
Image by
kyeung808 via
Flickr
One topic that came up during the panel discussion at the Under the Radar
event was about the recently released McKinsey report. Everyone was unanimous
in highlighting the “errors” in the report’s conclusion. Many people have
written extensively debunking McKinsey’s claims and the report also had support
for its claim from pundits who make a career talking about traditional software
world. Of all the analysis, I was impressed by the model put forward by Joe Weinman. He correctly pointed out to
the role of Cloud Computing in the enterprise IT, at least, in the near
future.
If McKinsey had studied, say, automotive transportation, the conclusion they
might have come to would be that, since enterprises and “cloud service
providers” — rental car companies — use the same infrastructure
components — cars — there really aren’t any economies of scale and the service
provider premium then makes rental cars financially unattractive. In fact, the
rental car industry may have been “getting away” with an even higher premium
than the cloud computing industry; a Chevy that one can finance or own
for $10 a day might rent for four or five times that.So wouldn’t the analogous conclusion then be that enterprises should be wary
of using rental cars, until that industry comes up with a better value
proposition?
He then correctly points out how the Cloud services fit in the enterprises
today
Rental cars and other “meatspace” utility services create value for customers
by cost-effectively addressing unpredictable or variable demand with dynamically
provisioned resources on a pay-per-use basis. So do cloud services. And, as
enterprises consolidate data centers, they’d best leverage content delivery
clouds and other dispersed infrastructure services available from cloud service
providers to meet interactive application requirements.
I am thoroughly impressed with Joe’s model to describe the role of Cloud
Computing in the enterprises. On the Cloud evangelist side, let us be clear
about certain aspects of enterprise adoption. It is totally unreasonable for us
to expect the enterprises to give up all their past investments and jump into
enterprise bandwagon overnight. If anyone thinks that enterprises will
immediately embrace the Cloud Computing because of the advantages it offers,
they are fooling themselves. On the enterprise side, I do agree that there are
some who are unwilling to change the way they are doing IT and they are the ones
I compare to dinosaurs on the path to extinction. A big chunk of enterprises are
reluctant to jump in for two reasons.
- They want the Cloud Computing landscape pass over the hype cycle and mature.
We are already seeing signs of maturity and as we go by, it will only get
better. We will see more and more enterprises jumping in after this stage - Then, there are those companies with huge investments on datacenters and
related infrastructure. We cannot expect them to write off all these investments
and embrace Cloud Computing. The best they can do is to consolidate their prior
investments and start using Clouds for the non critical processes initially and,
then, slowly embrace Clouds over a period of time.
This will be the normal evolution of enterprise IT and any expectations to
accelerate the process is unreasonable. At the same time, if the critics of
Cloud Computing use this (slow) normal evolution process as an evidence to
debunk Cloud Computing, it just shows their ignorance. Enterprises will slowly
move towards Cloud Computing and while the transition happens, Joe’s rental car
model is the correct model to explain what is happening. I hope analysts take
some time to understand this before they come up with another report talking
about the irrelevance of Cloud Computing.
I like the rental car analogy too – but maybe draw different conclusions from it about Enterprise computing vs Cloud. Its clear from the analogy that short term rental offers something different from long term purchase or lease arrangements – flexibility, rapid deployment, fine grained usage based costing. And its also clear that if what you want is stable, long term, predictable then renting a car on this basis (vs purchase or long term lease) is a fairly bad idea economically. There are some down sides – turning up at the rental desk and having to queue because demand is high, or the vehicle you want is not available…
Same with Enterprise IT. Large enterprises running ERP systems need predictable , reliable capacity and can predict their demands fairly well over long time periods. For these uses, the benefits of cloud do not apply. So it is not a question of ‘replacing’ enterprise IT with cloud, or some inevitable migration to a better model, but it is more a question of complementing enterprise IT with cloud services. Just like a large company may have a fleet of leased cars as well as a contract with Hertz for short term, unpredictable needs.
Mark, I am with you on it. The rental car analogy doesn’t hold if you want to use Cloud for a longer term. That is why I have argued that the rental car model is a valid model in the near future when enterprises are getting comfortable with the idea of Cloud Computing and market forces bring down the cost of infrastructure. For me, this model explains the current state of enterprise cloud adoption.