HAUFE‘S TECH STRATEGY IN PRACTICE
Marco SeifriedLead Platform Architect@marcoseifried [email protected]
A story on how to develop modern age systems and help business to reach their goals
Haufe Group
Digital Media Group providing digital workplace and business solutions
Online services for HR managers, accounting apps for the self-employed, Internet portals for tax experts, online and offsite trainings
Founded 1930, family owned still, now 1500 employees
Locations in Germany, Spain, Switzerland, US, China and Romania
WE NEED TO TRANSFORM OUR BUSINESS
...CONSTANTLY
EFFICIENCY IN LEGACY BUINESS
SPEND MONEY ON VALUE
SPEED TO MARKET GLOBALIZATION
DIGITALIZATION OFBUSINESS MODELS
SPARKINGINNOVATION
CHALLENGES IN TRANSFORMING BUSINESS
HOW TO FOSTER INNOVATION?
Culture
StructureTechnology
ALIGNMENT & BALANCE
Values• Values is what we value• High Performance• Freedom & Responsibility• Context, not control
Skills• Curiosity• Courage• Passion• Selflessness
Culture @Netflix
A must read slidedeck: http://de.slideshare.net/reed2001/culture-1798664
Technical Strategy@Haufe
BASELINE – ARCHITECTURAL PRINCIPLES
Business value over technical strategyStrategic goals over project-specific benefitsComposability over silos
Shared services over specific-purpose implementationsEvolutionary
refinement over pursuit of initial perfection
Design for obsoleteness over building for eternity
Good enough over best of breedDeclarative processes over implicit knowledge
Data over opinion
02.05.2023 11
MICROSERVICES
• Vertical Decomposition • Shared Nothing Architecture
• Independent Deployable, Scalable and Evolvable• Do One Thing and Do One Thing Well
AUTOMATE YOUR ECOSYSTEM
• Build• Infrastructure as Code -> git• Pipelines -> go.cd, Jenkins• Docker (container, docker compose), Kubernetes, Consul• Docker Registry
• Deploy• Feature Flags• Blue / Green• Rolling Updates• Immutable Infrastructure
• Test• martinfowler.com/articles/microservice-testing
AUTOMATE YOUR ECOSYSTEM
• Operate• Docker Swarm, Kubernetes, Amazon / Azure Container Service, Rancher• Autoscaling ? • Service Discovery external (AWS Elastic Loadbalancer) & internal (Kubernetes)
• Monitor, Logging, Reporting• Health endpoints per service• Central Log Management -> fluentd, Graylog• Monitor -> Grafana, Graphite, InfluxDB,
Prometheus
COMBINE THAT WITH…• Cloud to make use of IaaS, PaaS & SaaS
• Watch out: Lock-in vc. benefits of offerings• Docker as base to be independent• Do it yourself where there is unique business value• Don‘t just ‚lift & shift‘ – think about cloud architecture
• APIs • to decouple systems / microservices • outside-in design: don‘t expose what you have, expose what your customer
needs• Self serviced
LAST BUT NOT LEAST - PRODUCT TEAMS
• Responsible• Empowered• Cross functional• DevOps• Conway‘s Law
Spotify is a prime example: http://www.full-stackagile.com/2016/02/14/team-organisation-squads-chapters-tribes-and-guilds/
Some real life experience
Introducing – Our Service Platform
http://fineartamerica.com/featured/big-ball-from-a-cable-twisted-pair-aleksandr-volkov.html
- 5 to 10 days to deploy- 2 releases a year- Months to test- White box testing - Deployed on hosted hardware- test != prod- Not an ESB, but worse- Serves a wide range of
services from a single entity
Lets start simple and extract (just) User Management
9 months later - it was a complete failure
- Lost test coverage (Remember white box testing?)
- Scope creep (lets fix all the things which bugged us)
- Project complexity grew out of control (Leading Indicator: “We just need X more developers”)
- Agile became frAgile- Team morale and spirit were destroyed
Stabilize Optimize Transform
The 3 Stages of Lean Transformation
So we reset the project (but kept the team)
• Strictly time-boxed phases (3 months)• Additional focus on team culture, attitudes, and challenge
• Start with the basics – Automation, to gain time for the more valuable features
• Result: Time to deploy from 5-10 days to 30 min• Infrastructure as Code (Docker)• Rapid deployment via Cloud (Azure)• Build and deployment automation through Go.CD• Inmutable Server
Next up: Haufe Publishing System• Haufe‘s factory: Producing our content• Monolithic application, grown over the
years• Max 2 releases per year
Work @Haufe karriere.haufe-lexware.com/en/
. You’re full of energy, looking for a challenge and willing to learn on the job? Great! We give you space to mobilize that energy and channel it into exciting projects.
What we offer: Basic Information:
Internships 1 – 6 month(s), 39 hrs per week
Work Semesters max. 20 hrs per week/semestermax. 40 hrs per week /holiday
Traineeships esp. Sales, Marketing, Product and Project management12 -14 months, full time
Bachelor- / Master Thesis
Positions Software Developer, - Architect, Tester, Project Manager
Find out more.
For job opportunities check out: karriere.haufe-lexware.com/en/
For daily news about us, our teams, twitter.com/HG_Jobsour work and jobs follow us: www.linkedin.com/company/haufe-gruppe
www.xing.com/company/haufe-gruppe
See what other people have to say about us: www.kununu.com/de/haufe-gruppe
Technical blog, GitHub, Twitter: dev.haufe.comgithub.com/Haufe-Lexware@HaufeDev
Get in contact: [email protected]
Further information on our work and projects
Project example from this deck dev.haufe.com/the-automated-monolith
OSS API Mgnt wicked.haufe.io
Dockergithub.com/Haufe-Lexware/docker-style-guide
API Styleguide github.com/Haufe-Lexware/api-style-guide
Slideshare www.slideshare.net/HaufeDev
Bedtime reading
@danielbryantuk
