Gordon E. Moore, co-founder of Intel, wrote in 1965 his observation that over the history of computing hardware, the number of transistors on integrated circuits doubles approximately every two years. His prediction has proven to be accurate, in part because the law is now used in the semiconductor industry to guide long-term planning and to set targets for research and development. But in the age of the Internet and the Cloud, I sometimes wonder whether his statements still hold true. Take the case of cloud computing, would you believe that its roots can be traced to an idea of an "intergalactic computer network" thought of in the sixties by J.C.R. Licklider? Licklider was responsible for enabling the development of ARPANET (Advanced Research Projects Agency Network) in 1969. Over the years countless events have contributed to the development of the cloud into what we know of today. Are we there yet? I doubt it.
Today, we hear of anecdotal implementations of private cloud computing in response to the security problems associated with public cloud. What we are hearing as the next stage of the evolution is hybrid cloud computing. Why hybrid? Because many organizations, either by virtue of industry regulations or by their own insecurities, would still prefer to have parts of their computing requirements holed up at their secure-as-Fort Knox facilities or even in the public cloud for the least critical applications.
Some will argue that hybrid cloud today is where private cloud was three years ago – something to aspire and just that – the Holy Grail if you want to be Arthurian about it – at least until the next idea pops up in the general public.
Gartner defines a hybrid cloud computing service as a cloud service that spans both private and public cloud implementations or both on-premises private and off-premises private or public cloud implementations. A hybrid cloud service can, among other things, synchronize and/or replicate data between public and private clouds, and migrate services on a continuous basis between public and private clouds.
Milind Govekar, managing vice president at Gartner, wrote on his blog the benefits and challenges of hybrid cloud computing and examined its critical success factors. In his view, hybrid cloud computing extends the benefits that can be gained from cloud computing:
Organisations that have deployed private clouds will likely be the first to get to the hybrid cloud because they already have a foot in the door – with virtualization and private cloud experience and infrastructure in place. Perfecting private clouds remain a challenge not only with issues like the escalating war on security, data privacy and government or industry regulatory compliance, but the associated costs of building, deploying and managing a rapidly expanding cloud will always require businesses and IT to justify the investments in people, processes and technology.
The complicating factor today is the involvement of business units in the buying of technology – analysts estimate that this non-tech influence is growing. Without the oversight from IT leadership, who becomes accountable for the security and compliance of the company as a whole? [point your finger to the CIO].
The CIO and indeed the IT organization is shifting to a brokerage of IT services – some of it in-house and some of the cloud – the mix depends on the organisation’s vision and willingness to take a leap of faith. By adopting a formal internal cloud service brokerage role with responsibility for governance, demand management and delivery, IT can take back some semblance of control and accountability – well, no matter what happens they will be accountable. Let’s face it (and you can blame it on the tablet evolution exacerbated by the Apple iPad), business managers will circumvent IT in order to access the IT services they need. The result is a further disaggregation of IT services and reduced value from the remaining shared IT services.
But just so we are clear here – there is interest in hybrid clouds. A 2012 survey InformationWeek survey makes it clear (see chart below).
I don’t mean to rant even more about this because it’s hard to accurately predict where we are headed because a software-defined revolution is upon us. Moore’s Law needs to be recalibrated to a software-dominated world – hardware is now seconded to software.
Watch the video below for some basic understanding of the different kind of clouds.