By John Dixon, Consulting Architect, LogicsOne

A few weeks ago, I took part in another engaging tweetchat on Cloud Computing. The topic: the enterprise cloud. Transcript here: http://storify.com/CloudCommons/cloudviews-tweetchat-enterprise-cloud

I’ll be recapping the responses to each question posed in the Tweetchat and giving an expanded response from the GreenPages perspective. As usual with tweetchats hosted by CloudCommons, five questions are presented a few days in advance of the event. This time around, the questions were:

1. How should an enterprise get started with cloud computing?
2. Is security still the “just because” reason for not migrating to the cloud?
3. Who is responsible for setting the cloud strategy in the enterprise?
4. What’s the best way for enterprises to measure cloud ROI?
5. What are the top 3 factors enterprises should consider before moving to a cloud model?
6. How should an enterprise measure the success of its cloud implementation?

Before we jump in to each question, let me say that the Cloud Commons Tweetchats are getting better and better. I try to participate in each one, and I find the different perspectives very interesting. The dynamic on Twitter makes these conversations pretty intense, and we always cover a lot of ground in just 60 minutes. Thanks to all of the regulars that participate. And if you haven’t been able to participate yet, I encourage you to have a look.

How should an enterprise get started with cloud computing?

I’m sure you’d agree that there are lots of different perspectives on cloud computing, especially now that adoption is gaining momentum. Consumers are regularly using cloud services. Organizations large and small are using cloud computing in different ways. Out of the gate, these different perspectives came to the surface. Here’s a recap of the first responses (with my take in parentheses). I don’t disagree with any of them; I think they’re all valid:

1. “Ban new development that doesn’t use cloud … as a means to help development teams begin to learn the new paradigm” (maybe a little harsh, but I can see some policy and governance coming through in that point – after all, many corporate IT shops have a virtualization policy that kind of works this way, don’t they?)
2. 2.       “Inventory applications, do some analysis, and find cloud candidates” (this is definitely one way to go, and maybe the most risk-averse; this perspective holds “the cloud” as a destination)
3. 3.       “Use SaaS” (certainly the easiest and quickest way to start using cloud, if that’s a mandate from management)
4. 4.       “Enterprises are already using cloud, next question” (I definitely agree with this one, enterprises are already using cloud for some things, no doubt about it)
5. 5.       “Look at rogue IT, then enhance and wrap some governance around the best ideas” (again, I definitely agree with this one as a valid method; in fact, I did a recent blog post on the same concept,
6. 6.       “Know what you need from a cloud provider and be prepared” (the Boy Scout model, be prepared! I definitely agree with this one as well! In fact, look here.)
7. 7.       “Partner with the business to determine how cloud fits in the COMPANY strategy, not the IT strategy” (this was from me; and maybe it is obvious by now that cloud has huge business benefits, not just benefits for corporate IT)

 

There was lots of talk about the idea of identifying the “rogue IT” groups and embracing the unique things they have done in the cloud. All in all, these are ALL great ways to get started with cloud. In hindsight, I would add in another method of my own:

1. Manage your internal infrastructure as if it were already deployed to the cloud. Some tools emerging now have this capability – to manage infrastructure through one interface whether it is deployed in your datacenter, with Rackspace, or even Amazon. This way, if and when you do decide to move some IT services to an external provider, the same tools and processes can be applied.

 

Your organization may have some additional methods to get started with cloud (and I’d love to hear about them!). So, why not use all of these methods in one concerted effort to evaluate cloud computing technology?

 

Is security still the “just because” reason for not migrating to the cloud?

The short recap on this topic: yes, organizations do use security as a convenient way to avoid acting on something. The security justification is more prevalent in large organizations, for obvious reasons. I’d like to point out one of the first responses though:

“…or is security becoming a reason to move to the cloud? Are service providers able to hire better security experts?”

I think this is a fantastic, forward looking response. History has seen that specialization in markets does occur. Call this industrial specialization: eventually…

• “The price of infrastructure services will be reduced as the market becomes more competitive. Providers will compete in the market by narrowing their focus on providing infrastructure in a secure and reliable way – they specialize or go out of business.” To compete, service providers will find/attract the best people who can help them design, build, and test infrastructure effectively
• Thus, the best people in IT security (a.k.a., the people most interested in security) will be attracted to the best jobs with service providers

Who is responsible for setting the cloud strategy in the enterprise?

Common answer was C-level, either CIO or even CEO. Cloud computing should enable the strategy of the overall business, not only IT. I think that IT should own the cloud strategy, but that more business-oriented thinkers should be in IT!

 

What’s the best way for enterprises to measure cloud ROI?

Lots of perspectives popped up on these topics. I don’t think the group stood behind a single answer. Here are some of the interesting responses for measuring ROI of cloud:

• IT staff reduction
• Business revenue divided by IT operations expense
• Improving time to market for new applications

Measuring the value of IT services is, excuse the pun, tricky business. I think cloud adoption will undoubtedly accelerate once there is a set of meaningful metrics that is applicable across industries. Measuring ROI of a virtualization effort was fairly easy math – reduction in servers, networking, datacenter floor space, etc. Measuring ROI of cloud is much more difficult, but the prize is up for grabs!

 

What are the top 3 factors enterprises should consider before moving to a cloud model?

This goes back to the Boy Scout model of proper preparation, which I wrote about a few months ago. I saw a few responses that were especially interesting, and yet unsolved:

• Repatriation, or portability of applications
• Organizational change (shouldn’t cloud be transparent?)
• Investments in legacy technology, goes to timing and WHEN to consider cloud
• Security, location of data, etc.

 

At GreenPages, we think of cloud computing more as a management paradigm, and as a New Supply Chain for IT. Considering that perspective, the points above are less of an issue. GreenPages  Cloud Management as a Service (CMaaS) offering was designed specifically for this – to view cloud computing as the New Supply Chain for IT. In a world of consumers (enterprises) and providers (the likes of Amazon, Rackspace, and Terremark), where competition drives prices down, cloud computing, like other supply chains, can be thought of as the way to take advantage of market forces to benefit the business.

Thanks to Cloud Commons for another great conversation…looking forward to the next one!