Слайд 2The Agile Admins
Ernest Mueller (@ernestmueller)
James Wickett (@wickett)
Karthik Gaekwad (@iteration1)
Who Are We?
Слайд 4Things DevOpsers Care About
Source: Devopsdays Austin Survey 2015 (500 participants)
Слайд 5Topics covered today
What is DevOps?
Why would I use it?
What’s hot in
DevOps?
What does the future look like?
Слайд 6DevOps Defined
DevOps is the practice of operations and development engineers participating
together in the entire service lifecycle, from design through the development process to production support.
DevOps is also characterized by operations staff making use many of the same techniques as developers for their systems work.
Слайд 8The Layers of DevOps
DevOps Principles
DevOps Tools
DevOps Practices
Слайд 9DevOps Principles
The Three Ways (Gene Kim)
Systems Thinking
Amplify Feedback Loops
Culture of Continual
Experimentation
CAMS (John Willis)
Culture – People > Process > Tools
Automation – Infrastructure as Code
Measurement – Measure Everything
Sharing – Collaboration/Feedback
Informed by the values in the Agile Manifesto and Lean Theory of Constraints
Слайд 10DevOps Practices
•Version Control For All
•Automated Testing
•Proactive Monitoring and Metrics
•Kanban/Scrum
•Visible Ops/Change Management
•Configuration
Management
•Incident Command System
•Continuous Integration/Deployment/Delivery
•“Put Developers On Call”
•Virtualization/Cloud/Containers
•Toolchain Approach
•Transparent Uptime/Incident Retrospectives
Слайд 11DevOps Tools
Hardware (System provisioning/automation)
Software (Software delivery pipeline)
Wetware (Process/human element)
Monitoring/Metrics (of all
the above)
Слайд 12DevOps Adoption Impact
2014 State of DevOps survey by Puppet Labs, Thoughtworks,
and IT Revolution Press
Top indicators of high organizational performance:
IT performance - Leaders here are twice as likely to exceed their profitability, market share, and productivity goals. Deployment frequency, lead time for changes, mean time to recover from failure are positively impacted by DevOps practices.
Organizational culture and climate for learning
Job satisfaction
Слайд 13What’s Hot Now
Security and DevOps
Internet of Things
Containers and Docker
Microservices and 12-Factor
Apps
ChatOps
Unicorns to Horses to Donkeys
Слайд 14Security is where ops was 5 years ago...
Слайд 15Epiphanies in Security and DevOps
Smaller batches make you more secure
Departure from
the perimeter mentality
Instrument your runtime environment and correlate disparate metrics (e.g. transactions to csrf token generations)
Emphasis on MTTD and MTTR
Слайд 16Epiphanies in Security and DevOps
We can go fast and still do
compliance
Major vulnerabilities like Heartbleed exposed weaknesses in our software supply chain
Security testing can be done in development
Слайд 17Security Companies to Watch
Signal Sciences
Threat Stack
Sonatype
Contrast Security
Слайд 19DevOps and IoT
Fully distributed compute on networks you can’t/don’t control
Software and
tracking components becomes really hard
Automation even more necessary
Security even more necessary
Слайд 21
Open source platform for developers and sysadmins to build, ship and
run distributed applications anywhere.
What is Docker?
Слайд 22Docker Components
Docker Engine
Runtime and packaging tool
Installed on hosts that run
Docker
Docker Hub
Cloud service for storing and sharing apps
Save your docker images (public/private)
Слайд 23Why is it so popular?
Portability
Write once, run anywhere
Standardized Environments
Dev/QA/Production can be
modeled exactly the same way
Rapid scale-up/scale-down
Containers take seconds to deploy
Ability to build a continuous delivery (CD) pipeline
Слайд 24Use cases for Docker
Continuous Integration & Delivery
Yelp, Spotify, Mailgun, Cambridge Healthcare
PaaS
Yandex,
Baidu
Easier Development & Deployment
Ebay, Bleacher Report, New Relic
Hosting Legacy Applications
https://www.docker.com/resources/usecases/
Слайд 27
“Independent processes communicating with each other with well defined API’s to
form larger more complex applications”
Microservices
Слайд 28Microservices characteristics
Do one thing, and do it well (fine grained)
Independently built-
common language to communicate
Independently deployable
Fault tolerant and reliable
Слайд 2912 Factor App
Use declarative formats for setup automation; minimize time and
cost for new developers joining the project
Clean contract with the underlying OS, offering maximum portability between execution environments;
Suitable for deployment on modern cloud platforms
Minimize divergence between development and production- enabling continuous deployment
Scale up without significant changes to tooling, architecture, or development practices.
Слайд 30Relationship with containers
Microservices and containers are a great fit.
Single service on
a container.
Isolates service and makes it easy to manage
Emerging best practice for new architectures with containers.
Consider building your architecture in this manner.
Слайд 32Unicorns to Horses to Donkeys
Gartner: “By 2016 25% of the G2000
will be employing DevOps.”
“By 2018, 90% of I&O organizations attempting to use DevOps without specifically addressing their cultural foundations will fail.”
Слайд 34Future
More Lean Extension
Security Fully DevOpsed
Cloudification/Compute Fabric
Testing and Monitoring
Devops - The New
Normal
Containers Rule The World
Слайд 35Nice (Software Supply) Chain!
Extension of Lean
Слайд 36Delivery Pipeline extension to Lean
Your BOM is part of your product
Ability
to track where all code came from which products are using it just like automotive and aeronautical industries
Quality metrics extending to customer usage
Слайд 38Security into Devops
Security rebranded in language of Rugged as features (ruggedsoftware.org)
Growth
around integrating security at all phases of the delivery pipeline
Instrument runtime environment with security metrics
Movement from security training to security testing
Searching for aspirational moment like “10 deploys a day” did for devops
Слайд 39Cloud/Compute Fabric
Rightscale 2015 State of the Cloud Survey reveals 93% of
organizations are running apps on or experimenting with IaaS
82% of enterprises have a hybrid cloud strategy
IoT and Big Data mean large heaps of data it’s hard to move around the network
Containerization and microservices mean it’s easier to move compute to the data than vice versa
Eventual: Complex, ultra-distributed compute
Слайд 40Testing and Monitoring
Automated testing is a best practice
Light/fast testing feeds CD
App
and system monitoring is a best practice
Logging, analytics being absorbed into it
There is no difference between monitoring and regression testing theoretically - only historically
Eventual: Convergence of regression testing and monitoring
Слайд 41Containers rule the world
Yesterday: Docker didn’t exist
Today: Engineers playing with Docker
in dev/test.
Tomorrow: Production Docker
Even lesser emphasis on hardware/cloud.
Eventual: Microservices and 12 factor apps inside containers will be the defacto standard to deploy apps.
Слайд 42DevOps: The New Normal
Devops is penetrating enterprises.
Yesterday: Unicorns
Today: Horses & Donkeys
Tomorrow:
Everyone
Backlash: Where’s the silver bullet? What is the correct implementation of DevOps?
Eventual: Devops will be the new normal