2
$\begingroup$

With the advent of cloud computing, data storage and scaling has become extremely cheap when compared to the strictly physical server alternative (which require a dedicated facility and substantial fixed/variable costs). Implicitly, we could view cloud computing as having comparatively big ecological advantages over dedicated servers. From this perspective, companies with server/data center needs will have huge incentives to choose cloud computing.

My question is, is the verdict really this simple? Wouldn't some companies be reluctant to use cloud computing for security / conflict of interest reasons? Instead, might they prefer to be the sole custodians of their data? For example, investment banks, hedge funds, ect?

Are there any other compelling reasons that might hold the server / data center industry away from adopting the cloud?

| improve this question | |
$\endgroup$
  • 1
    $\begingroup$ Why are you sure that cloud computing has "comparatively big ecological advantages over dedicated servers"? With cloud computing, you still need a server to store your data, it is just not your server. So you pay somebody to buy that server and a "dedicated facility" instead of buying them yourself. Without additional considerations, the opportunity cost of "living without cloud" is zero. IMO, a clear formulation of technological advantages is a key to answering your question. $\endgroup$ – beroal Apr 20 '17 at 17:49
  • $\begingroup$ @beroal I didn't elaborate on that point too much, but I did find a nice graph that shows what I mean from a Forbes article. You can see it here. Basically, it shows how much we could save if we used servers as communal resources and dynamically utilize them moment to moment. $\endgroup$ – Arash Howaida Apr 21 '17 at 6:44
2
$\begingroup$

Institutions like banks for many years now keep back-up archives of their data in physical locations owned and operated by other legal entities. This is a standard case of outsourcing to specialists a requirement of a business process that is not a directly value-added activity. So the "sole custodian of one's own data" has been proven in the market not to be strong enough against cost-reduction considerations (or against price/value optimization if you wish).

Cloud computing is comparatively still new. The cost incentives to adopt it are indeed very strong. In practice, I believe we will observe, as in the past with other innovations, a certain degree of selectiveness: as the risks and rewards of cloud-computing will become more well understood (and measured), companies may select certain parts of their activities to keep off the-cloud, while enjoying the cost reductions for all the rest.

This by the way, can be squarely put in the framework of the, dear to us economists, marginal optimization - it is not a "yes-or-no" decision but a process where each company will adopt the Cloud up to the degree where marginal benefit equals marginal cost.

| improve this answer | |
$\endgroup$
  • $\begingroup$ @AlecosPapadopoulous I take your point that if there were no obligation, to the legal entities, they might not take it upon themselves. I'm still curious what activities a company would keep off the cloud, if you have any thoughtful speculation on that. $\endgroup$ – Arash Howaida Apr 19 '17 at 23:31
  • $\begingroup$ @ArashHowaida In today's digitized word, there still exist "things" that are considered so "sensitive", they can be found only in hard copy, after being typed in a old-fashioned typewriter. Make the analogy between dedicated servers and the cloud. Which "things", depends on the organization: what is the inside information that will incur the biggest costs if it gets public, or stolen, for each entity? For some may be trade secrets, formulas and blueprints. For others a very big and important contract and its provisions. Yet for others, their transactions... it depends $\endgroup$ – Alecos Papadopoulos Apr 20 '17 at 0:44
  • $\begingroup$ Yes, trade secrets and other sensitive information will conceivably stay on site. Another opportunity cost that just occurred to me is the issue of latency. Going back to investment banks, if they have HFT operations, then adopting the cloud will not help them win the race to zero. Perhaps some kind of private cloud could lower cost alternatives in the marginal utility solution space. $\endgroup$ – Arash Howaida Apr 20 '17 at 8:34
  • $\begingroup$ @ArashHowaida Good point. Good marginalist thinking. $\endgroup$ – Alecos Papadopoulos Apr 20 '17 at 8:47

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.