At a recent event where the topic of cloud management was being discussed, I was able to ask a panel of experts for their thoughts on where cloud computing, both public and private, might be heading.
I proffered the idea that in the future, all but the largest of businesses would likely go for public cloud services and that private cloud would become the preserve of very large organisations that had the skills, the budget and the need to keep direct control of infrastructure and systems.
One voice from a major technology vendor in the sector reported they were already seeing a divergence whereby smaller companies were going toward the use of public cloud services by way of hybrid cloud initially. So once private cloud had been implemented, bringing all IT together into a unified resource that allowed IT to be delivered as a service, that infrastructure could be, via APIs, allowed to flex or burst into public cloud infrastructure.
This divergence is being prompted by a greater understanding of what cloud computing means, and can do, both internally and externally, and is being strengthened by the experience of more and more organisations going to the cloud and having satisfactory experiences and fears allayed. As usage grows, so does acceptance and one can see it becoming the norm, with more organisations willing to give up their own infrastructure internally.
The larger companies, with the skills and resources to maintain the systems and infrastructure themselves, are still opting for total control, albeit in the form of private cloud, but also with a view to occasional public cloud usage, again with hybrid being the preferred option.
“The experts agreed that if you had a specific need around ensuring data remained in a specific geography, then the best option was likely to be one of the less than web-scale providers.”
The analogy that is often used in these cases is the power generation industry and how when power generation became cheap and reliable on the mass scale, businesses stopped generating their own power in favour of utility consumption. A national grid developed to distribute the power and what emerged was the situation that persists to this day.
Looking at IT as a service, there appear to be even more parallels. Today, the likes of data centres, hospitals and other critical services still have their own power generation capabilities, in other words, specialist needs mainly. I worked for a small ISP and hosting service that had its own diesel generator for emergencies. But outside of such instances, most businesses seem happy to go with a service provider for their main needs, and use their own generation capability as a back-up. In IT terms, the same thing seems to be happening insofar as when the services became cheap, reliable and of a certain uniform quality, businesses seemed happy to take them up and reduce the implementation and maintenance headaches. Some chose to have their own infrastructure as a sort of hot back-up option, with the main service coming from the service provider. However, even this is expected to wane as the quality of services offered combined with increasing trust and confidence will likely see the internal infrastructure allowed to go end of life without replacement.
There are other developments too, that the experts seem to think were not far off. One of these developments was highlighted when the experts were asked about data residency, and how the largest service providers still seemed a little lacking in accommodation. The experts agreed that if you had a specific need around ensuring data remained in a specific geography, then the best option was likely to be one of the less than web-scale providers.
Subscribers 0
Fans 0
Followers 0
Followers