Colo vs. Cloud

For sometime, I have been meaning to compare the costs of colocation to cloud. Somebody beat me to it. Some will try to use this model as the definitive guide as to which solution to choose, without their own inputs. I think the key takeaway from the post is:

This model presents an analytical framework which you are free to copy and then sharpen up with your own business model and cost structure.

Translation: It all depends on your situation. But now you have a framework to start with.

I also find the comments to thepost definitely worth reviewing.

I guess it is all about services after all …

From Terremark data:

On the revenue side, the managed hosting space generates $20,000 a year in revenue for each square foot, compared to $12,000 a year for cloud computing and $1,000 a year for colocation. That works out to an annual return of $2.50 for every $1 invested in managed hosting, compared to $1.50 for cloud computing and 90 cents for colocation.

While colocation offers the lowest cost of entry, it also has the lowest return. Cloud computing and managed hosting offer a healthy return on investment, but also require an investment of $8,000 per square foot (at least in Terremark’s operations; others may do this for less).

That’s why we won’t see the landscape overrun by cloud computing data centers. Cloud economics offer faster, cheaper deployment for end users, but the barrier to entry remains high on the back end. And the $8,000 a square foot investment in infrastructure only pays off quickly if you can fill the space with customers. For many providers eyeing the cloud, a strong track record in acquiring customers is a prerequisite to access to capital.

From Rackspace data:

But Rackspace says that the efficiency of cloud computing, and the ability to pack many users on each physical server, makes cloud computing a winner in a revenue per server analysis. The ability to automate tasks that previously required support staff is another key benefit of the cloud model. Lower supports costs don’t appear to be factored into the Terremark data, and certainly would narrow the profitability gap between cloud and managed hosting.

The Conclusion

The higher return on capital for managed hosting is among the reasons why this hosting model will never be fully superseded by cloud computing. Many customers will continue to pay a premium for high-touch support and service for their mission-critical IT assets.

But cloud computing offers a middle path, offering cost and usability advantages for customers, as well as an attractive return for providers. For established data center builders and operators, this creates a “best of both worlds” scenario: a fast-growing customer base and a high barrier to entry for new players.

My conclusion:

Become a managed cloud provider, infrastructure, os, select applications, whatever.  But find a managed cloud niche.

Full article link.

Educate First, Then Sell

I have always felt that the best customer calls, sales or otherwise, occur when you are able to provide some education to your customer about the services you provide.  I do not mean your specific version of service.  I mean the service you provide generally.  When we first introduced our co-location offerings, it was just an important to be able to show our customers how you can use co-location to improve IT generally as well as the features our specific co-location offering.  We had to educate first, then sell.

Looks like cloud service providers should be focusing on this as well:

More than four in five IT managers do not use cloud technologies, despite the growing interest in the subject.

That’s according to research from web hosting company Peer 1, which also revealed that more than a third of those that don’t use the cloud admitted it was because they didn’t have enough knowledge of the technology.

So you really have to sell many customers twice, once on what cloud computing offers them, specifically, and then once more on why your cloud solution is the best one for them. I would certainly be focusing on cloud computing education, whether whitepapers, webinars, seminars, or general advertising.

What do your customers need to know?

• The general benefits of cloud computing
• What uses make sense
• How to plan for costs
• How to migrate to cloud solutions

The opportunities are many. Education makes you a thought leader. Do it now when there is no clear thought leader.

Measuring VM Energy Usage

Propeller-head alert:  This post is a little technical/geeky.

Microsoft has a new application to measure VM energy usage.  While the details may be beyond the concern of most business owners, energy usage is certainly paramount in operating cloud services.  I will leave the blog post and related links to stand on their own:

http://www.greenm3.com/2010/03/microsoft-research-paper-measuring-energy-use-of-a-virtual-machine.html

http://www.greenm3.com/2010/03/microsoft-research-images-of-measuring-energy-use-of-a-virtual-machine-joulemeter-part-2.html

http://research.microsoft.com/en-us/projects/joulemeter/default.aspx

http://research.microsoft.com/en-us/um/redmond/events/techfest2010/posters/Posters/128-Energy-Aware-VMs.pdf

What managed services is really all about

Managed services, and to a certain extent, cloud computing, are about people.  If you have the people that can efficiently architect and operate your IT infrastructure, consider yourself lucky.  But rather than resting on you laurels, perhaps those kind of people are better off selecting and implementing applications or other IT projects that add business value and competitive differentiation.  But don’t just take it from me.  How about the CIO and CTO of PricewaterhouseCoopers:

“The problem is, not that many people out there right now are versed in the data center of today,” he says.

By “data center of today,” Ancona means a highly automated, dense and virtualized IT infrastructure that relies on the most advanced electrical and mechanical components for ultra power efficiency.

The modern data center demands that IT professionals understand multiple disciplines as well as the facilities infrastructure — and that’s one of the big issues giving rise to the hiring difficulties.

The entire article is worth a read, but the key takeaways are:

• Virtualization and storage skill sets are in high demand with low availability
• Engineers need to understand the applications being deployed
• Generalists are just as (if not more) important as specialists
• 1/3 of organizations in a recent survey reported projects on hold because of skill shortages.
• If PWC can’t find and hire people, what makes you think you can (MSP or Corporate IT)?
• What are you doing to make sure PWC doesn’t hire your good people?

MSP Wanted: Must like ITIL

I have been speaking with several MSPs and companies that want to become an MSP or acquire an MSP.   The number one question I get is, what exactly is it that we should be looking to acquire/build.

My first thought is, it is hard to find good acquisition candidates at this point. Perhaps it is the relative fragmentation and immaturity of the provider landscape. Perhaps it is because there are a lot of regional players that have other lines of business along with their managed services practice. Perhaps it is the large recovery in the public market and the difficulty of setting valuations on this “newer” business line.

But, if I were looking to acquire an MSP, here is what I would be looking for:

• A proven implemented operational framework (ITIL, COBIT, ISO) that has been implemented and audited against.  From my experience, this something that you can NOT quickly build.  It has to seep its way into the operational fabric over a period of time. Operators will see it as a pain at first, but will soon see at is necessary to maintaining the availability and performance of client systems.
• Mature monitoring and reporting tools.  A minimal number of tools to manage. Tools that provide technical monitoring as well as application monitoring. Usage reporting, dashboard reporting, and SLA reporting.  And, a program to minimize the monitoring noise and optimize the data the tool provides.
• Customer Portal. Can the customer see information about their network and systems from a website? Configuration information? Operations information? Dashboard reports? Service requests? Invoices?  A single portal is definitely a sign of maturity.
• Continuous Improvement Program.  How does the MSP measure success and progress?  What are the progress metrics and do they truly measure against them?

This list is certainly not exhaustive, but gives a flavor of what I think are important.  What am I missing? I will discuss sales and marketing competencies in a separate post.

Cloud Progression

I met with a managed services provider earlier this week and we had a discussion about cloud computing and their strategy. The provider is a medium-sized managed services provider, fairly mature in their management capabilities, but limited offerings in the hosted/cloud space. During our discussions, they drew my attention to the following diagram:

from this post.

I found it very useful to show the “probable” natural progression to cloud computing, especially for back-office applications. The entire post and the comments are worth reading.

Architecting for the Cloud

One of the bigger opportunities, both from a consulting perspective as well as an employment perspective, is to be able to understand how cloud computing works and how to architect applications and IT infrastructure in general to take advantage of it.  This whitepaper from Amazon is a very good summary of these issues and is beneficial for cloud computing understanding in general, not just Amazon.  However, I think everyone working with cloud computing needs to understand what the giant in the room is doing, and I am not sure if anyone else but Amazon qualifies for that role at this time.

I am especially intrigued by the Amazon Virtual Private Cloud and the Security Best Practices.  I will follow-up over the next several weeks with separate posts on each.

No more CIOs? Part II

Part 2 of the article that prompted the post “No More CIOs?” can be found here.

Here are some relevant snippets with my thoughts following:

Right now, more than 15 million computer servers are operating worldwide. Yet, 80% remain idle every day. You see, every organization has to build its network for its peak demand.

I have used this 80% idle thought as well.  Does anyone know what the real, verified rate is?  If nothing else, I would think it has come down since the widespread implementation of VMWare and other virtualization software within IT departments.

Fortunes can be made by catching a massive, long-term trend like this as it migrates from concept to reality. But timing is critical. The full reality of cloud computing is at least a decade in the making. The transition of cost and operational burdens of IT into the cloud will happen slowly and incrementally. [My emphasis added.]

Not exactly sure how the fortunes will be made, but I do think there are opportunities, obviously.

I like the handful of “infrastructure as a service” players like Equinix (EQIX), Rackspace (RAX), and Terremark Worldwide (TMRK). These shops offer data-center real estate, as well as hosting services on a “pay as you go” basis.

I did not realize Equinix was an IaaS (Infrastructure as a Service) Provider. If so, we really need to get more precise and demanding with our definitions.

To be honest, I don’t expect any of the above names to be independent companies in five years. Just as the budding electricity utility industry underwent massive consolidation in the 1920’s, I believe the pioneers in the cloud computing space will be gobbled up by the biggest tech firms as a cheap way to gain a foothold in this revolutionary trend.

If this is true, do you, as an MSP or cloud provider, have an exit plan? Or a plan to compete with this trend?

Finally, for those of you stock traders/investors:

P.S. I dedicated my most recent Phase 1 issue to cloud computing. Longtime readers know I rarely pursue opportunities outside of the health care and biotech space. I only venture out when three-, five-, even 10-bagger returns are possible. I’ve lined up two cloud computing names that fit the bill. To learn more about Phase 1, click here.

Any guesses as to the providers?  If I had to guess, it would be someone in the enablement space. And public, obviously.  Shouldn’t be too many that fit that bill.  I will follow up if I make any headway in identifying them.

No more CIOs?

Ran across this article from one of the investment emails I subscribe to. I wonder if it is a sign of a current short-term top in the cloud computing hype cycle.  The relevant snippet I found interesting/telling is as follows:

Now ask yourself another question: Does your company employ a vice president of information technology? Or a chief information officer?

Doubtless your company has hired someone to maintain all the computing infrastructure your company requires to function… In 10 years, that position will be as obsolete as vice president of electricity.

Now 10 years is a long-time to predict anything.  And I am relatively certain that a large majority of the IT infrastructure in 10 years will be cloud-based. BUT, isn’t there more to IT than Infrastructure?  And even for cloud services, doesn’t someone still have to:

• Research and select cloud providers
• Migrate data and applications to the providers
• Integrate into the rest of the infrastructure, whether premise-based or cloud-based with other providers
• Monitor the performance of the cloud providers
• Manage and optimize relationships with existing cloud providers

Seems to me that their is still infrastructure management to be done, it is just a different type of management.

More importantly, it seems like cloud services should will put serious pressure on CIOs (and IT in general) to add value to the business.  And isn’t that what IT is supposed to do in the first place? Seems like the beginning of great opportunities for CIOs (or VPs of Infrastructure) not a threat.  Unless you do not know how to add value.

But, if I am the VP or Director or Manager of IT Infrastructure, the next 2 years, much less 10 years, are certainly going to see significant changes to how I have done my job over the last 10 years. More discussion on that in later posts.