Six and a half ridiculous things to do with Quarkus

Let’s abuse the tools! Everyone knows Quarkus is computationally efficient, expressive, and rock-solid for production. But did you know that we can use Quarkus efficiency to build applications that shouldn’t go anywhere *near* production? In this demo-driven session, Holly will put the joy into “developer joy”. She’ll show you all sorts of things you can do with Quarkus that you probably shouldn’t:

- Build an LLM-powered app that’s *guaranteed* to hallucinate, because you can do more than you think with guardrails, and truth is so boring
- Write your business code in rockstarlang, because everything should be a hair metal ballad
- Use Minecraft as your observability client, because the LGTM stack doesn’t have enough explosions
- Write a CLI for generating memes faster, because everything is better on the command-line
- Benchmark an application against a grapefruit, because metric units aren’t tasty

Business value? Learning? If you insist. As well as absurd demos, you’ll leave this session with a deeper understanding of how to get the most out of Quarkus and Java. There will be new Java 25 language features, Quarkus best practices, powerful integrations, and nifty use cases alongside the silly explosions and grapefruit.

Transcript

1. 6 ½ ridiculous things to do with Quarkus Holly Cummins IBM
3. quarkus is used in production by:
4. quarkus is used in production by: • telcos
5. quarkus is used in production by: • telcos • retailers
6. quarkus is used in production by: • telcos • retailers • banks
7. quarkus is used in production by: • telcos • retailers • banks BORING!
9. https://github.com/holly-cummins/gen-alpha-iser
10. 6 ½ skibidi Ohio things to do with Quarkus Holly Cummins IBM + Red Hat
11. was this an easy demo?
12. was this an easy demo? super easy
13. tell your boss you learned… @ServerResponseFilter
15. https://github.com/phillip-kruger/quarkus-jokes
16. was this an easy demo?
17. was this an easy demo? yes, phillip krüger wrote all of it :)
18. tell your boss you learned… the dev ui now has assistants to help you write + debug code
27. https://github.com/quarkiverse/quarkus-bon-jova-rockstar
28. https://github.com/quarkiverse/quarkus-bon-jova-rockstar
29. was this an easy demo?
30. was this an easy demo?
31. was this an easy demo? -friends don’t let friends write compilers
32. was this an easy demo? -friends don’t let friends write compilers -bytecode generation :(
33. was this an easy demo? -friends don’t let friends write compilers -bytecode generation :( -parsing :( :(
34. tell your boss you learned… -the quarkiverse is very welcoming to new extensions -writing a compiler is a bad idea (but if you do go there, use antlr)
37. spec
38. spec (by dylan beattie)
39. spec (by dylan beattie) cost: 5 beers
40. spec (by dylan beattie) cost: 5 beers time: 1 night
41. spec (by dylan beattie) cost: 5 beers time: 1 night implementation
42. spec (by dylan beattie) cost: 5 beers time: 1 night implementation cost: $0
43. spec (by dylan beattie) cost: 5 beers time: 1 night implementation cost: $0 time: 3 weeks
44. spec spec (by dylan beattie) cost: 5 beers time: 1 night implementation cost: $0 time: 3 weeks
45. spec none, obvs spec (by dylan beattie) cost: 5 beers time: 1 night implementation cost: $0 time: 3 weeks
46. spec none, obvs spec (by dylan beattie) cost: 5 beers time: 1 night implementation cost: $0 time: 3 weeks
47. spec none, obvs implementation spec (by dylan beattie) cost: 5 beers time: 1 night implementation cost: $0 time: 3 weeks
48. spec none, obvs implementation cost: $36,312.03 spec (by dylan beattie) cost: 5 beers time: 1 night implementation cost: $0 time: 3 weeks
49. spec none, obvs implementation cost: $36,312.03 time: 3 months spec (by dylan beattie) cost: 5 beers time: 1 night implementation cost: $0 time: 3 weeks
50. gen ai: what could possibly go wrong?
52. https://www.computerworld.com/article/4059383/openai-admits-ai-hallucinations-are-mathematically-inevitable-not-just-engineering-flaws.html rate of hallucination when summarizing public information:
53. o1: 16% https://www.computerworld.com/article/4059383/openai-admits-ai-hallucinations-are-mathematically-inevitable-not-just-engineering-flaws.html rate of hallucination when summarizing public information:
54. o1: 16% o3: 33% https://www.computerworld.com/article/4059383/openai-admits-ai-hallucinations-are-mathematically-inevitable-not-just-engineering-flaws.html rate of hallucination when summarizing public information:
55. o1: 16% o3: 33% o4-mini: 48% https://www.computerworld.com/article/4059383/openai-admits-ai-hallucinations-are-mathematically-inevitable-not-just-engineering-flaws.html rate of hallucination when summarizing public information:
56. solution: guardrails
58. was this an easy demo?
59. was this an easy demo?
60. was this an easy demo? -adding guardrails is simple
61. was this an easy demo? -adding guardrails is simple -but … the demo needed
62. was this an easy demo? -adding guardrails is simple -but … the demo needed -a prompt the llm would hallucinate about 50% of the time
63. was this an easy demo? -adding guardrails is simple -but … the demo needed -a prompt the llm would hallucinate about 50% of the time -a reliable guardrail
64. gemma3 - 1 b parameters how many bs in blueberry?
65. "Okay, let’s tackle this! As of today, October 1st, 2025, there are approximately **12** “BS”s in the word “blueberry.” It’s a surprisingly common phenomenon! 😊 Do you want to explore why this happens?" gemma3 - 1 b parameters how many bs in blueberry?
66. llama3.2- 1 b parameters how many bs in blueberry?
67. "There are no \"bs\" in Blueberry. It's spelled B-E-L-U-E-B-R-I-E-R-Y." llama3.2- 1 b parameters how many bs in blueberry?
68. deepseekr1- 1.5 b parameters when did brexit happen?
69. <think> Hmm, the user is asking about something related to 20-30-0-3 but brexit, probably typo or20-10- something, correction. User meant brexit or2-1- meaning Brexit. Let me confirm that. User is20- user2 is inquiring about Brexit. UK leaving EU thing. But what time period20 2020-3 might be typo for20- 20 twenty021203. 20-20 or something. Perhaps20-202201. UK left. UK left EU. UK left20 on01 Jan201202022-021201021 I think22. Wait no1. Let me212012 January22202021. So January. Yes2 user might be21 probably20 asking about that. User22. Better2 just confirm. Brexit201 20 answer. Say20 simply22 date2. Not much else0 else0. Keep2 simple1 short1 friendly2. Okay2 polite2 user. </think> It seems you're asking about the date the Brexit, right? The UK left EU on 29th January21st of February2020221 deepseekr1- 1.5 b parameters when did brexit happen?
70. the value of a second opinion? chatgpt.com
71. the value of a second opinion? chatgpt.com the statement we’re validating
72. the value of a second opinion? chatgpt.com the statement we’re validating “oh yes, definitely true”
73. the value of a second opinion? chatgpt.com the statement we’re validating “oh yes, definitely true” “what answer would you give?”
74. the value of a second opinion? chatgpt.com the statement we’re validating “oh yes, definitely true” “what answer would you give?” but if you knew, why did you … ?
75. tell your boss you learned… •model choice makes a big difference •“more LLM” may not be the right solution to “LLM” •simple guardrails are not great at detecting hallucinations
76. observability
78. user code
79. user code quarkus framework
80. application user code quarkus framework
81. application user code quarkus framework extensions
82. application user code quarkus framework extensions
83. application user code quarkus framework client extensions
84. application user code quarkus framework client extensions
85. application user code quarkus framework client extensions
86. application user code quarkus framework client extensions
87. application user code quarkus framework client extensions observability stack
88. application user code quarkus framework client metrics extensions observability stack
89. application user code quarkus framework client metrics traces extensions observability stack
90. application user code quarkus framework client metrics traces logs extensions observability stack
91. application user code quarkus framework observability client client metrics traces logs extensions observability stack
92. application user code quarkus framework observability client client metrics traces logs extensions observability stack
93. application user code quarkus framework observability client client dev service (testcontainers) extensions observability stack
95. application user code quarkus framework client extensions
96. application user code quarkus framework client extensions
97. application user code quarkus framework client extensions minecraft server
98. application user code quarkus framework minecraft client client extensions minecraft server
99. application user code quarkus framework minecraft client client extensions minecraft server
100. application user code quarkus framework minecraft client client extensions minecraft server mod
101. application user code quarkus framework minecraft client client dev service (testcontainers) extensions minecraft server mod
102. https://github.com/quarkiverse/quarkus-observability-minecraft
104. tell your boss you learned… -you can do a lot with interceptors -dev services are awesome
106. https://github.com/quarkiverse/quarkus-observability-minecraft
107. https://github.com/quarkiverse/quarkus-observability-minecraft
108. was this an easy demo?
109. was this an easy demo?
110. was this an easy demo? -extension was easy, minecraft mod was hard
111. was this an easy demo? -extension was easy, minecraft mod was hard -minecraft APIs
112. was this an easy demo? -extension was easy, minecraft mod was hard -minecraft APIs -lack documentation
113. was this an easy demo? -extension was easy, minecraft mod was hard -minecraft APIs -lack documentation -change between minor versions
114. was this an easy demo? -extension was easy, minecraft mod was hard -minecraft APIs -lack documentation -change between minor versions -client-server sync of injected behaviour is hard
115. tell your boss you learned… -typesafe objects are better than a chat box -… but small models may struggle with structured data -gen ai adds latency
116. how quick can quarkus go?
118. thanks to markus eisele for the graphics code https://github.com/holly-cummins/meme-generator
119. was this an easy demo?
120. was this an easy demo?
121. was this an easy demo? -writing the cli was easy -native compilation was easy -native compilation + awt + mac is very hard
123. i tried:
124. i tried: native awt on mac
125. i tried: native awt on mac liberica vm mac + tuning of build-time and runtime classes
126. i tried: native awt on mac liberica vm mac + tuning of build-time and runtime classes graalvm community container
127. i tried: native awt on mac liberica vm mac + tuning of build-time and runtime classes graalvm community container mandrel container
128. tell your boss you learned… •picocli + quarkus is fire •native is clever tech •…but native has dragons •only go native if you actually need it
129. is quarkus fast?
131. was this an easy demo?
132. was this an easy demo? i used julien ponge’s scripts and eric deandrea’s code
133. https://github.com/quarkusio/spring-quarkus-perf-comparison
134. don’t do perf testing on a laptop don’t drive load from the machine being measured
135. here’s what we see in our controlled lab 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 3.22.3 3.4.5 3.4.5 3.22.3 3.4.5 Build time (s) 13.15 7.60 172.95% 17.02 77.26% 223.85% 178.99 279.45 64.05% Av. RSS after startup (MB) 192.31 298.09 64.51% 290.40 66.22% 97.42% 71.22 205.74 34.62% Av. time to 1st req (ms) 4652.67 8770.33 53.05% 8180.67 56.87% 93.28% 135.67 916.67 14.80% Av. RSS after 1st req (MB) 244.61 339.52 27.05% 331.78 73.73% 97.72% 79.12 208.21 38.00% Av. throughput (req/sec) 30,397.95 6,495.20 468.01% 6,427.64 472.93% 98.96% 16,753.94 6,146.27 272.59% Max throughput density (req/sec/MB) 35.79 7.95 450.50% 7.89 453.88% 99.26% 59.82 15.94 375.27% Build RSS (GB) 5.62 157.04 57.92% Binary Size (MB) 90.95 157.04 57.92% https://github.com/edeandrea/quarkus-for-spring-devs-examples/tree/quarkus-for-spring-devs-talk-sb3 8 cores, 14GB RAM, GraalVM CE 21.0.2
136. it’s not really about throughput
137. it’s not really about speed
138. it’s about hardware
139. … and hardware is money
140. spring boot quarkus
141. spring boot quarkus 40 586 req/s
142. spring boot quarkus 40 586 req/s 17 209 req/s
143. spring boot quarkus 40 586 req/s 17 209 req/s 1 342 errors
144. spring boot quarkus 40 586 req/s 17 209 req/s 1 342 errors 2 vCPUs 0.5 GiB
145. spring boot quarkus t3.nano $0.0052 40 586 req/s 17 209 req/s 1 342 errors 2 vCPUs 0.5 GiB
146. spring boot quarkus t3.nano $0.0052 40 586 req/s 17 209 req/s 1 342 errors 2 vCPUs 2 GiB 2 vCPUs 0.5 GiB
147. spring boot quarkus t3.nano $0.0052 t3.small $0.0208 40 586 req/s 17 209 req/s 1 342 errors 2 vCPUs 2 GiB 2 vCPUs 0.5 GiB
148. spring boot quarkus t3.nano $0.0052 t3.small $0.0208 40 586 req/s 17 209 req/s 1 342 errors $3.87 a month 2 vCPUs 2 GiB 2 vCPUs 0.5 GiB
149. spring boot quarkus t3.nano $0.0052 t3.small $0.0208 40 586 req/s 17 209 req/s 1 342 errors $15.47 a month $3.87 a month 2 vCPUs 2 GiB 2 vCPUs 0.5 GiB
150. spring boot quarkus t3.nano $0.0052 t3.small $0.0208 40 586 req/s 17 209 req/s 1 342 errors $15.47 a month $3.87 a month x2 2 vCPUs 2 GiB 2 vCPUs 0.5 GiB
151. spring boot quarkus t3.nano $0.0052 t3.small $0.0208 40 586 req/s 17 209 req/s 1 342 errors $15.47 a month $3.87 a month $30.94 a month x2 2 vCPUs 2 GiB 2 vCPUs 0.5 GiB
152. tell your boss you learned… -quarkus uses 2 - 10x fewer compute resources than spring -quarkus is 2 - 10x cheaper to run than spring
153. it’s also about energy
154. it’s also about energy BORING!
159. 9000 potatoes
161. 0.0007 W
163. t3.nano t3.small 2x t3.small Watts per hour 12.5 14 28 Lemons per hour 17,857 20,000 40,000
164. spring boot 40,000 lemons/hour quarkus 17,857 lemons/hour
166. jbang wrk2@hyperfoil -t2 -c100 -d30s -R 200 -- latency http://localhost:8080/fruits sudo powermetrics --samplers cpu_power,tasks --show- process-samp-norm --show-process-gpu --show-usage- summary -i 0 -o output.txt
167. https://github.com/quarkusio/spring-quarkus-perf-comparison
168. https://github.com/quarkusio/spring-quarkus-perf-comparison
169. tell your boss you learned… -use wrk2 or hyperfoil for load testing -use powermetrics to measure energy by pid on mac
170. was this an easy demo?
171. was this an easy demo?
172. was this an easy demo? -bringing 77 ,857 lemons to devoxx turned out to be hard
173. was this an easy demo? -bringing 77 ,857 lemons to devoxx turned out to be hard -no time for powermetrics demo
174. what have we done?
175. what have we done? 1. translated output with a filter
176. what have we done? 1. translated output with a filter 2. transformed code with an extension
177. what have we done? 1. translated output with a filter 2. transformed code with an extension 3. compiled rock ballads to bytecode
178. what have we done? 1. translated output with a filter 2. transformed code with an extension 3. compiled rock ballads to bytecode 4. amplified hallucinations
179. what have we done? 1. translated output with a filter 2. transformed code with an extension 3. compiled rock ballads to bytecode 4. amplified hallucinations 5. used minecraft as an observability client
180. what have we done? 1. translated output with a filter 2. transformed code with an extension 3. compiled rock ballads to bytecode 4. amplified hallucinations 5. used minecraft as an observability client 6. made memes with a cli
181. what have we done? 1. translated output with a filter 2. transformed code with an extension 3. compiled rock ballads to bytecode 4. amplified hallucinations 5. used minecraft as an observability client 6. made memes with a cli 7 . confirmed quarkus does a lot with very few resources
182. gen-alpha translation services: asher hodkinson i’ll help you but you need to put me in the closing credits
183. https://hollycummins.com/quarkus-ridiculous-things-devoxx-be/