
I think I can institutionalize Cisco UCS
Institutionalizing a new technology means successfully adopting it so that it becomes a part of business as usual, though it might change the meaning of business as usual in the process!
Back in 2004 I first went through this process with a customer who was adopting ESX 2.0. I was challenged by a senior executive:
Steve, this VMware technology looks great and I don’t doubt that it can deliver what it promises in some organizations. But I think you’ll find that we, as an institution, will not be able to adapt to adopt this new technology: for this reason, the project is likely to fail.
This was the best start to any project. You heard me right: being told we would fail was not only a great incentive to a contrary, awkward person like me, but it also highlighted where I needed to focus my time.
In this case, and ever since, it was less about the technology and more about the culture, people and process. Guess what? Nothing has changed since 2004.
Today’s familiar challenge is: how do you institutionalize something new like Cisco UCS?
The UCS technology is not the issue here: it’s the institution. I believe there is a ratio between how successful you will be with UCS and the amount you change the institution to take advantage of it.
- At the low end of the change spectrum, you just plug the 6100 LAN uplinks into your 6500, and your Fibre uplinks into your MDS, and nothing much changes other than fitting even more virtual machines into a smaller, tidier space.
- At the high end of the change spectrum, you go beyond the technology integration and use your business-centric IT culture (what?!) to optimize your people and process too. But how? Where do you start? What’s possible? How do you get there?
Cisco have an answer for this, and it’s a services engagement called Virtual Operations Management Assessment (VOMA) plus a series of transformational projects that deliver specific, operational, outcomes.
What I love about VOMA (I’m one of the consultants who delivers this for Cisco) is that it has codified the things I have done for VMware in a really cool approach. One of the cool things is the way it looks at six dimensions:
- Organization: The organization element defines the new organizational models, new roles and responsibilities, and organizational interactions required to support data center virtualization and unified computing technologies.
- People: People elements include the skills needed to successfully architect, implement, and manage your virtualized data center environments.
- Processes: Operational processes include the service lifecycle management framework and processes required to support the virtualized data center, Cisco Nexus™ platform, unified computing solution (UCS), and overall Data Center 3.0 architecture.
- Tools and technologies: This element includes hardware and software tools and technologies that affect and enable the virtualized data center.
- Governance: Governance and the standards required to manage, monitor, and transform operating models, policies, and guidelines are included in this element.
- Metrics: This element includes data center virtualization metrics that are aligned with the governance framework and unify the integrated silos together. Service management and key performance indicators (KPIs) drive improvement and help mitigate risk and ensure success.
What you get out of this is a defined future state to aim for and a prescriptive roadmap on how to get there. It’s like ITIL but with meat on it.
But wait, there’s more. Imagine if you mixed all this cool stuff with the cool stuff from the Visible Ops guys, Behr/Kim/Spafford? Here’s just one example, a one line snippet from Kevin Behr’s blog post IT Ops or IT Slops
Technology CAN (not does) provide value IF and only IF it diminishes a business constraint.
I share Kevin’s appreciation of the Toyata Production System, Value Chain, Lean and The Goal – though I’m a mere beginner compared to him.
Bringing Cisco’s VOMA intellectual property together with the Visible Ops methodology would be incredibly powerful and help IT organizations remove business constraints through the adoption of Cisco UCS. I’m convinced of it. Now to make it happen.
Related posts:
- Measuring a Unified/Useful Computing System
- Who’s got the best Unified Computing System?
- Unified Computing = Unified People, Process and Technology
- Unified Computing Operational Processes
- Accept failure and build resilience and recovery into the system

Steve,
Let me know what I can do to help!
Kevin Behr
Hello Steve,
Could you share any sample metrics or “key performance indicators” used in step 6 ?
Do you merge the network,system and storage one or use new metrics, plus datacenter ones (Pue, DCIE) ?
Thanks.
Julien
Hey Julien, the metrics are a mix of technology metrics (e.g. capacity management will look at %READY for ESX) and process metrics (e.g change management will look at %successful changes).
I’m new to the PUE/DCiE/DCP measurements and in my travels I’ve found the input metrics hard to get. Often all I can get is an end result like “Watts per square foot” – it’s almost as hard as finding OpEx numbers :-/
Some time ago I provided a sample of virtualization-related metrics, based upon the itSMF “Metrics for Service Management” (with permission from “The Church of the itSMF” of course) – at the VIOPS site I used to run @ VMware – http://bit.ly/SE4Tw.
In the VOMA service engagement these metrics are much more comprehensive but I can’t share them at the moment as it’s not my IP to share (it belongs to Cisco). My hope is that, one day, the VOMA IP is in a public, standard assessment but that’s not the case today.
If you are interested in more detail, please send me a direct message at http://twitter.com/stevie_chambers.