1/23/25 - Montreal JUG - Quarkus Compare and Contrast

Are you a Spring developer and interested in learning something new ?

In this session I will show concepts and conventions familiar to Spring developers and how those same concepts and conventions can be implemented in Quarkus, all while highlighting similarities and differences between them, including how testing is done.

Additionally, I will highlight performance characteristics between them and prove, live, that you could save 50% in your operational costs and reduce your carbon emissions 2-3x simply by switching your Java runtime.

1. @edeandrea for Developers Eric Deandrea Sr. Principal Developer Advocate Red Hat
2. @edeandrea Who’s on Stage Today? Eric Deandrea 🤘 25+ years software development experience Java Champion 🤘 Contributor to Open Source projects Quarkus Spring Boot, Spring Framework, Spring Security LangChain4j (& Quarkus LangChain4j) Wiremock Microcks 🤘 Boston Java Users ACM Chapter Board Member 🤘 Published author
3. @edeandrea • How Quarkus enables modern Java development & the Kubernetes-native experience • Introduce familiar Spring concepts, constructs, & conventions and how they map to Quarkus • Emphasis on testing patterns & practices https://red.ht/quarkus-spring-devs
4. @edeandrea Java,The Enterprise Workhorse
5. @edeandrea Java,The Enterprise Workhorse
6. @edeandrea Monolith J2SE / J2EE Java / Jakarta EE Java,The Enterprise Workhorse
7. @edeandrea Monolith J2SE / J2EE Java / Jakarta EE Java,The Enterprise Workhorse
8. @edeandrea Monolith J2SE / J2EE Java / Jakarta EE Cloud-Native Microservices Spring Boot MicroProfile Java,The Enterprise Workhorse
9. @edeandrea Monolith J2SE / J2EE Java / Jakarta EE Cloud-Native Microservices Spring Boot MicroProfile Java,The Enterprise Workhorse
10. @edeandrea Monolith J2SE / J2EE Java / Jakarta EE Cloud-Native Microservices Spring Boot MicroProfile Java,The Enterprise Workhorse
11. @edeandrea Java / Jakarta EE J2SE / J2EE Monolith Cloud-Native Microservices Spring Boot MicroProfile Java,The Enterprise Workhorse
12. @edeandrea Java / Jakarta EE J2SE / J2EE Monolith Cloud-Native Microservices Serverless Event-Driven Microservices Cloud-Native Spring Boot MicroProfile Java,The Enterprise Workhorse
13. @edeandrea The Warmup Issue with Java Simon Ritter - Azul Systems - https://youtu.be/bWmuqh6wHgE (first 13 minutes)
14. @edeandrea Build Time Runtime Packaging (maven, etc) gradle…) Load config file from file system Parse it Classpath scanning to find annotated classes Attempt to load class to enable/disable features Build its model of the world. Start the management (thread, pool…) @ @ </> How Does a Framework Start?
15. @edeandrea The Quarkus Way Runtime Build Time @ @ </> Package model
16. @edeandrea JVM Build Time @ @ </> Package model Native The Quarkus Way enables Native Compilation
17. @edeandrea What’s the Secret Sauce? Holly Cummins - https://youtu.be/vnrXXvnqriA
18. @edeandrea What’s the Secret Sauce? 🤘Don’t be dynamic Holly Cummins - https://youtu.be/vnrXXvnqriA
19. @edeandrea What’s the Secret Sauce? 🤘Don’t be dynamic Holly Cummins - https://youtu.be/vnrXXvnqriA
20. @edeandrea What’s the Secret Sauce? Holly Cummins - https://youtu.be/vnrXXvnqriA
21. @edeandrea What’s the Secret Sauce? 🤘Don’t be dynamic 🤘Dead code elimination / inlining Holly Cummins - https://youtu.be/vnrXXvnqriA
22. @edeandrea What’s the Secret Sauce? 🤘Don’t be dynamic 🤘Dead code elimination / inlining Holly Cummins - https://youtu.be/vnrXXvnqriA
23. @edeandrea What’s the Secret Sauce? 🤘Don’t be dynamic 🤘Dead code elimination / inlining Holly Cummins - https://youtu.be/vnrXXvnqriA
24. @edeandrea What’s the Secret Sauce? 🤘Don’t be dynamic 🤘Dead code elimination / inlining Holly Cummins - https://youtu.be/vnrXXvnqriA
25. @edeandrea What’s the Secret Sauce? 🤘Don’t be dynamic 🤘Dead code elimination / inlining Holly Cummins - https://youtu.be/vnrXXvnqriA
26. @edeandrea What’s the Secret Sauce? Holly Cummins - https://youtu.be/vnrXXvnqriA
27. @edeandrea What’s the Secret Sauce? 🤘Don’t be dynamic 🤘Dead code elimination / inlining 🤘Extensible build process so libraries can participate Holly Cummins - https://youtu.be/vnrXXvnqriA
28. @edeandrea What’s the Secret Sauce? 🤘Don’t be dynamic 🤘Dead code elimination / inlining 🤘Extensible build process so libraries can participate 🤘Jandex Indexing (offline reflection) Holly Cummins - https://youtu.be/vnrXXvnqriA
29. @edeandrea What’s the Secret Sauce? 🤘Don’t be dynamic 🤘Dead code elimination / inlining 🤘Extensible build process so libraries can participate 🤘Jandex Indexing (offline reflection) Holly Cummins - https://youtu.be/vnrXXvnqriA
30. @edeandrea What’s the Secret Sauce? Holly Cummins - https://youtu.be/vnrXXvnqriA
31. @edeandrea What’s the Secret Sauce? 🤘Don’t be dynamic 🤘Dead code elimination / inlining 🤘Extensible build process so libraries can participate 🤘Jandex Indexing (offline reflection) 🤘Don’t make humans tell the computer what it knows Holly Cummins - https://youtu.be/vnrXXvnqriA
32. @edeandrea What’s the Secret Sauce? 🤘Don’t be dynamic 🤘Dead code elimination / inlining 🤘Extensible build process so libraries can participate 🤘Jandex Indexing (offline reflection) 🤘Don’t make humans tell the computer what it knows Holly Cummins - https://youtu.be/vnrXXvnqriA
33. @edeandrea What’s the Secret Sauce? 🤘Don’t be dynamic 🤘Dead code elimination / inlining 🤘Extensible build process so libraries can participate 🤘Jandex Indexing (offline reflection) 🤘Don’t make humans tell the computer what it knows Holly Cummins - https://youtu.be/vnrXXvnqriA
34. @edeandrea What’s the Secret Sauce? 🤘Don’t be dynamic 🤘Dead code elimination / inlining 🤘Extensible build process so libraries can participate 🤘Jandex Indexing (offline reflection) 🤘Don’t make humans tell the computer what it knows Holly Cummins - https://youtu.be/vnrXXvnqriA
35. @edeandrea What’s the Secret Sauce? Holly Cummins - https://youtu.be/vnrXXvnqriA
36. @edeandrea What’s the Secret Sauce? 🤘Don’t be dynamic 🤘Dead code elimination / inlining 🤘Extensible build process so libraries can participate 🤘Jandex Indexing (offline reflection) 🤘Don’t make humans tell the computer what it knows 🤘Continuous profiling & shaving Holly Cummins - https://youtu.be/vnrXXvnqriA
37. @edeandrea What’s the Secret Sauce? 🤘Don’t be dynamic 🤘Dead code elimination / inlining 🤘Extensible build process so libraries can participate 🤘Jandex Indexing (offline reflection) 🤘Don’t make humans tell the computer what it knows 🤘Continuous profiling & shaving 🤘Be reactive without forcing humans to do reactive Holly Cummins - https://youtu.be/vnrXXvnqriA
38. @edeandrea Unification of Imperative & Reactive Unification of Imperative and Reactive
39. @edeandrea 🤘Zero-config Live coding 🤘Auto-provision services 🤘Continuous testing 🤘Dev UI 🤘CLI Enhancing Developer Joy
40. @edeandrea IT’S STILL JAVA!
41. @edeandrea AOT/Native vs JVM Mode
42. @edeandrea
43. @edeandrea Jam Time!
44. @edeandrea Migration Toolkit For Applications “Simplifies the Migration of Spring Apps to Quarkus” 🤘 Automate Application Analysis 🤘 Estimate Level of Effort 🤘 Accelerate Code Transformation & Migration 🤘 Includes Rules for DI, Metrics, Security, Web, Shell, & More https://developers.redhat.com/products/mta
45. @edeandrea https://www.techempower.com/benchmarks/#section=data-r22&test=db&f=zijunz-zik0zj-zik0zj-zik0zj-zik0zj-zik0zj-zik0zj-v2qiv3-xamxa7-zik0zj-zik0zj-zik0zj-zik0zj- zik0zj-1ekf&hw=ph&l=zik0vz-cn3 https://www.redhat.com/en/blog/key-findings-idc-red-hat-quarkus-lab-validation https://www.reddit.com/r/java/comments/o0ewar/do_quarkus_performance_benefits_scale Don’t Take My Word For It!
46. @edeandrea JVM Native Quarkus Spring Boot Ratio (Quarkus / SB) Spring Boot (w/AOT) Ratio (Quarkus / SB AOT) Ratio (SB AOT / SB) Quarkus Spring Boot Ratio (Quarkus / SB) Framework version 3.17.5 3.4.1 3.4.1 3.17.5 3.4.1 Build time (s) 12.88 7.82 164.75% 16.88 76.34% 215.81% 172.8 283.39 60.97% Av. RSS after startup (MB) 190.6 297.92 63.98% 293.88 64.86% 98.64% 71.09 206.47 34.43% Av. time to 1st req (ms) 4,684 8,885.33 52.72% 7,940 58.99% 89.36% 132.67 924.33 14.35% Av. RSS after 1st req (MB) 241.29 344.86 69.97% 335.76 71.86% 97.36% 78.95 208.87 37.8% Av. throughput (req/sec) 30,769.99 6,713.7 458.32% 6,750.04 455.85% 100.54% 16,189.26 6,024.12 268.74% Max throughput density (req/sec/MB) 41.77 8.58 487.08% 8.75 477.32% 102.04% 61.1 14.33 426.38% Build RSS (GB) 6.0 7.83 76.63% Binary Size (MB) 90.47 156.58 57.78% Don’t Take My Word For It! Some real numbers (8 cores, 14GB RAM, GraalVM CE 21.0.2)
47. @edeandrea JVM Native Quarkus Spring Boot Ratio (Quarkus / SB) Spring Boot (w/AOT) Ratio (Quarkus / SB AOT) Ratio (SB AOT / SB) Quarkus Spring Boot Ratio (Quarkus / SB) Framework version 3.17.5 3.4.1 3.4.1 3.17.5 3.4.1 Build time (s) 12.88 7.82 164.75% 16.88 76.34% 215.81% 172.8 283.39 60.97% Av. RSS after startup (MB) 190.6 297.92 63.98% 293.88 64.86% 98.64% 71.09 206.47 34.43% Av. time to 1st req (ms) 4,684 8,885.33 52.72% 7,940 58.99% 89.36% 132.67 924.33 14.35% Av. RSS after 1st req (MB) 241.29 344.86 69.97% 335.76 71.86% 97.36% 78.95 208.87 37.8% Av. throughput (req/sec) 30,769.99 6,713.7 458.32% 6,750.04 455.85% 100.54% 16,189.26 6,024.12 268.74% Max throughput density (req/sec/MB) 41.77 8.58 487.08% 8.75 477.32% 102.04% 61.1 14.33 426.38% Build RSS (GB) 6.0 7.83 76.63% Binary Size (MB) 90.47 156.58 57.78% Don’t Take My Word For It! Some real numbers (8 cores, 14GB RAM, GraalVM CE 21.0.2)
48. @edeandrea JVM Native Quarkus Spring Boot Ratio (Quarkus / SB) Spring Boot (w/AOT) Ratio (Quarkus / SB AOT) Ratio (SB AOT / SB) Quarkus Spring Boot Ratio (Quarkus / SB) Framework version 3.17.5 3.4.1 3.4.1 3.17.5 3.4.1 Build time (s) 12.88 7.82 164.75% 16.88 76.34% 215.81% 172.8 283.39 60.97% Av. RSS after startup (MB) 190.6 297.92 63.98% 293.88 64.86% 98.64% 71.09 206.47 34.43% Av. time to 1st req (ms) 4,684 8,885.33 52.72% 7,940 58.99% 89.36% 132.67 924.33 14.35% Av. RSS after 1st req (MB) 241.29 344.86 69.97% 335.76 71.86% 97.36% 78.95 208.87 37.8% Av. throughput (req/sec) 30,769.99 6,713.7 458.32% 6,750.04 455.85% 100.54% 16,189.26 6,024.12 268.74% Max throughput density (req/sec/MB) 41.77 8.58 487.08% 8.75 477.32% 102.04% 61.1 14.33 426.38% Build RSS (GB) 6.0 7.83 76.63% Binary Size (MB) 90.47 156.58 57.78% Don’t Take My Word For It! Some real numbers (8 cores, 14GB RAM, GraalVM CE 21.0.2)
49. @edeandrea JVM Native Quarkus Spring Boot Ratio (Quarkus / SB) Spring Boot (w/AOT) Ratio (Quarkus / SB AOT) Ratio (SB AOT / SB) Quarkus Spring Boot Ratio (Quarkus / SB) Framework version 3.17.5 3.4.1 3.4.1 3.17.5 3.4.1 Build time (s) 12.88 7.82 164.75% 16.88 76.34% 215.81% 172.8 283.39 60.97% Av. RSS after startup (MB) 190.6 297.92 63.98% 293.88 64.86% 98.64% 71.09 206.47 34.43% Av. time to 1st req (ms) 4,684 8,885.33 52.72% 7,940 58.99% 89.36% 132.67 924.33 14.35% Av. RSS after 1st req (MB) 241.29 344.86 69.97% 335.76 71.86% 97.36% 78.95 208.87 37.8% Av. throughput (req/sec) 30,769.99 6,713.7 458.32% 6,750.04 455.85% 100.54% 16,189.26 6,024.12 268.74% Max throughput density (req/sec/MB) 41.77 8.58 487.08% 8.75 477.32% 102.04% 61.1 14.33 426.38% Build RSS (GB) 6.0 7.83 76.63% Binary Size (MB) 90.47 156.58 57.78% Don’t Take My Word For It! Some real numbers (8 cores, 14GB RAM, GraalVM CE 21.0.2)
50. @edeandrea JVM Native Quarkus Spring Boot Ratio (Quarkus / SB) Spring Boot (w/AOT) Ratio (Quarkus / SB AOT) Ratio (SB AOT / SB) Quarkus Spring Boot Ratio (Quarkus / SB) Framework version 3.17.5 3.4.1 3.4.1 3.17.5 3.4.1 Build time (s) 12.88 7.82 164.75% 16.88 76.34% 215.81% 172.8 283.39 60.97% Av. RSS after startup (MB) 190.6 297.92 63.98% 293.88 64.86% 98.64% 71.09 206.47 34.43% Av. time to 1st req (ms) 4,684 8,885.33 52.72% 7,940 58.99% 89.36% 132.67 924.33 14.35% Av. RSS after 1st req (MB) 241.29 344.86 69.97% 335.76 71.86% 97.36% 78.95 208.87 37.8% Av. throughput (req/sec) 30,769.99 6,713.7 458.32% 6,750.04 455.85% 100.54% 16,189.26 6,024.12 268.74% Max throughput density (req/sec/MB) 41.77 8.58 487.08% 8.75 477.32% 102.04% 61.1 14.33 426.38% Build RSS (GB) 6.0 7.83 76.63% Binary Size (MB) 90.47 156.58 57.78% Don’t Take My Word For It! Some real numbers (8 cores, 14GB RAM, GraalVM CE 21.0.2)
51. @edeandrea JVM Native Quarkus Spring Boot Ratio (Quarkus / SB) Spring Boot (w/AOT) Ratio (Quarkus / SB AOT) Ratio (SB AOT / SB) Quarkus Spring Boot Ratio (Quarkus / SB) Framework version 3.17.5 3.4.1 3.4.1 3.17.5 3.4.1 Build time (s) 12.88 7.82 164.75% 16.88 76.34% 215.81% 172.8 283.39 60.97% Av. RSS after startup (MB) 190.6 297.92 63.98% 293.88 64.86% 98.64% 71.09 206.47 34.43% Av. time to 1st req (ms) 4,684 8,885.33 52.72% 7,940 58.99% 89.36% 132.67 924.33 14.35% Av. RSS after 1st req (MB) 241.29 344.86 69.97% 335.76 71.86% 97.36% 78.95 208.87 37.8% Av. throughput (req/sec) 30,769.99 6,713.7 458.32% 6,750.04 455.85% 100.54% 16,189.26 6,024.12 268.74% Max throughput density (req/sec/MB) 41.77 8.58 487.08% 8.75 477.32% 102.04% 61.1 14.33 426.38% Build RSS (GB) 6.0 7.83 76.63% Binary Size (MB) 90.47 156.58 57.78% Don’t Take My Word For It! Some real numbers (8 cores, 14GB RAM, GraalVM CE 21.0.2)
52. @edeandrea A Real Example
53. @edeandrea 🤘 Spring Boot 3 application contents: ◦ 39,533 classes ◦ 84,792 fields ◦ 285,156 methods ◦ Reflection: ◦ 11,143 classes ◦ 1,637 fields ◦ 13,032 methods A Real Example
54. @edeandrea 🤘 Spring Boot 3 application contents: ◦ 39,533 classes ◦ 84,792 fields ◦ 285,156 methods ◦ Reflection: ◦ 11,143 classes ◦ 1,637 fields ◦ 13,032 methods 🤘 Quarkus 3 application contents: ◦ 23,814 classes ◦ 47,373 fields ◦ 190,150 methods ◦ Reflection: ◦ 6,691 classes ◦ 202 fields ◦ 4,971 methods A Real Example
55. @edeandrea 🤘 Spring Boot 3 application contents: ◦ 39,533 classes ◦ 84,792 fields ◦ 285,156 methods ◦ Reflection: ◦ 11,143 classes ◦ 1,637 fields ◦ 13,032 methods 🤘 Quarkus 3 application contents: ◦ 23,814 classes ◦ 47,373 fields ◦ 190,150 methods ◦ Reflection: ◦ 6,691 classes ◦ 202 fields ◦ 4,971 methods 🤘 The Quarkus application has: ◦ 15,719 (40%) less classes ◦ 37,419 (44%) less fields ◦ 87,519 (33%) less methods ◦ 4,452 (40%) less classes using reflection ◦ 1,435 (88%) less fields using reflection ◦ 8,061 (62%) less methods using reflection A Real Example
56. @edeandrea Don’t Take Our Word For It! https://www.redhat.com/en/resources/etat-du-valais-customer-case-study
57. @edeandrea https://www.youtube.com/live/YNxz3KaTz5Y
58. @edeandrea https://quarkus.io/blog/tag/user-story Don’t Take Our Word For It!
59. @edeandrea https://quarkus.io/blog/tag/user-story Don’t Take Our Word For It!
60. @edeandrea https://www.redhat.com/en/resources/greener-java-applications-detail Setup 🤘 AWS (us-east-1) 🤘 SLA > 99% 🤘 800 req/sec over 20 days 🤘 50% load The Cost / Carbon Impact
61. @edeandrea
62. @edeandrea https://github.com/quarkusio/quarkus-super-heroes https://quarkus.io/quarkus-workshops/super-heroes Quarkus Superheroes
63. @edeandrea Thank You!