When we run across an absolute and unchangeable law of nature, isn’t it fascinating to ponder its splendor for a moment and let its simple truth shape our minds. Take gravity; we don’t see it or hear it, but we certainly feel it. Without it, where would we be? In a different vein, how about that unequivocal law called the “iron triangle”? What? You haven’t heard of the iron triangle? Well it’s been around for centuries and it is as true today as when first discovered.
The iron triangle is best understood with an example. Let’s say you want to build a new piece of software that will solve world hunger and most likely usher in an era of global peace. Now that you have this phenomenal idea, you can’t wait to roll it out and start selling it to nations on all 5 continents. So, to get rolling you take your specs to a great software developer and inform her that you need the production version of this app in your hands fast, it has to be cheap, and it has to be good. She looks at you and says, “fast, cheap AND good . . . sorry buster, you can have only two of those three requirements.” Welcome to the absolute world of the iron triangle.
I’m sure you know what I&rs
quo;m talking about. And, as a responsible IT professional you are probably guided in your technology decisions by this immutable law, right?
Let’s get practical. Let’s move from the world of applied philosophy to the world of ‘you know what’ happens. As we noted earlier, gravity is for real, nothing we can do to change that. On the other hand, we try to bend and shape some laws (e.g. the iron triangle) to suit our needs, goals, and dreams. To highlight a familiar example, do you remember from U.S. history the time when land was given to the citizens of our growing country if they would leave their homes back east and move west to the frontier. The pioneers were promised a tract of land the day they arrived (fast), it was free (cheap), and it was tillable (good). I think we know how things turned out for most of the early pioneers; they got two of the three points on the iron triangle – fast and cheap. But, the land was either covered with huge trees, it was open prairie that was hard as a rock, or it was on a mountainside that actually was a rock. Add a few hostile Native Americans who weren’t in any mood to give up their lands, and the deal was not quite as ‘good’ as the brochure promised.
All that said, the pioneers still went west, they worked hard, sacrificed and over the course of many years they met with some success. So the lure of ‘fast, cheap and good’ works on some levels, like to attract pioneers. In reality it takes lots of time and lots of patience to reach a stated goal. Unfortunately lots of time and lots of patience are never planned for in the beginning. The iron triangle is a demanding mistress, and she will be obeyed.
In technology we have an enormous problem with the iron triangle. How often have you heard about the next BIG thing that will change our world forever and make life worth living? Well, we have just such a big thing sitting on our doorstep right now. It’s called in-memory computing. SAP has an in-memory product called HANA, that could be part of an SAP hosting environment, Oracle has one called Exalytics. I suppose there are other companies working on their own flavor of in-memory computing too. But for now let’s look at HANA and Exalytics. Notice I did not say HANA versus Exalitics; I don’t think we know enough about either product yet to pursue that line of thought.
To help you understand what in-memory computing is, I selected an excerpt from an article out of the online publication – The Times of India, dated September 21, 2011.
Computers typically store data on the hard disk, and when you want to perform a task, it pulls out the relevant data and applications from the computer's hard disk and processes everything in the main memory, which is where computations happen.
With the emergence of multi-core processo
rs and the sharp decline in prices of processors and memory, German business software maker SAP developed a technology that made it possible for even large enterprises to dispense with hard disks and store and perform all operations on the main memory. It boosted performance enormously compared to systems based on retrieving data from hard drives.
Now back to my main point, both HANA and Exaltyics are messing with the absolute integrity of the iron triangle. So far HANA is staying away from the ‘cheap’ point on the triangle, but Exalytics is hitting all the tips on the iron triangle. Here is an excerpt from an Oracle press release:
“…customers can expect to reduce the complexity and cost of their IT infrastructure with Oracle Exalytics. Offering dramatically faster analytics performance, Oracle Exalytics enables our clients to gain deep and immediate insight across all key areas...” – Senior Practice Director, Oracle Partner
SAP is not far behind Oracle with its claims that could test the veracity of the iron triangle, see the example below:
“HANA is not an experimental thing, we got lucky and we found something and it runs things 10, 100, 1,000 times faster. HANA is a platform and is on the way to become the foundation for most of our applications, whether they’re on-premise or in the cloud.” – Hasso Plattner, Chairman SAP
Larry Ellison, not to be outdone, tried to position Oracle as the clear leader when he introduced Exalytics at Oracle’s OpenWorld:
"We're determined to deliver best-of-breed in every aspect of our computing architecture," Ellison said unveiling Exalytics. &
nbsp;"We're in the business of catching up with IBM in the microprocessor business. If we don't pass them we'll be very, very close. Overall we move data around a hundred times faster than anyone else [veiled reference to SAP] in the business.” – Larry Ellison, CEO Oracle.
In-memory computing is probably leading the way to major breakthroughs in computing speed, but how long will it take to realize the vision championed by the titans of technology? How much blood, sweat and tears will be spilled before in-memory computing lives up to the hype and hysteria that Mr. Ellison and Mr. Plattner are encouraging? Is it possible that in-memory computing is just the latest fast, cheap and good technological thingamabob to lure the adventurous and restless among us out to the frontier? Probably, but metaphorically speaking, there will be many wagon-train wrecks along the way and that is to be expected.
Who will bear the brunt of the problems and disappointment that botched in-memory computing projects leave in their wake? It will probably be the worker bee, the guy or gal who is handed this extraordinary piece of technology and told…’make it happen’. But when it doesn’t happen, or it takes too long, or it costs too much, or it doesn’t perform as expected…remember the iron triangle…it cannot be bent or shaped to suit our needs.