DevOps 101

Concepts and Overview

© 2013 IBM Corporation

 Software Delivery is critical to success

54 %

of companies believe software delivery is critical

69
outperform
%

But only…
those who don’t

25 %
leverage software delivery effectively today
leverage
software delivery
effectively

Source: “The Software Edge: How effective software development drives competitive advantage,” IBM Institute of Business Value, March 2013
2
 And a lack of continuous delivery
impacts the entire business
CHALLENGES
CHALLENGES

Costly, error prone Slow deployment Upgrade risk due to

manual processes and to development and test managing multiple application
efforts to deliver software environments leave teams configurations and versions
across an enterprise waiting and unproductive across servers

Business Development/ Operations/

Customers Owners Test Production

Software glitch costs New Zealand’s biggest phone company, A bad software upgrade
trading firm Knight Telecom paid out $2.7 million to some at RBS Bank left
Capital $440 million 47,000 customers who were millions unable to access
in 45 minutes overcharged after a software glitch money for four days

3
 Patterns of challenges
Differences in dev Backlog of agile Manual (tribal) Lack of feedback and
and ops releases that Ops processes for quality metric leads
environments cannot handle release lack to missed service
cause failures repeatability/speed level targets

Dev Who did

this last
time?
Daily
Build Dave…

Prod Dave’s
not here
Monthly
Delivery man…
DevOps: The time is now
Four key drivers are making DevOps an imperative for all organizations.

Business
Agility

Cloud Agile
DevOps
Computing Development

Operational
Discipline
Why DevOps?
§Time to value
–Deploy faster. Deploy Often
–Reduce cost/time to deliver

§Developer ‘Self-service’
–Allow Developers to Build and Test against ‘Production-like’ systems

§Increase Quality
–Reduce cost/time to test
–Increase test coverage

§Increase environment utilization

–Virtualize Dev and Test Environments
Why DevOps?
§Deployment
–Minimize deployment related downtime
–Minimize roll-backs of deployed Apps

§Defect Resolution
–Increase the ability to reproduce and fix defects
–Minimize ‘mean-time-to-resolution’ (MTTR)
–Reduce defect cycle time

§Collaboration
–Reduce challenges related to Dev and Ops collaboration
–Dev vs. Ops
Stakeholders
§Development
§QA
§Operations
–Dev and QA Environments
–Production Environment

§Security, Data, Cloud, Enterprise Architecture…

DevOps is a Philosophy

Source: http://virtualplatosacademy.blogspot.com/p/great-philosophers.html
A blueprint for continuous delivery of software-
driven innovation
dev·ops noun \'dev-äps\
Enterprise capability for continuous software delivery that enables clients
to seize market opportunities and reduce time to customer feedback.

DevOps Lifecycle

Customers Business Development/Te Operations/Produ

Owners st ction

Continuous Feedback and Improvements

§ Accelerated software delivery § Improved governance across the lifecycle

§ Reduced time to obtain and § Balanced quality, cost and speed
respond to customer feedback

10
DevOps Principles and Values

§ Develop and test against a production-like

system

§ Iterative and frequent deployments using People

repeatable and reliable processes
Process
§ Continuously monitor and validate operational
Tools
quality characteristics

§ Amplify feedback loops

Key Concepts
The key technical Capabilities of DevOps
1. Continuous Integration : CI – Jenkins, GitLab, CircleCI, Ansible
2. Continuous Delivery : CD - Deploy Manual – Jenkins, AWS CodeBuild,
Azure Devops, Argo CD
3. Continuous Deployment – Deployment Automated – Docker, Spinnaker,
Chef, Bamboo, AWS CodeDeploy
4. Continuous Test – Selenium, Jira, Puppet, Junit
5. Continuous Feedback – Buddy, Cruisecontrol
6. Continuous Monitoring – Akamai, AppDynamics, Splunk, Dynatrace,
PagerDuty, New Relic, Sumo Logic, Datadog
7. Continuous Optimization – “
8. Continuous release – CI Tools
9. Build and Delivery Pipeline / CI-CD Pipeline
10. Continuous Development – Maven
11. Continuous Business Planning – Requirements, Rational Focal Point
12. Continuous collaboration: Confluence, Slack, Trello, Mural
 C2R (Concept to Realization) Model

Concept/ Vision

Biz IT

Business to IT Alignment
Met
Exceeds Values
Not Met

APM
Enterprise Architecture
Initiatives IT Strategy
Monitoring Technology Management
DevOps

Realization/Software System
13
1. Continuous Integration

http://bit.ly/PRQ4a7
2. Continuous Delivery

http://bit.ly/PRQ4a7
3. Continuous Test

http://bit.ly/PRQ9dQ
4. Continuous Monitoring

http://bit.ly/PRQ9dQ
6. Build & Delivery Pipeline
Delivery Pipeline

Build,
Package,
& Unit Test
.jsp .html Application
Binaries &
Platform Deploy
.java Configuration

.sh chef
recipes
Environment
Deployable Artifacts Running System
Source Artifacts
Source Control Library
Management
7. Organizational Change

‘Shift Left’ – Operational Concerns

Build ‘Application aware’ Environments
DevOps Adoption
Ø Identify the Business Value
Ø Build a Business Case

Ø Create a DevOps Culture

Ø People – Processes – Tools

Ø Identify Capabilities to Adopt/Enhance

DevOps Adoption (1 of 2)
§ Requirements Management
–Requirements Management and communication across Development and
Operations
§ Versioning of all DevOps assets
–Versioning of Deployment Scripts and Source Code
§ Access to Production-like Environments
–Documentation of Production-like environments as Patterns
–Developers have ability to launch and destroy production-like environments
from these patterns
§ Deployment Automation
–Pattern based reusable deployment scripts
–Ability to deploy applications in One-step
–Daily deployment and verification of applications to a production-like
environment
Source: 12 Steps to Better DevOps – Michael Elder
DevOps Adoption (2 of 2)
§ Change Management
–Linking bugs, issues and work items to application changes
–Linking production issues to associated deployment bugs
§ Automated Testing
–Automated testing is used to validate application and platform function and
characteristics
§ Monitoring
–Monitoring Deployed applications to validate performance and reliability
§ Delivery Pipeline
–Having a dashboard to track application stages thru the delivery pipeline and
track deployment velocity
Continuous Delivery Adoption Maturity
Ø Common Source Control
Ø Automated Builds (Build Definitions)
Ø Continuous Integration (CI)
Ø Automated Delivery
Ø Continuous Delivery to Test (CD)
Ø Continuous Delivery to Production-like
Systems (Infrastructure as Code)
Ø Continuous Delivery thru Prod
(Cloud based)
Continuous Delivery flow

Test Automation
Cloud Platform Provider

Developer Tools Execute

Request
tests
cloud
resources
Provision
Deliver resources
changes Automation Agent
Post results (execute delivery process)

Source Control and Change

Management server Publish
packages
Retrieve
packages
Trigger
delivery Artifact Library
Post
changes Virtual System

Publish
Build Server packages

25
IBM DevOps Reference Architecture
Developer Communities

Intelligent Expert
Mobile Cloud Big Data Social Devices Systems Security Industry

Implementation Services
Eco System

Develop and Test Release and Deploy Monitor and Optimize

DevOps Foundation
OSLC (enabled by the Jazz Platform) Linked Data

TOSCA
Deployment Platforms (PaaS, IaaS, Physical Infrastructure)