Associates, an industry research firm focused on IT management. Dennis can be reached at [email protected].