Myth as myth
One of those mythical statements is; “the public cloud is the most inexpensive way to procure IT services.”
Companies who promulgate this style of myth often do so for their own ends and begin by freely conceding that, yes indeed, a characteristic of the public cloud is a relatively inexpensive ‘pay-as-you-use’ model, before proceeding to ‘bust the myth’ – on their terms.
Would-be myth busters
Initially, setting the scene, these ‘axe grinders’ will indicate that the starting price for basic, on-demand instances within Amazon’s EC2, for example, is less than 10¢ per hour, based on metrics like system size, operating system, etc. Such companies usually conclude their introduction by suggesting that it’s easy to see why people think all delivery from the public cloud is cheaper than that delivered by internal IT.
It’s no surprise that the punch line comes next, when a ‘myth buster’ will typically state, with a leading ‘However’, that if you probe further, the picture will change. The changed picture presented is painted thus: for resources that are needed constantly, the private cloud is more cost-efficient than the pay-as-you-use public cloud model.
A common analogy
What usually follows is a pithy example by way of analogy, such as the comparison between renting and buying a car. The argument goes that for short-term use, a car rental is cost-effective, because you only pay for what you consume. The clincher seems to be the statement that if you drive frequently and/or for longer, owning a vehicle makes better financial sense. Does it, indeed?
Let’s explore that idea.
Truth rather than myth
The myth is stated as; “the public cloud is the most inexpensive way to procure IT services.”
In reality, what you’ve just read is a misrepresentation of the situation. The claims made for cloud computing are not presented in that fashion. It is a myth that such claims are made.
The real statement should be presented as; “the public cloud is the most cost-effective way to procure IT services.” Therein lies truth and no myth.
Unsurprisingly, the myth busters chose not to present it that way, because it’s more difficult to argue against it being cost-effective. They prefer self-fulfilling prophecies in their desire to influence behaviour.
The analogy explored
To buy the analogous car, you first need a deposit, typically thirty percent of the purchase price, which is a non-trivial investment, assuming you don’t have to borrow the cash in the first place, in which case the proposition becomes even less attractive. To retain ownership of the car, you are then committed to paying the balance of the price of the vehicle, in monthly instalments, over an extended period of time, not atypically several years.
During that hire purchase contract term, you are also likely to take up a maintenance or service package, which may or may not include provision for consumables e.g., tyres, and you may even extend the warranty for an extra cost.
It isn’t that simple
At the end of the term, you own a car. Bully for you! However, the car is worth far less than you paid for it and (don’t forget this is merely an analogy) probably no longer ideal for the task for which it was first procured. So you buy a new car and if you’re lucky, the residual value will suffice for the deposit on the new one, but don’t count on it.
Now stop and think. You’re on the merry-go-round. You’re locked into buying car after car, forever – and a day.
Let’s look at the hypothetical outlay involved, substituting a PBX system for the car.
Total payment (p)
Total cost (d+p)
The key cost for comparison is the monthly payment of $1,500.00 as that’s what we have to spend on an equivalent, cloud-based IT service, in this case an IP-PBX. And don’t forget, that monthly cost is payable in perpetuity as long as you keep replacing and funding each new PBX in the same way. We can assume market price pressure rather neatly counteracts inflation, for the sake of simplicity.
The false trail
The myth busters would have you believe that you will pay a flat fee per seat per month for an IP-PBX, just as you do with certain software-as-a-service offerings. Thus, if you’ve got 250 employees at $10 per seat month, your monthly budget is exceeded to the tune of $1,000.00 and the idea that the public cloud is the most inexpensive way to procure IT services is, therefore, held up as a myth – because, in pure monetary terms, it’s plainly not the “most inexpensive” option.
But they’re barking up the wrong tree. You’ve already read that they’ve made a myth of a myth. Consider now, again, the real claim, which is, “the public cloud is the most cost-effective way to procure IT services.”
The whole truth
In fact, you can argue that the ‘per seat’ method is cost-effective, however, there is another model. If you are not paying a flat rate monthly fee and, instead, you’re paying on an ‘as-you-go’ usage basis, the picture is quite different. If you’re paying 3¢ per minute for platform time and your usage is reasonably typical e.g., you are not an outbound collections agency, your enterprise’s monthly call volume would be in the region of 12-15,000 minutes, which is less than $500 per month.
To put it another way, you can consume up to 50,000 minutes per month before you exceed the above notional budget. You may call that inexpensive if you wish, but you surely can’t argue against its being cost-effective.
Little more needs to be written. You can sum it up yourself and substitute your own figures. Whichever way you approach the issue, it’s reasonably clear that, in many typical scenarios, the public cloud can be a very cost-effective way to procure IT services.
Monthly outlay per above example
Capital purchase over term
Flat fee per seat
That is what is claimed; nothing more, nothing less – and there’s no myth about it.
Of course, there are other concerns beyond mere price, such as performance, security, compliance, service level agreements, availability, etc. However, we shall consider those another day. See you next time.