Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Don't Fear the Walking Dead - How to get rid of...

Avatar for Sebastian Heuer Sebastian Heuer
May 10, 2016
Technology
0
74

Don't Fear the Walking Dead - How to get rid of a monolith

Avatar for Sebastian Heuer

Sebastian Heuer

May 10, 2016
Tweet

More Decks by Sebastian Heuer

See All by Sebastian Heuer
The State of Asynchronous PHP
Avatar for Sebastian Heuer sebastianheuer
0
33
Typescript - A Marriage Counselor for Frontend and Backend
Avatar for Sebastian Heuer sebastianheuer
0
41
The Myth of Untestable Code
Avatar for Sebastian Heuer sebastianheuer
0
29
Deployment without the Hassle @ IPC 2016
Avatar for Sebastian Heuer sebastianheuer
1
53
99% is not enough
Avatar for Sebastian Heuer sebastianheuer
0
120
Don't Fear the Walking Dead - How to get rid of a zombie
Avatar for Sebastian Heuer sebastianheuer
0
110
A Developer's Survival Guide for the Microservice Era
Avatar for Sebastian Heuer sebastianheuer
2
380
PHPUGHH - Deployment with OS Packages
Avatar for Sebastian Heuer sebastianheuer
0
79
How to manage your Project's Toolchain in 2016
Avatar for Sebastian Heuer sebastianheuer
1
48

Other Decks in Technology

See All in Technology
Okta Identity Governanceで実現する最小権限の原則
Avatar for 株式会社出前館 demaecan
0
210
OPENLOGI Company Profile for engineer
Avatar for OPENLOGI hr01
1
46k
メールやSlack通知をトリガーにした非同期APIテスト基盤を作ってみた / async-test-platform-for-automated-testing
Avatar for bun bun913
0
120
AI時代の発信活動 ～技術者として認知してもらうための発信法～ / 20251028 Masaki Okuda
Avatar for SHIFT EVOLVE shift_evolve
PRO
1
120
生成AI時代のPythonセキュリティとガバナンス
Avatar for abenben abenben
0
160
CLIPでマルチモーダル画像検索 →とても良い
Avatar for Motoi Washida wm3
1
650
abema-trace-sampling-observability-cost-optimization
Avatar for tetsuya28 tetsuya28
0
380
デザインとエンジニアリングの架け橋を目指す OPTiMのデザインシステム「nucleus」の軌跡と広げ方
Avatar for OPTiM optim
0
120
DSPy入門
Avatar for TomuHirata tomehirata
6
680
JAWS UG AI／ML #32 Amazon BedrockモデルのライフサイクルとEOL対応/How Amazon Bedrock Model Lifecycle Works
Avatar for quiver quiver
1
130
ViteとTypeScriptのProject Referencesで 大規模モノレポのUIカタログのリリースサイクルを高速化する
Avatar for did0es shuta13
3
230
AIでデータ活用を加速させる取り組み / Leveraging AI to accelerate data utilization
Avatar for okiyuki okiyuki99
6
1.5k

Featured

See All Featured
The Art of Delivering Value - GDevCon NA Keynote
Avatar for David Neal reverentgeek
16
1.7k
XXLCSS - How to scale CSS and keep your sanity
Avatar for Zaharenia Atzitzikaki sugarenia
249
1.3M
How To Stay Up To Date on Web Technology
Avatar for Chris Coyier chriscoyier
791
250k
The Success of Rails: Ensuring Growth for the Next 100 Years
Avatar for Eileen M. Uchitelle eileencodes
46
7.7k
Visualizing Your Data: Incorporating Mongo into Loggly Infrastructure
Avatar for mongodb mongodb
48
9.7k
The Language of Interfaces
Avatar for Des Traynor destraynor
162
25k
Leading Effective Engineering Teams in the AI Era
Avatar for Addy Osmani addyosmani
7
670
Docker and Python
Avatar for Tania Allard trallard
46
3.6k
How Fast Is Fast Enough? [PerfNow 2025]
Avatar for Tammy Everts tammyeverts
2
160
Practical Tips for Bootstrapping Information Extraction Pipelines
Avatar for Matthew Honnibal honnibal
PRO
23
1.5k
10 Git Anti Patterns You Should be Aware of
Avatar for Lemi Orhan Ergin lemiorhan
PRO
658
61k
Cheating the UX When There Is Nothing More to Optimize - PixelPioneers
Avatar for Stéphanie Walter stephaniewalter
285
14k

Transcript

  1. HOW WE GET RID OF OUR MONOLITH DON’T

  2. None
  3. None
  4. None

  5. How we started • built on top of a standard

    e-commerce platform • a lot of custom code to support a wide combination of product options Feature X Feature Y Feature Z Core Code Customization

  6. The challenges of a growing business • serious performance issues

    • own developers don’t feel comfortable with the code • adding individual features is too complex • upgrading is hard

  7. Finding a way out

  8. You don’t have to replace everything at once!

  9. What part of the software causes the highest amount of

    pain? • category pages • product detail pages

  10. Planning

  11. Current Setup Legacy System Session Storage RDBMS Webserver

  12. Idea Legacy System Session Storage RDBMS Webserver New Software

  13. None
  14. None

  15. FURY Frontend Key-Value Storage getCategoryPage(341); "<html><head>…</head><body>…</body></html>"

  16. None

  17. FURY Frontend Key-Value Storage ["SBU06HE","TEOD3HE","ANW04HD"] Search search("{'category':341, 'filters':{...}}"); getTiles(["SBU06HE","TEOD3HE","ANW04HD"]); ["<div

    class="category-item">…</div>", …]

  18. Webserver Key-Value Storage Search FURY Frontend FURY Backend FURY Components

  19. Webserver Key-Value Storage Search FURY Frontend FURY Backend Legacy RDBMS

    Legacy System Session Storage Collect & Export

  20. Webserver Key-Value Storage Search FURY Frontend FURY Backend Legacy RDBMS

    Legacy System Session Storage FURY Requests 200 OK

  21. Webserver Key-Value Storage Search FURY Frontend FURY Backend Legacy RDBMS

    Legacy System Session Storage Requests to Legacy System 404 NOT FOUND

  22. Webserver Key-Value Storage Search FURY Frontend FURY Backend Legacy RDBMS

    Legacy System Session Storage Requests to Legacy System 404 NOT FOUND 200 OK

  23. Learnings • we have full responsibility now • no framework

    needed • no relational database needed

  24. Goods and Bads • fully object-oriented • easy refactoring thanks

    to 100% code coverage • not enough automated acceptance tests • dependencies to legacy software (database changes, API calls)

  25. Deployment

  26. Webserver (Router) Webserver FURY Frontend Server A K/V Storage Search

    FURY Backend Webserver FURY Frontend Server B K/V Storage Search FURY Backend

  27. Webserver (Router) Webserver FURY Frontend Server A K/V Storage Search

    FURY Backend Webserver FURY Frontend Server B K/V Storage Search FURY Backend active = A

  28. Webserver (Router) Webserver FURY Frontend Server A K/V Storage Search

    FURY Backend Webserver FURY Frontend Server B K/V Storage Search FURY Backend active = B

  29. RDBMS Read Slave Webserver FURY Frontend Server A K/V Storage

    Search FURY Backend Webserver (Router) Build Server Deploy Code active = B

  30. RDBMS Read Slave Webserver FURY Frontend Server A K/V Storage

    Search FURY Backend Webserver (Router) Build Server Deploy Code Collect & Export active = B

  31. RDBMS Read Slave Webserver FURY Frontend Server A K/V Storage

    Search FURY Backend Webserver (Router) Build Server Deploy Code Collect & Export Smoke Tests active = B

  32. RDBMS Read Slave Webserver FURY Frontend Server A K/V Storage

    Search FURY Backend Webserver (Router) Build Server Deploy Code Collect & Export Smoke Tests Switch to A active = A

  33. RDBMS Read Slave Webserver FURY Frontend Server A K/V Storage

    Search FURY Backend active = B Webserver (Router) Build Server Deploy Code Collect & Export

  34. We are very close to launch!

  35. What next? • party hard! • identify the next pain

    points we want to work on • replace the next parts of the legacy system with new, self-contained components / services

  36. https://www.instagram.com/kartenmacherei/ Q&A https://www.facebook.com/kartenmacherei/ +49 40 468996861 [email protected] http://www.kartenmacherei.de/recruiting