The Industrialization Cycle Explained

Introduction

Fifteen years ago, some of the largest internet companies (Facebook, Google, Amazon...) realized they could not afford to purchase and run their IT infrastructure the same way as traditional enterprises. Margins were razor thin and the infrastructure scale and investment required was too large. The extreme economics forced a different approach on how to design, supply and manage massive digital infrastructure operations. In 2006, Bill Baker explained the mindset shift by coining the phrase "cattle, not pets". Previous IT approaches treated all servers as individually valuable (pets). Scale operations requires all servers to be managed as a herd rather than a number of individuals (cattle).

Pets:?Servers get unique names and special treatment – when they are ‘sick’ they are carefully nursed back to health, often with a substantial financial investment and allocated resource. There is an undoubted preference for keeping the server as it is, or at least in the healthiest version.

Cattle:?Servers are part of an identical group and numbered, not named.?They receive no special treatment and when they go wrong, they are replaced rather than repaired in place. They are inherently a commodity and treated as such.Cloud is the . The industrialization cycle explains the process required to continuously increase output and efficiency, whilst eliminating cost and delays in any business of any industry. For reference and to learn more, read here.

From pets to cattle became the best know analogy for what was happening in the IT industry to achieve massive scale with minimized cost. However there is a more generic understanding of the process involved in achieving maximum output and minimum waste in any industry. This is the Industrialization Cycle and this is what we will explain here. This model is attributed back to Jason Hoffman , CTO and Founder of Joyent, the original public cloud company that started before AWS and was the original provider for Facebook and Twitter (Joyent was acquired by Samsung).

Cloud is the application of the industrialization cycle to the IT industry. And the application of the industrialization cycle to the telecom industry is what is coming next. I write this article for the benefit of people in telecom that might not be aware of what has happened in the IT industry because of cloud. We may see the outcomes without understanding the underlying principles. If we wish to understand the winning operational strategy for telecom we need to understand the Industrialization Cycle.

What is the Industrialization Cycle?

The industrialization cycle is a cyclical process that when applied, leads to continuous improvement in terms of efficiencies gained and outputs produced versus inputs required. The steps are as follows.

Standardize

Minimize the amount of variability in the business. Standardize as much as possible around one practice, be it facility, hardware, software, operations and business.

Combine

Combine as much as possible so the number of entities and operations under management reduces. Continuously consolidate at the unit level, replacing existing components with higher capacity components as technology and capabilities, leading to the highest possible levels of occupation, utilization and density.

Abstract

Create a normalization layer above all entities that allows common management practice be applied universally rather than managing exceptions. In the technology world, make sure this abstraction is remotely accessible through common Application Programming Interfaces.

Automate

Automate everything, using the previous abstraction layer. If there are exceptions to the abstraction, then these have to be handled as specialist automation exceptions at best and manually at worse. No interaction requires a human, lowering OPEX costs and increasing agility.

Govern

Governance ensures that the implemented cycle is achieving the needed outcomes. There are six dimensions of needed governance - performance, scalability, quality, economics, compliance and security. Without adequate governance the machine is running the business and moving faster than the human management.

Standardize...

Once the cycle is complete it starts again. Every industry continues to evolve in terms of sophistication and improvement and each cycle builds and improves on the previous cycles outcomes. The only way to compete against the industrialization cycle for any mass market business, is to execute the cycle better and faster than the competitors.

The Industrialization Cycle Applied to Dairy Farming

To explain the model I have applied the industrialization cycle to dairy farming.

If I have a pet cow then I give my pet cow a name.? Let's call her Daisy.? I keep Daisy in my backyard, I feed her and I milk her and she produces enough milk for my family to live on.

If I choose to start to industrialize my milk production, my relationship with Daisy changes. First we are going to decide that we are going to produce milk from cows. This is a choice.? All animals produce milk but we are deciding to standardize our milk production using cows.??

We take Daisy out of the backyard and place her in a field full of other cows.? The effect of this is that we no longer have to invest in individual infrastructure to support the life of one cow, Daisy. ? The field has one fence, the field has common ground with common grass,? with many cows sharing. There is one trough for food, there's one food supply, there's one water supply so here we start to combine the supporting infrastructure costs to reduce the cost per cow.? There is then the opportunity for potential consolidation. There might be a breed of cow that produces twice as much milk as Daisy so maybe we should replace Daisy with the super cow and have less cows for the same milk result.?

The third industrialization concept is abstraction.? If we now have? thousands of cows then how do we manage those cows as a herd rather than individuals. We start by removing specialized names.? Daisy no longer exists, Daisy has become cow XY1238. Daisy has common characteristics with all other similar but uniquely labeled cows.? They all have temperatures, they all should move a certain amount a day, eat a certain amount of food.? We start to abstract the concept of a cow into what we define as being important in a cow.? We create an instrumented perspective of a cow.? This allows us to start understanding a programming the herd of cows.? When a bell chimes all cows move to the milking facility so they can all automatically be milked,

This moves us to the next phase, one of automation rather than human management.? No human ever touches a cow anymore,? the cows are managed as a herd.? There are a thousand cows and they all move through the milking facility and they are all automatically milked to produce the required volumes.? This is possible because all cows are now programmable because of the abstraction layer so all cows can be treated the same.?

We are now highly efficient but how do we know the cows are producing the volumes of milk wanted and the cows are healthy and producing healthy milk? ? We must apply the six aspects of governance.?

1. Performance - Are the cows producing the amount of milk that we believe the cows should produce?
2. Scalability - Is the herd producing the amount of milk we can actually sell in the market?? Are we producing too much or too little??Can we increase and decrease as we need?
3. Quality - Is the quality of the milk what we expect?
4. Economics - Are we meeting our economic targets for milk production versus herd cost?
5. Compliance - Are we compliant to health regulations so we can prove that our milk is actually safe to sell?
6. Security - Is the herd secure so we can ensure nobody can come in and introduce a virus to cause damage in the milk and cause damage to the people we supply?

This cycle is never ending - standardize, combine/consolidate, abstract, automate, govern. By the end of one rotation of the cycle, new alternatives will exist to further optimize and improve operation. The cycle starts again.