Samos is an island in the eastern Aegean Sea, where Pythagoras is thought to have been born in 569 B.C. SAMOS© is also my acronym for Simple– Automatic – Modular – Organic —Synchronous. For thirty years, it has been the guiding principle that helps me determine if a new technology is truly innovative and could disrupt the market place, or just a waste of my time.
Why is simple so important? “Because the unpredictable happens,” explains Tom Kolditz ) of West Point. “No plan survives contact with the enemy. No amount of planning, no sophisticated algorithm, can prepare you for every circumstance.” The Army relies on something called the Commander’s Intent (CI). A commander knows that if he tries to spell out all the details of a plan it will fail. So instead, he tries to convey his intent. CI is SIMPLE = core + compact.
Using the SAMOS principle, a system must be simple. Technology must be able to execute the ‘intent’ of business. Like, “we’ve got more customers, we need more storage.” One of the unintended consequences of distributed computing and Service Oriented Architecture (SOA) is that things get more complicated. Complicated is not good. When things get complicated, bottlenecks develop, and systems get more difficult to maintain.
Nutanix is a great example of simple. It’s important not to conflate “simple” with “simplistic.” Nothing about Nutanix is simplistic. Like many solutions, Nutanix converges “storage,” “compute” & virtualization. Nutanix is “hyper” converged. Nutanix has converged just about everything in the data center. In addition to storage, compute, and virtualization, they converged, DNS, DHCP and their own hypervisor. However, Nutanix did not stop there. They converged traditional symmetrical computing (SMP) with parallel computing. Nutanix employs Map/Reduce, Cassandra, Paxos, and Zookeeper to create a “distributed everything” architecture.
With Nutanix, each node is very sophisticated. You can start with just three nodes, and then keep adding more nodes one by one, just like Lego—that’s simple.
When you use modular units, a larger system is built from smaller units. In good designs, modules are reusable and systems scale easily. In the best designs, modules are object-oriented and systems can be complex, but remain simple. Many technologies other than IT use modular designs. Office cubicles are modular. They use smaller units to build larger systems. By contrast, computers are not modular. The typical computer has seventeen distinct components and none of them is modular.
In 1790, gunsmith Honoré Blanc produced a thousand muskets with interchangeable parts. In 1798, Eli Whitney was awarded a contract to sell the US Government 10,000 muskets, but by 1801, he had failed to deliver any. When he finally did deliver the muskets, they were NOT interchangeable. Eli Whitney was credited with the “idea” of interchangeable parts, but it is documented that he faked his presentation to Congress!
While U.S. manufacturing became world famous, the industrial revolution failed to produce interchangeable parts. In the first industrial revolution, things were still made by hand and not very modular. Sewing machine needles were ubiquitous, but bobbins could still only be used in specific models. It wasn’t until the standardization of machine tools and dies that the US was finally able to produce interchangeable parts. It wasn’t until the introduction of the assembly line that modular design began to fuel the amazing growth of the 20th century. All modular parts are interchangeable, but not all interchangeable parts are modular. Gun parts today are finally interchangeable. Yet the rest of the manufacturing world still hasn’t caught up. I still can’t put the lid from my peanut butter on my jelly jar.
When a product is both modular and interchangeable, it achieves ‘technology transparency’ and is then inducted into the modular hall of fame. Lego is the perfect example of both modularity and interchangeability.
Lego bricks are modular units that can be assembled in countless ways, but more importantly, their interlocking system allows them to be taken back apart. Danish mathematician Søren Eilers calculated that there were 915,103,765, different combinations that could be built just from the original six bricks. It was in 1958 that Lego patented the ‘Automatic binding brick’, and yet today, modern Lego bricks are still compatible with those original bricks. This is modular design at its best.
Lego bricks embody the SAMOS principle of modularity because they allow you to create virtually anything. Det bedste er ikke for godt is the Lego motto. It means, “The best is never too good.”
Imagine if IT were that modular!
The original Automat was a self-service cafeteria in New York City. When we refer to something as being automatic, it doesn’t mean that something will happen without being controlled by a person, but that it has a self-regulating mechanism. The patented name for a Lego brick is the “Automatic Binding Brick.” By simply applying a little pressure, blocks automatically join to one another. Just like automatic windows in a car, a human still has to push the button for window to go up. The person doesn’t have to manually wind or crank it up. In the best designs, modular, interchangeable parts are automatically assembled into larger units. Automatic is a key element of SAMOS. However, automation by itself is not always a good thing. Automation can make bad processes go wrong faster. Remember what happened to Mickey Mouse as the Sorcerer’s Apprentice.
In chemistry, organic refers to the presence of carbon—humans are an example of carbon-based life forms. In the military, organic has a different meaning. A unit or asset is organic when it is a permanent part of a larger unit. In the SAMOS principle, units are modular, interchangeable and are automatically combined into larger systems. The units are part of the whole, and the whole is made up of the parts. We can say that the system is organic.
In IT synchronous can have a dual meaning.
In digital communications, when something is synchronous, both ends of the network have a clock that is synchronized. If it is asynchronous, there is only one clock and the ends are not in perfect sync. The most famous example of this is SONET, Synchronous Optical Network. SONET rings made up the backbone of Public Switched Telephone Network and are used by long haul carriers. When your clocks are synchronized, you can run more data on the line. This is why SONET gets so much speed out of Time Division Multiplexing. Ethernet, the most ubiquitous communication medium in IT, uses an underlying technology like “Manchester encoding” which is also synchronous. Synchronizing two clocks is more expensive than using just one asynchronous clock. But if you are going to do something, it’s worth doing it right.
Programmers define synchronous and asynchronous a little differently. Imagine if your boss comes into your office and asks you to create a diagram. If he waits in your office until it’s done, we say that is synchronous. If your boss leaves and lets you email it to him when you’re done, we say that’s asynchronous. Programmers like asynchronous communication because they can start one task and before it finishes, they can start a second task. Things can happen in parallel instead of serially.
Since I was trained as an engineer, I use the networking definition. In SAMOS, to have the ultimate speed in communications, I prefer that things be synchronous. It makes things fast—very fast.
With the rapid pace of innovation and the consumerization of IT, it can be difficult to keep up with all the latest trends and technology: cloud computing, mobile computing, big data, internet of things, etc. So whenever I encounter a new technology, I use the SAMOS principle to guide me. Is the technology simple? Can it scale automatically? Does it use a modular design? Do the units become a permanent part of a larger whole—does the system grow organically? And can it be synchronous?
When presented with new technology, the question you always want to ask yourself is, “Is this going to make my business operate better?
Latest posts by Nathaniel Crocker
- Presenting Tech To Decision Makers: Be Bold, Be Brief, Be Gone…. - February 14, 2017
- Seven Trends and Predictions for 2017 - January 21, 2017
- Hyper-segmentation – How to Avoid Cyber Disasters - November 2, 2016