Cloud computing: What is it?

Every so often a new buzzword hits the industry. Occasionally, the concept is pretty good. Cloud computing is one of those ideas whose time has come -- depending of course upon how it's defined.

What is cloud computing?

For industry old-timers who remember, cloud computing is 2010s version of time-sharing.

For those of you who are not a technological relic of a bygone era, cloud computing is a systems service architecture in which computer resource is provided to you, and you pay for that service, on demand. This model works particularly well with computer processing capability, and within the right parameters, data storage. Cloud computing is particularly useful when your storage and processing requirements change from order to order, or month-to-month.

When cloud computing makes sense.

Generally, cloud computing works better for smaller companies with limited economic resources then for large companies which can buy their own computer resources in bulk.

Cloud computing makes a lot of sense were there are periodic needs four higher than normal computer resource utilization. A very simple example is financial reporting. in financial reporting there were monthly, quarterly and annual reporting cycles. There are probably just as several days during each of these three cycles in which processing utilization and data storage requirements will be particularly high. Let's say that in each time frame there is a three day window in which computer utilization outstrips the average need a large amount. This means that you have extreme computer utilization requirements for 51 days a year. Historically, the company would have purchased enough computing power to handle peak periods. However, with cloud computing, you can pay for that additional resource only when needed. This can be a significant financial benefit to the company.

Ask your self... are there particular Windows in which computer utilization, because of the applications being run, significantly outstrips the balance of the year. If the answer to that question is yes, your application is a candidate for cloud computing.

Cloud computing -- marginal situations.

Data storage and computer processing requirements typically grow linearly, but are purchased in step functions. Cloud computing can be a way to handle linear growth. When your data and/or processing requirements do not yet scale to the size of an economical hardware purchase, cloud computing can be a much more economical way to handle growth. Only when the cloud computing resources become financially balanced with buying the next increment of hardware does the purchase make sense. In this way, incremental utilization due to growth can be made much more economically palatable. You don't have a significant amount of unused resource, for which you've paid, using this technique.

When cloud computing doesn't work.

Cloud computing may not be the most logical alternative for growing organizations were computing resource is growing at a steady rate. Cloud computing can be, but is not designed to be, a permanent solution for increasing data storage and processing requirements. If you need a terrible eight of data online, and it's not likely to do anything but grow, you're probably better off buying or leasing the terabyte rather than using cloud computing.

The magical number ...

George S. Miller wrote a paper in 1956 called "The magical number 7 ± 2: Some limits on our capacity to process information." I first came across this paper in the mid-1980s when I was the vice president of education services running the technical training division of a software company.

My associate, Vaughn Frick, and I devised a couple of experiments that we actually repeated many times over the course of a two year period. The experimental results not only confirmed Miller's original hypothesis, but allowed us to push the envelope on the concept even further.

The instigation.

Miller's original report resulted in the now famous American seven digit phone number preceded by a three digit area code. You can even repeat the original experiment yourself. Write down a series of numbers from three digits to 10 digits. it doesn't matter what the numbers are.

You can even use the list below, unless your friends are likely to read this blog and memorize the results! (Wishful thinking.)

you will inevitably find that when the string of digits reaches 10, there was a sharp drop-off in memory. You will find also that there is a much smaller drop-off between five and nine. The difference however in drop-off between the 5/9 range and the 10 and over rate is greater by a factor of five or more.

What this illustrates.

the human brain has a unique, but limited capacity for short-term storage of information. You can call information and remember up to three telephone numbers. Why? Because the telephone number itself is broken down into three smaller groups of numbers: the area code (three digits), the exchange (three digits) and the number (four digits). Some old folks like myself remember the vague and easiness when they first added new area codes. I tracked down that uneasiness to the pattern of numbers used. Originally, the area code and a zero or one as the middle digit. The exchange never had a zero or one.

What does this mean?

it means, basically, that we have an extraordinary capacity to remember information. Think about it. We can remember seven groups of things. Each group of things can have seven elements. Each of the seven elements can be up to seven digits long!

I had a stroke at the end of 2008. It took nearly a year to recover. Even at that, my recovery was not complete -- not back to my pre-stroke self. This knowledge of how the brain uses Miller's magic number in able to me to vastly improve my post stroke of memory.

example data

number list 1

847

2195

64723

539261

3984271

75143268

139624857

4382769150

82734061529

number list 2

203 . 561 . 8947

How to explain anything to anyone: Braess's Paradox, the volvox colony, and a dragon eats his tail

Sometime in the 1970s I saw a movie about a spy, very much like James Bond. I just can't remember his name. The movie was pivotal in my success in communicating big business and technological concepts to average people. This technique is, perhaps, something they can benefit you.

At one point in the movie, the big, evil CEO of the company gets up to give a speech. What shows on the projector is a Dragon eating his own tale. Evil CEO starts his speech, not by talking about a new business topic, but about the dragon eating his tail -- and why. After the tail eating story, the evil CEO starts talking about the business, and tying the business points, point by point, to the story of the tail eating dragon. Brilliant. So brilliant, it's the only thing out of the movie that I can remember with any clarity. The day, I adopted this technique for many of my public speaking engagements.

The "bucket of green slime" speech.

For several years during the early 1980s, I gave what became known as the "Green Slime Speech." At the beginning of the speech, I told the story of volvox colony. Volvox are unicellular microorganisms that have a nasty habit of turning water into a brownish green sludge. However, given the right provocation, enough time, and the right temperature, the brown green sludge condenses itself into a multicellular colony of creatures in which the individual one celled animals take on particular roles based on where they end up living in a colony. The water turns crystal clear.

See! What could a bucket of green brown sledge possibly have to do with high-tech anything, particularly computer software? Nothing actually. It was an analogy. Most computer projects, in that day and age, were late! Late projects are like a bucket of green slime. They are messy, and generally left a bad taste in your mouth, just like the green slime. However, given some time, organization, and management principles, projects could be organized and controlled, much like the orderd volvox colony that ultimately congealed, with the right provocation, and left the water clear.

The loop around the city.

One day, a city road engineer got annoyed at taking this outer loop highway around the city. Wouldn't it be faster, for a least a third of the highway users if they just put a new North/South arterial right through the middle of the city? Sounds good.

Whoops.

Traffic bogged down on the other loop -- writing at the junction with the north-south arterial! Now, traffic was, in general, slower everywhere. A paradox! An unintended consequence. Something which mere mortal logic could not foresee. How many times in our daily business lives do we think up a quick answer to a problem?

One third of the time, that brilliant idea actually improves things.

One third of the time, that same brilliant idea does nothing with respect to productivity.

One third of the time, the idea actually makes the situation noticeably worse!

This story actually drew upon an analogy called Braess's Paradox which explains why extra transaction handling circuitry placed inside a computer chip, at random, actually slows the computer chip down. Very esoteric. The Road's story is actually easier to understand!

However, Braess's Paradox and the story of the highway loop may a great analogy for why two thirds of our off the cuff business decisions have, at best, no impact on the business. It also shows what type of analysis must be done in order to make sound, rational decisions. A great analogy.

The power of analogy.

You will see analogies in my writing from time to time. They are there to make you think, to look at a problem from a new angle, or just to shake the brain awake. Think of ways to use analogies in your explanations. Most importantly, analogies make discussions and explanations memorable. 20 years later, somebody still reminds me of that "green slime speech."

What was I thinking?

Some of you know me. Most of you probably do not. So, for the benefit of the "great body of uninitiated," who have never had the benefit of my prognostications and observations, I'll give you a little background here.

Background.

I've been in a number of industries. I've spent most of my career in some type of technology business. I've done the usual techie stuff: programmer, analyst, VP of IT, etc. Then, I got tired of technology. I was tired of being a geek!

I had spent a lot of time, when my children were young and I was poor, hanging round universities as an adjunct professor. I decided to make some money and capitalize on those skills with commercial education. I was VP of consulting and/or education for a couple of firms.

Then, in October of 1989, I joined an amazing company -- GartnerGroup.

(Aside: One day, midway through my career there, the founder, Gideon Gartner had left and some bright eyed marketing executives decided the Gartner Group sounded pretentious and renamed the company Gartner, Inc. given that today, more than 10 years later, people still call it the Gartner Group, it was a huge and foolish waste of money.)

I spent the next 11 years there working my way from analyst to division president. Eventually, all my old friends and business associates left, including my "patron saint"at the time, the firm's third ever CEO, Michael Fleischer. It was time to move on. I just had no idea what to do with my life.

An old and dear friend.

Long before Gartner Group, I befriended Dr. N. Adam Rin (... and yes, I will figure out what the and stands for one of these days!).

Adam and I probably knew each other for 10 years or so before Gardner. The truth of the matter is, one of Gartner's VP's, Peter Levine, tried to hire me in 1988, the year before I joined. I didn't. That was not one of my better decisions, in retrospect. But anyway, I had known Peter for a while, and he asked me if I knew Adam. I told him of course I did. Peter asked me what I thought. I told him Adam was a great guy and a consummate professional. Peter hired Adam in 1988. I caved in to Peter's pressure, in a rare fit of sanity, and joined Adam there in 1989.

Adam climbed the ladder faster than I did, because he was willing to play their game long before I was. As he became division president of Gartner | Research before leaving to join them CEO, Manny Fernandez, full-time at Gartner's venture company. While Adam was nearing the end of his tenure as division president of Gartner | Research, I became division president of Gartner | eMetrix (that was back in the day than if you were cool, you had an E in your name).

Off on our own.

Knowing that Michael was leaving, and having, for the fourth time, a wad of cash waived before me big enough to choke a horse (or maybe an elephant), I bid adieu to Gartner.

I pondered my future, and did some consulting, and caught up with Adam a couple months later. We talked. He left Manny's investment firm. We've been partners, ever since! our biggest successes are Training JumpStart Ltd., which we started with a third partner, Dr. Hadas Rin, and InfoTollgate Ltd., both of which are alive and well. Visit them on the web!