Cloud Cloud Cloud. Whenever you talk to a provider, or IT analyst, consultant or investor these days, the C word inexorably oozes out during whatever dialogue or monologue is taking place. And what we love most about the Cloud, is when we ask that person to define it, there is this inevitable pause before he or she stumbles through various unstructured sentences, before asking us how we define it.
Are you really ready for the "HfS Private Cloud Challenge"?
Never before have we come across such a vague, poorly-defined piece of terminology so many people utter so freely without really understanding it, but feel an irrepressible urge to vocalize it at every given opportunity. At HfS, we’ve become so depressed by the repeated fuzziness of the Cloud, and the startling inability of anyone to put forward a convincing, proven case for the Private Cloud, that our hot-blooded analyst, Esteban Herrera, has deemed it high-time to lay down the gauntlet to you all, so over to you Esteban to explain our new challenge…
It has suddenly become almost as popular to bash the Cloud, as it has to hype it. Personally, I have been a skeptic all along, and don’t even get Phil started on Cloud BPO!
The Public Cloud mostly makes sense, but all of the technologies required have existed in some form for quite some time—they are available and people use them today, so I am not sure what is so “new” about it. At least the economics make sense: lots of companies invest in a single infrastructure with overcapacity, but because of the consolidation and standardization of the offerings, it costs them a whole lot less than developing their own infrastructure, and gives them the ability to scale. Given its limitations, the Public Cloud has been mostly popular with small and medium size.
But the Private Cloud makes my blood boil. Private Clouds are a cynical oxymoron. The whole point of a Cloud is that you share resources and don’t have to own the capacity you need, because its available on demand, so you can pay by the drink. Well, if you own the resources and the capacity, it is inherently limited to what you own, and you’ve already paid for everyone’s drinks at the bar whether they consume them or not!
Check out Wikipedia’s definition:
Private Clouds have attracted criticism because users “still have to buy, build, and manage them” and thus do not benefit from lower up-front capital costs and less hands-on management,[52] essentially “[lacking] the economic model that makes Cloud computing such an intriguing concept”.[54] [55] Enterprise IT organizations use their own private Cloud(s) for mission critical and other operational systems to protect critical infrastructures. [56] Therefore, for all intents and purposes, “Private Clouds” are not an implementation of Cloud computing at all, but are in fact an implementation of a technology subset: the basic concept of virtualized computing.
But HfS is about dialog, and believe it or not we don’t mind being proven wrong, so we are launching the HfS Research Private Cloud Challenge—a fun contest for true innovators in the Private Cloud. Here are the rules:
- Entries must describe a Private Cloud solution that has been implemented. Not a server virtualization initiative, not the purchase of Software as a Service, but a true, full-on Cloud model where the technical infrastructure, software, and services are Cloud-based, available on demand, and can scale up or down painlessly.
- Each entry must be accompanied by a business case. That’s right, real numbers on spreadsheets, folks! Entries must show in real money the benefits of the Private Cloud implementation. You may also share other benefits that are hard to measure in dollars, but assume we are taking a CFO perspective here.
- Your story must be verifiable. We are happy to protect the innocent and will not disclose any name you don’t want us to, but we want to talk to real people who are close enough to the solution and to the numbers to be able to tell us its real.
- Do not send us more than 10 pages plus a spreadsheet with the REALIZED business case (not the expected business case). The legions of HfS minions who will have to sort through your thousands of entries are quite busy.
What do you get if you “win”? If you are the first person or organization to convince us that there is business value in a Private Cloud, you will get unparalleled fame in the form of:
- Your story will be published on this very blog, where 50,000+ loyal and adoring readers will understand the benefits of private Cloud. You will be tweeted and re-tweeted the world over and every major social media outlet will know you have won the challenge.
- You will earn the satisfaction of having crushed my gargantuan ego by proving me wrong, and demonstrating conclusively that I am an ornery, luddite fool.
- I will publicly eat my fedora in front of all said subscribers, and have to admit that the Private Cloud is indeed not an oxymoron, but a valuable new way for enterprises to do business.
Email your entries to me at [email protected]. I’m fully prepared for a flooded inbox.
Posted in : Cloud Computing, Confusing Outsourcing Information, IT Outsourcing / IT Services
I think that the discussion on the Private Cloud is an interesting one. The purist’s argument is absolutely valid in recommending that the whole value underlying cloud computing is its ability to pool the compute capacity across the internet into a compute-workload agnostic resource, akin to the electricity from the utility company we draw from the electric sockets in our homes and businesses. The concept is to “plug-into” this compute capacity pool – a.k.a., the cloud, for performing computing tasks, that are no longer tethered to or limited by the computing capacity available within the enterprise.
However, for a large public enterprise with global operations, that is under constant threat of attacks on its compute infrastructure, simply throwing its operational, employee, customer, sales and numerous other sensitive data into the cloud is not the most palatable approach – atleast not for now! C-suite executives of the enterprise are answerable to their shareowners and customers on how they secure the electronic perimeter of the enterprise and protect the privacy of the sensitive data within the enterprise.
Multiple successful attacks on Sony’s network and more recently, the hacking of Citibank and HP’s networks will probably continue to drive the large enterprise away from adopting the public cloud in the near term.
However, it is not all bad news. A large enterprise constitutes of a conglomeration of computing infrastructure that was built over time using engineering policies driven to build the computing capacity of the infrastructure, based largely on the peak capacity requirements of applications within the application portfolio. The impact of this policy is reflected in the average server utilization metric, which stands at a low 15-20% across the industry. In other words, the private, public and government enterprises around the world are bearing the procurement costs associated with the computing assets, the power and cooling expenses and the real estate and facilities expenses for a whopping 80-85% of their compute capacity that is rarely used during the life of the infrastructure.
Virtualization technologies have been adopted by enterprises across all segments, leading to better utilization metrics. However, according to metrics gathered across the industry, there is only about a 10% “penetration” of virtualization across the install base.
Bringing this discussion within the boundaries of an individual enterprise, a similar picture can be painted. In today’s large enterprises, compute infrastructure is broken down between the Corporate and business unit IT organizations. This leads to “compute-capacity-silos” across the enterprise, that contributes to the low capacity utilization metric referred-to above.
With the advent of cloud computing technologies, Corporate IT finally has the ability to build out compute-workload agnostic environments that can be realistically utilized across the business unit silos within the enterprise.
This type of capacity pooling has the potential to drive up utilization of the computing capacity within the corporation to order-of-magnitude levels higher in the short term and make the OPEX dollar go further. However, the real value is in the achievement of longer term savings associated with the ability to postpone procurement of computing equipment until utilization levels rise up to 80-90%. Future procurements of infrastructure to boost the capacity of the “private cloud” may continue to remain low based on the ability to pool capacity across the enterprise.
Net net, aggressively architecting and deploying an enterprise-wide private Cloud Computing infrastructure can drive significant savings in the enterprises’ IT operating expenses and can be viewed as Act 1 of 2. Transition or “bursting out” from the private to the public cloud, can be envisioned as the second act for the large enterprise during the next 2-4 years. This 2 Phase transition may also enable the security infrastructure underlying the public cloud infrastructure platforms to mature enough to allow the large enterprise to seriously consider using it.
[…] a single provider or buyer could answer Esteban’s warcry to prove they had a “Private Cloud”. Private Cloud makes my blood boil. Private Clouds are a […]