The Unicorn Project and The Five Ideals (Updated Dec 2019)

@RealGeneKim
Session ID:
Gene Kim
The Unicorn Project
And The Five Ideals

@RealGeneKim
My Definition of DevOps
The architecture, technical practices, and cultural norms
that enable us to…
increase our ability to deliver applications and services...
quickly and safely, which enables rapid experimentation
and innovation, and the fastest delivery of value to our
customers…
while ensuring world-class security, reliability, and stability...
…so that we can win in the marketplace.

@RealGeneKim
Another Definition of DevOps
Better Value,
Sooner, Safer, Happier
Jon Smart, Partner, Enterprise Agility, Deloitte

@RealGeneKim
The Downward
Spiral

@RealGeneKim
IT Ops And Dev At War
10

@RealGeneKim
IT Operations
CBS Photo Archive/Star Trek: The Original Series/Getty Images

@RealGeneKimSource: Flickr: birdsandanchors
The Product Managers

@RealGeneKim“Hot Fuzz”: Rogue Pictures

@RealGeneKim
The Product Managers
Source: Flickr: birdsandanchors
Architects

@RealGeneKim
The Problems That Still Remain
 Absence of all the invisible structures needed to
enable developer productivity
 The orthogonal problem of getting data from
where it resides to where it needs to be used
 Strong opposition to support new ways of
working
 Ambiguity on what behaviors needed to support
during a transformation

@RealGeneKim
The Five Ideals
1. Locality and Simplicity
2. Focus, Flow, and Joy
3. Improvement of Daily Work
4. Psychological Safety
5. Customer Focus

@RealGeneKim
Session ID:
The Business Value Of DevOps
Is Even Higher Than We Thought

@RealGeneKim
Elite Low Difference
Deployment Frequency
On-demand
(multiple times per day)
Monthly or quarterly 208x
Deployment Lead Time < 1 hour 1 day to 1 week 2,555x
Deploy Success Rate 0-15% 46-60% 7x
Mean Time to Restore < 1 hour 1 week to 1 month 2,604x
Elite vs. Low Performers
Source: Google/DORA: 2019 State Of DevOps Report: https://cloud.google.com/devops/state-of-devops/

@RealGeneKim
On-demand
Deployment Lead Time < 1 hour 1 week to 1 month 106x
Mean Time to Restore < 1 hour 1 week to 1 month 2,604x

@RealGeneKim
On-demand
Deployment Lead Time < 1 hour 1 week to 1 month 106x
Mean Time to Restore < 1 hour Less than one day 2,604x

@RealGeneKim
High Performers Are More Secure And
Controlled
2x 29%
less time spent
remediating
security issues
more time spent
on new work
Source: Google/DORA: 2018 State Of DevOps Report: https://cloudplatformonline.com/2018-state-of-devops.html

@RealGeneKim
High Performers Win In The Marketplace
2x 2xmore likely to
exceed profitability,
market share &
productivity goals
more likely to achieve
organizational and
mission goals, customer
satisfaction, quantity &
quality goals

@RealGeneKim
High Performers Win In The Marketplace
2.2xhigher employee
Net Promoter Score
50%higher market
capitalization growth
over 3 years*

@RealGeneKim
The Opposite Of
Technical Debt Is…

@RealGeneKim
When we can safely, quickly,
reliably, securely achieve
all the goals, dreams and
aspirations of our business…

@RealGeneKim
Session ID:
Ideal #1:
Locality and Simplicity

@RealGeneKim
The Birth And Death Of Etsy Sprouter
 A story about teams of engineers implementing
changes
 2008: Devs and DBAs
 2009: Devs and DBAs and Sprouter team
 2010: Devs

@RealGeneKim
Lesson:
The Organization and
The Architecture Of Our Software
Must Be Congruent

@RealGeneKim
Lead Time = 9 months
Source: Damon Edwards (@damonedwards)

@RealGeneKim
Architecture Enables Teams To…
 …make large scale changes to the design of its system without the
permission of someone outside the team, or depending on other
teams
 ...complete its work without fine-grained communication and
coordination with people outside the team
 ...deploy and release its product or service on demand, independently
of other services the product or service depends upon
 ...do most of its testing on demand, without requiring an integrated
test environment
 ...perform deployments during normal business hours with negligible
downtime
Source: Puppet/DORA: 2017 State Of DevOps Report: https://puppet.com/resources/whitepaper/state-of-devops-report

@RealGeneKim
The First Ideal: A Measure
 Bus factor
 Lunch factor

@RealGeneKim
How Many People Do You Need To Feed?
 Two pizza team
 Feeding everyone in the building
 Schedule lunch with 43 different people

@RealGeneKim
The First Ideal: Code
 Ideal: anyone can implement what they need by
looking at one file or module, and make the
needed change
 Kubernetes sidecars
 Spring (http-retry, Dependency Injection)
 Aspect Oriented Programming
 Not Ideal: to make your needed change, you
have to understand and change all the files and
modules

@RealGeneKim
The First Ideal: Code
 Ideal: changes can be independently
implemented and tested, isolated from other
components (composability)
 Not Ideal: in order for changes to be
implemented and tested, the entire system must
be present (e.g., integrated test environment)

@RealGeneKim
The First Ideal: Organization
 Ideal: every team has the expertise, capability
and authority to satisfy customer needs
 Not Ideal: in order to satisfy customer needs,
every team must escalate up two levels (and over
two, and down two)

@RealGeneKimSource: Manu Cornet: Bonkersworld

@RealGeneKim
Team of Teams
 Story of Joint Special Forces
Task Force battling a smaller,
nimbler adversary in Iraq in
2004
 Pushing decision making to
the edges

@RealGeneKim
The First Ideal: Data
 Ideal: every team has access to the data they
need, on-demand, quickly, accurately, and
securely
 Not Ideal: in order to get the data they need,
teams must wait months, and hope that every
report won’t break

@RealGeneKim
Session ID:
Ideal #2:
Focus, Flow, and Joy

@RealGeneKimSource: https://itrevolution.com/love-letter-to-clojure-part-1/

@RealGeneKim
As Your Ambassador From Dev
 For decades, I self-identified as an Ops person…
 2 years ago, I’ve started to self-identify as Dev
 Clojure / ClojureScript
 LISP, functional programming, immutability
 3000 lines of Objective C -> 1500 lines of TypeScript/React -
> 500 lines of ClojureScript
 Development is so fun, and these days, you can do
miraculous things with so little effort

@RealGeneKim
Why Functional Programming
 The famous French philosopher Claude Lévi-Strauss
would say of certain tools, ‘is it good to think with?’
 Core FP concepts
 Immutability
 Pure functions
 Composability
 Pioneered by Haskell and Ocaml. Popularized by
Clojure, Erlang, Elm, Elixir, ReasonML

@RealGeneKim
Interestingly, It Portends Future Of Ops
 Core concepts
 Immutability
 Pure functions
 Composability
 Look at…
 Docker, Docker Compose
 Kubernetes
 Kubernetes sidecars
 Event streams: Apache Kafka
 Git

@RealGeneKim
The Second Ideal: Focus and Flow
 Ideal: your energy and time is focused on solving
the business problem, and you’re having fun
 Not Ideal: all your time is spent trying to solve
problems you don’t even want to solve (e.g.,
YAML files, Makefile and spaces in filenames,
bash)

@RealGeneKim
Never Have I Valued Infrastructure More
 Things I detest now
 Everything outside of my application
 Connecting to anything to anything
 Updating dependencies
 Secrets management
 Bash
 YAML
 Patching
 Building kubernetes deployment files (mostly by Googling)
 Why my cloud costs are so high

@RealGeneKim
The Value Of Platforms
 Enable developer productivity
 Self-service
 On-demand
 Immediacy and fast feedback
 Focus and flow
 Joy
 Monitoring, deployment, environment creation,
security scans, orchestration…

@RealGeneKim
There’s Never Been A Better Time
for Infrastructure and Operations

@RealGeneKim
Two Types Of Learning
● Procedural Learning
● One-shot Learning
Source: Dr. Mihaly Csikszentmihalyi

@RealGeneKim
“What is your lead time
for changes?”
“How long does it take to go from
code committed to code successfully
running in production?”

@RealGeneKimSource: The DevOps Handbook
Product Design and Development Product Delivery
(Build, Test, Deploy)
Create new products and services that solve
customer problems using hypothesis-driven
delivery, modern UX, design thinking
Enable fast flow from development to
production and reliable releases by
standardizing work, reducing variability and
batch sizes
Feature design and implementation may
require work that has never been done before
Integration, test and deployment must be
performed continuously, as quickly as possible
Estimates are highly uncertain
Cycle times should be well-known and
predictable
Outcomes are highly variable Outcomes should have low variability
Change Committed Into Version Control

@RealGeneKimSource: The DevOps Handbook
Change Committed Into Version Control
Product Design and Development Product Delivery
(Build, Test, Deploy)
Create new products and services that solve
customer problems using hypothesis-driven
delivery, modern UX, design thinking
Enable fast flow from development to
production and reliable releases by
standardizing work, reducing variability and
batch sizes
Feature design and implementation may
require work that has never been done before
Integration, test and deployment must be
performed continuously, as quickly as possible
Estimates are highly uncertain
Cycle times should be well-known and
predictable
Outcomes are highly variable Outcomes should have low variability

@RealGeneKim
 Ideal: when you can implement and test your
feature on your Dev laptop, and learn whether it
worked in seconds
 Not Ideal: when the only way you can determine
whether you feature worked is waiting minutes,
hours, or days… or weeks…

@RealGeneKim
 Ideal: trunk based development
 Not Ideal: 5 days merging, 50 people in
conference rooms

@RealGeneKim
Session ID:
Ideal #3:
Improvement Of Daily Work

@RealGeneKim
Not Ideal
“In manufacturing, the absence of effective feedback often
contribute to major quality and safety problems. In one well-
documented case at the General Motors Fremont manufacturing
plant, there were no effective procedures in place to detect
problems during the assembly process, nor were there explicit
procedures on what to do when problems were found.
“As a result, there were instances of engines being put in
backward, cars missing steering wheels or tires, and cars even
having to be towed off the assembly line because they wouldn’t
start.”
Source: DevOps Handbook

@RealGeneKim
Create as much feedback in our system, from as
many areas in our system, sooner, faster, and
cheaper, with as much clarity between cause and
effect.
Why? Because the more assumptions we can
invalidate, the more we learn, improving our ability
to fix problems and innovate.
Source: DevOps Handbook
Ideal

@RealGeneKim
How many times per day is the andon cord
pulled in a typical day at a Toyota
manufacturing plant?
3,500 times per day
Source: http://www.gembapantarei.com/2008/04/how_many_times_do_you_pull_the_andon_cord_each_day.html

@RealGeneKim
Session ID:
Greatness Isn’t Free…
The Need To Pay Down Technical
Debt

@RealGeneKim
Fast Push To Market
Debts & Risks
Features
Quality
Defects

@RealGeneKim
Fast Push To Market — Continued
Features
Defects
Defect fixing dominates work
Site reliability tanks
Slower and slower velocity
Customers leave
Morale plunges
Devs leave because everything is hard
Quality
Debts & Risks

@RealGeneKimSource: https://twitter.com/johncutlefish/status/1046169469268111361
Who hasn’t felt this?
You hire a bunch of developers, but you
still can’t ship the features you
promised…
…and maybe you even have the feeling
that things are slowing down…

@RealGeneKim
Risto Siilasma, NOKIA
Source: The Unicorn Project (2019) / Transforming NOKIA (2019)

@RealGeneKim
Near Death Experiences
● Ebay (1999)
● Microsoft (2002): Bill Gates memo
● Google (2005): Automated testing culture
● Amazon (2004): Jeff Bezos memo
● Twitter (2008)
● LinkedIn (2009)
● Etsy (2009)

@RealGeneKim
2002 Microsoft Security
Standdown
 Famously, Microsoft after
SQL Slammer required
every product group to
freeze feature
Source: https://www.wired.com/2002/01/bill-gates-trustworthy-computing/

@RealGeneKim
The Feature Freeze / Standdown
Debt
Features
Quality
Defects
Features

@RealGeneKim
Quote from Marty Cagan from his book
Inspired
The deal [between product owners and] engineering goes like this: Product
management takes 20% of the team’s capacity right off the top and gives this to
engineering to spend as they see fit. They might use it to rewrite, re-architect, or
re-factor problematic parts of the code base…whatever they believe is necessary
to avoid ever having to come to the team and say, ‘we need to stop and rewrite [all
our code].’ If you’re in really bad shape today, you might need to make this 30% or
even more of the resources. However, I get nervous when I find teams that think
they can get away with much less than 20%.
Cagan notes that when organizations do not pay their “20% tax,” technical debt
will increase to the point where an organization inevitably spends all of its cycles
paying down technical debt. At some point, the services become so fragile that
feature delivery grinds to a halt because all the engineers are working on reliability
issues or working around problems.

@RealGeneKim
The Third Ideal: Enabling Greatness
 Ideal: 3-5% of developers dedicated to improving
developer productivity
 Google: likely 1,500+ devs ($1B+)
 Microsoft: likely over 3,000 devs
 Not ideal: assigned to summer interns and
“people not good enough to be developers”

@RealGeneKimSource: Satya Nadella, CEO, Microsoft (@satyanadella)

@RealGeneKim
Breaking The Bottlenecks In The Flow
 Environment creation
 Code deployment
 Test setup and run (mention @rohansingh)
 Overly tight architecture
 Development
 Product management

@RealGeneKim
"Automated tests transform fear into boredom."
-- Eran Messeri, Google
Google Dev And Ops (2013)
 15,000 engineers, working on 4,000+ projects
 All code is checked into one source tree
(billions of files!)
 5,500 code commits/day
 75 million test cases are run daily

@RealGeneKim
The Third Ideal: Improvement
 Not Ideal: No one cares if someone breaks the
build, or checks in code that breaks our tests
 Ideal: When someone breaks our build or our
tests, fixing it becomes the most important work
of the moment

@RealGeneKim
The Third Ideal: Improvement
 Not ideal: When someone needs a peer review,
that person has to wait until someone else frees
up
 Ideal: Whatever I’m working on, if someone
needs a peer review, I drop whatever I’m doing to
help

@RealGeneKim
Session ID:
Ideal #4:
Psychological Safety

@RealGeneKim
DevOps Enterprise: Lessons Learned
 In 2019, we’ll hold the sixth year of the DevOps Enterprise Summit, a conference for
horses, by horses
 Over the years, we’ve had nearly 350 leaders from:
 Capital One, KeyBank, Barclays, GE Capital, ING Bank, Fidelity, PNC, ADP, BofA, Western
Union, BBVA
 Nationwide Insurance, Zurich Insurance, Allstate, Hiscox, Aviva, LV=
 Walmart, Nordstrom, Target, Macy’s, Marks and Spencer
 Nike, Adidas, Sherwin Williams
 Verizon, Telstra, T-Mobile, Orange, CSG
 Raytheon, Lockheed Martin, Northrop Grumman, CSRA, Jaguar Land Rover, Fiat/Chrysler,
Cisco
 Disney, Ticketmaster, NBC/Universal, Comcast
 Kaiser Permanente
 US Citizenship & Immigration Services, UK HM Revenue Collection, DISA Forge.mil, NZ
Ministry of Social Development, UK Welfare and Pensions, US Joint Warfare Analysis Center
 Amazon PrimeNow, CA, Compuware, Google Search, IBM, MicroFocus, Microsoft, SAP

@RealGeneKimSource: Puppet/DORA: 2017 State Of DevOps Report: https://puppet.com/resources/whitepaper/state-of-devops-report

@RealGeneKim
One Of The Highest Predictors Of
Performance
Source: Typology Of Organizational Culture (Westrum, 2004)

@RealGeneKim
Google: Project Aristotle, Oxygen, re:Work
Source: https://rework.withgoogle.com/blog/five-keys-to-a-successful-google-team/

@RealGeneKim
Great Practices Enabled
 Blameless post-mortems
 Chaos Monkeys

@RealGeneKim
Modeling Continual Learning
 “When adult learners start trying to learn a new
skill, they will often do it in private, because of the
embarrassment associated with doing something
they’re not good at.”
 We can help by saying “I don’t know"

@RealGeneKim
Session ID:
Ideal #5:
Customer Focus

@RealGeneKim
The Fifth Ideal: Focus On The Customer
 Not ideal: Functional silo managers prioritize silo
goals over business goals
 Ideal: Functional silo managers make decisions
based on what the customer values, and helps
ensure their teams have the skills to thrive in the
long term

@RealGeneKim
The Fifth Ideal: Focus On The Customer
 Core vs. Context
 Enabled reallocation of $8MM back into R&D

@RealGeneKim
Why Do I Think This Is
Important?

@RealGeneKim
Want More Learn More?
To receive this presentation and the following:
 PDF and audio excerpts from The Unicorn Project
 Eight excerpts from Beyond The Phoenix Project audio
series w/John Willis
 The 140 page excerpt of The DevOps Handbook
 The 140 page excerpt of The Phoenix Project
 Videos and slides from DevOps Enterprise 2014-2019
 One hour excerpt of The Phoenix Project audiobook
Just pick up your phone, and send an email:
To: realgenekim@SendYourSlides.com
Subject: devops
realgenekim@SendYourSlides.com
devops

The Unicorn Project and The Five Ideals (Updated Dec 2019)

More Related Content

What's hot (20)

Similar to The Unicorn Project and The Five Ideals (Updated Dec 2019) (20)

More from Gene Kim (20)

Recently uploaded (20)

The Unicorn Project and The Five Ideals (Updated Dec 2019)

Editor's Notes