I am always interested at how we often fail to see ‘shades of grey’ and instead tend to view choices as being binary.
I have spent the past several years arguing that we are going to source the bulk of our IT resources from remote datacenters that are managed by third parties and that browsers will be the interfaces that we use for these activities. I still believe this to be the case except that increasingly, we will access these resources by using apps on our lighter and more portable devices.
Nevertheless, many still view cloud computing as a binary decision. Either adopt cloud computing across the board and expose yourself to all the risks associated with it, or you do not adopt it and continue to invest in on-premise resources. Clearly, this is not and never will be the case.
Cloud computing, in many ways, is making IT more complex. It is adding yet another dimension to heterogeneous IT environments. Corporate IT environments typically consist of a mix of mainframe technology from the 1960s and 1970s, client/server technology and cloud computing technology. This will continue. Cloud computing technology will increasingly be deployed in the form of private clouds and public clouds. It will soon be the dominant IT environment as client/server technology gets displaced.
However, client/server and mainframe technology will not disappear. It will instead account for a smaller proportion of IT infrastructures. In the meantime, large organizations will focus on integrating cloud computing with their existing infrastructures and this will be a major focus over the next few years. Already, systems integrators are massively benefiting from the huge opportunity that this additional complexity brings to them.
There will be a steady move to adopting IT resources, where appropriate from the public cloud. This is where the real disruption will occur. Why invest in on-site technology whether it is so called private cloud technology or other legacy technology, when you can access the necessary resources over the public Internet? In other words, why not source as much as possible from the cloud?
Some argue that there is a move back to processing on local devices. It is true that many apps carry out processing on local devices, but the bulk of processing and IT activity will occur remotely. Apps will offer some customization and, in my view, are a way of making the cloud work more effectively. They offer more personalized ‘windows to the cloud’ than is the case with traditional browsers. Indeed, browsers will be used much less often as a way of accessing the cloud.
In summary, more and more resources will be sourced from the public cloud. This is the real impact of cloud computing. Public cloud resources will need to be integrated with on-site investments which are made up of legacy mainframe and client/server systems as well as more recent datacenter or private cloud investments. By 2020, on-site IT activity will account for a very small proportion of an organization’s overall IT activity. In other words, the public cloud will be the dominant model of computing very soon.
I guess the bigger question would be what will it be called in 2020? If there is truth to IT spend being less, meaning that they are spending for lesser priced services than in-house, then buyers had better start asking the real serious questions about cloud services (dependability, security, scalability, adaption, recovery, contingency, service delivery levels, technology pacing, etc.). If a firm assurance can be reached on these facets it won't make much difference it's a cloud or 10,000 wombats manipulating calculators.
ReplyDeleteHi Jerry
ReplyDeleteThanks for the comment. I am sure that the term 'public cloud' will be extinct by 2020. But, I expect that many of the serious questions that you mention will be overcome. I expect that we will see most IT resources sourced from remote datacenters using the public Internet. Today, we call this the 'public cloud'.