DevOps job tasks (the work of DevOps)

Get DevOps done using job tasks and projects based on job tasks defined by professional certification

Overview

Amazon’s DevOps Professional Exam
Testing and other habits (values)
DevOps practices
DevOps Courses
Categories for technology management
Design a DevOps strategy
Implement DevOps development processes
Implement continuous integration
Implement continuous delivery
Implement dependency management
- Design a dependency management strategy
- Manage security and compliance
Implement application infrastructure
Implement continuous feedback
Resources
More on DevOps

Most job descriptons for DevOps roles focus on vendor product offerings (GitHub, Jenkins, etc.).

NOTE: Content here are my personal opinions, and not intended to represent any employer (past or present). “PROTIP:” here highlight information I haven’t seen elsewhere on the internet because it is hard-won, little-know but significant facts based on my personal research and experience.

However, this page lists the contributions, achievements, and outcomes from working DevOps (DevTestSecOps or DevXOps), based on Job Task Analysis done for the Implementing Azure DevOps Solutions exam AZ-400 which fuilfills requirements to be designated a Microsoft Certified: Azure DevOps Engineer Expert. Expand all (click on) these high-level statements of contributions:

Design a DevOps strategy
Implement DevOps development processes
Implement continuous integration
Implement continuous delivery
Implement dependency management
Implement application infrastructure
Implement continuous feedback

(Click on each link for its job tasks, detailed below.)

Amazon’s DevOps Professional Exam

Compare against the “domains” in Amazon’s AWS Certified DevOps Engineer – Professional Complete Exam: DOP-C00 (since December 2018):

SDLC Automation : CodeCommit, CodeBuild, CodeDeploy, CodePipeline ((22%)
Configuration Management and Infrastructure as Code : CloudFormation, Elastic Beanstalk, AWS Config, ECS, Lambda, OpsWorks (19%)
Monitoring and Logging “ CloudWatch, X-Ray (15%)
Policies and Standards Automation : Service Catalog, Trusted Advisor, Secrets Manager, Organizations, Macie, Cert Manager (10%)
Incident and Event Response : Guard Duty, Inspector Kinesis (18%)
High Availability, Fault Tolerance, and Disaster Recovery : SSO, CloudFront, AutoScaling, Route53, RDS, Aurora, DynamoDB (10%)

The above I also call “processes” addressed by products.

Testing and other habits (values)

The word “testing” is not among the above, perhaps because testing is pervasive, along with other “habits”:

Team Autonomy and Enterprise Alignment
Rigorous Management of Technical Debt
Focus on Flow of Customer Value
Hypothesis Driven Development
Evidence Gathered in Production
Live Site Culture (remediate immediately and drive to root cause)
Manage infrastructure as flexible resources

DevOps practices

But testing is a separate item among VIDEO: 7 DevOps practices as defined by Microsoft’s Sam Guckenheimer:

Configuration management [Course 4]
Release management
Continuous integration [Course 3]
Continuous deployment [Course 3]
Infrastructure as Code [Course 2]
Test automation [Course 5]
Application performance monitoring [Course 7]

All but “release management” track with Microsoft’s curriculum (below).

DevOps Courses

To obtain skill to make contributions listed above, there is a 10-course Microsoft Professional Program (MPP) in DevOps on Edx.org:

QUESTION: Is AZ-400 the same as exam: Implementing Microsoft Azure DevOps Solutions 70-538 “coming soon”.

Categories for technology management

A dashboard displaying progress toward DevOps implementation would instead be grouped by categories of technology:

A. Security
B. Networks
C. Persistance (storage, databases)
D. Compute (virtual machines)
E. Management (monitoring)
F. DevOps (development to operations)
G. Integration (messaging)
H. Analytics (BI, Visualization, Machine Learning & AI)

The above list is often used to also define organizational boundaries, where technicians report to a manager in charge of each category (such as the department of Networking, etc.). The concern with such an approach is that each category is dependent on others to be setup at the same time. Each of the above needs to be a part of each application.

But traditional enterprise communication tends to be funneled up through category managers rather than directly among workers.

“Agile” teams aim to put skills that address ALL categories into each small team. That’s a tall order considering the staggering number of technologies implemented, especially in companies which have relegated each employee to a very specific role for years and don’t make it easy to hold a job which spans all these categories, even in the course of an entire 40-year career.

Here is where culture, collaboration tools, and automation may help.

PROTIP: The hope is that, eventually, advances in automation would enable small Agile teams to work like a start-up on their own, addressing all aspects of a traditional data center without ceremonies to get permission and clearance through various management ranks. So the futuristic view of the job tasks below would be how to get a computer to do each on its own with minimal human interaction.

But, for now, people are involved in its setup, configuration, and maintenance, so they need to learn the specific skills below, both manually in the GUI and in automation scripts such as Bash, Python, PowerShell, and other scripts.

Design a DevOps strategy

analyze existing artifact (e.g. deployment packages, NuGet) and container repositories
analyze existing test management tools
analyze existing work management tools
recommend migration and integration strategies for artifact repositories, source control, test management, and work management

Design and implement an Agile work management approach

identify and recommend project metrics, KPIs, and DevOps measurements (e.g. cycle time, lead time, Azure Kubernetes Service, WIP limit)
implement tools and processes to support Agile work management
mentor team members on Agile techniques and practices
recommend an organization structure that supports scaling Agile practices
recommend in-team and cross-team collaboration mechanisms (Skype, Zoom)

Design a quality strategy

analyze existing quality environment
identify and recommend quality metrics
recommend a strategy for feature flag lifecycle
recommend a strategy for measuring and managing technical debt
recommend changes to team structure to optimize quality
recommend performance testing strategy

Design a secure development process

inspect and validate code base for compliance
inspect and validate infrastructure for compliance
recommend a secure development strategy
recommend tools and practices to integrate code security validation (e.g. static code analysis)
recommend tools and practices to integrate infrastructure security validation

Design a tool integration strategy

design a license management strategy (e.g. VSTS users, concurrent pipelines, test environments, open source software licensing, DevOps tools and services, package management licensing)
design a strategy for end-to-end traceability from work items to working software
design a strategy for integrating monitoring and feedback to development teams
design an authentication and access strategy
design a strategy for integrating on-premises and cloud resources

Implement DevOps development processes

Design a version control strategy

recommend branching models
recommend version control systems
recommend code flow strategy

Implement and integrate source control

integrate external source control
integrate source control into third-party continuous integration and continuous deployment (CI/CD) systems

Implement and manage build infrastructure

implement private and hosted agents
integrate third party build systems
recommend strategy for concurrent pipelines
manage Azure pipeline configuration (e.g. agent queues, service endpoints, pools, webhooks)

Implement code flow

implement pull request strategies
implement branch and fork strategies
configure branch policies

Implement a mobile DevOps strategy

manage mobile target device sets and distribution groups
manage target UI test device sets
provision tester devices for deployment
create public and private distribution groups

Managing application configuration and secrets

implement a secure and compliant development process
implement general (non-secret) configuration data
manage secrets, tokens, and certificates
implement applications configurations (e.g. Web App, Azure Kubernetes Service, containers)
implement secrets management (e.g. Web App, Azure Kubernetes Service, containers, Azure Key Vault)
implement tools for managing security and compliance in the pipeline

Implement continuous integration

Manage code quality and security policies

monitor code quality
configure build to report on code coverage
manage automated test quality
manage test suites and categories
monitor quality of tests
integrate security analysis tools (e.g. SonarQube, WhiteSource Bolt, Open Web Application Security Project)

Implement a container build strategy

create deployable images (e.g. Docker, Azure Container Registry)
analyze and integrate Docker multi-stage builds

Implement a build strategy

design build triggers, tools, integrations, and workflow
implement a hybrid build process
implement multi-agent builds
recommend build tools and configuration (e.g. Azure Pipelines, Jenkins)
set up an automated build workflow

Implement continuous delivery

Design a release strategy

recommend release tools
identify and recommend release approvals and gates
recommend strategy for measuring quality of release and release process
recommend strategy for release notes and documentation
select appropriate deployment pattern

Set up a release management workflow

automate inspection of health signals for release approvals by using release gates
configure automated integration and functional test execution
create a release pipeline (e.g. Azure Kubernetes Service, Service Fabric, WebApp)
create multi-phase release pipelines
integrate secrets with release pipeline
provision and configure environments
manage and modularize tasks and templates (e.g. task and variable groups)

Implement an appropriate deployment pattern

implement blue-green deployments
implement canary deployments
implement progressive exposure deployments
scale a release pipeline to deploy to multiple endpoints (e.g. deployment groups, Azure Kubernetes Service, Service Fabric)

Implement dependency management

Design a dependency management strategy

recommend artifact management tools and practices (Azure Artifacts, npm, maven, Nuget)
abstract common packages to enable sharing and reuse
inspect codebase to identify code dependencies that can be converted to packages
identify and recommend standardized package types and versions across the solution
refactor existing build pipelines to implement version strategy that publishes packages

Manage security and compliance

inspect open source software packages for security and license compliance to align with corporate standards (e.g. GPLv3)
configure build pipeline to access package security and license rating (e.g. Black Duck, White Source)
configure secure access to package feeds

Implement application infrastructure

Design an infrastructure and configuration management strategy

analyze existing and future hosting infrastructure
analyze existing Infrastructure as Code (IaC) technologies
design a strategy for managing technical debt on templates
design a strategy for using transient infrastructure for parts of a delivery lifecycle
design a strategy to mitigate infrastructure state drift

Implement infrastructure as code (IaC)

create nested resource templates
manage secrets in resource templates
provision Azure resources
recommend an Infrastructure as Code (IaC) strategy
recommend appropriate technologies for configuration management (ARM Templates, Terraform, Chef, Puppet, Ansible)

Manage Azure Kubernetes Service infrastructure

provision Azure Kubernetes Service (e.g. using ARM templates, CLI)
create deployment file for publishing to Azure Kubernetes Service (e.g. kubectl, Helm)
develop a scaling plan

Implement infrastructure compliance and security

implement compliance and security scanning
prevent drift by using configuration management tools
automate configuration management by using PowerShell Desired State Configuration (DSC)
automate configuration management by using a VM Agent with custom script extensions
set up an automated pipeline to inspect security and compliance

Implement continuous feedback

design practices to measure end-user satisfaction (e.g. Send a Smile, app analytics)
design processes to capture and analyze user feedback from external sources (e.g. Twitter, Reddit, Help Desk)
design routing for client application crash report data (e.g. HockeyApp)
recommend monitoring tools and technologies
recommend system and feature usage tracking tools

Implement process for routing system feedback to development teams

configure crash report integration for client applications
develop monitoring and status dashboards
implement routing for client application crash report data (e.g. HockeyApp)
implement tools to track system usage, feature usage, and flow
integrate and configure ticketing systems with development team’s work management system (e.g. IT Service Management connector, ServiceNow Cloud Management, App Insights work items)

Optimize feedback mechanisms

analyze alerts to establish a baseline
analyze telemetry to establish a baseline
perform live site reviews and capture feedback for system outages
perform ongoing tuning to reduce meaningless or non-actionable alerts

Resources

Amazon’s Study Guide for AZ-400

ACoud.guru vidoes

More on DevOps

This is one of a series on DevOps:

Wilson Mar