The update of kubernetes client from 6.13.4 to 7.0.1 has significant impact in build-time RSS usage and binary size

[zakkak]

Describe the bug

#45259 introduced the following changes in build-time metrics:

Quarkus version	cd9d9f0	72dd7dc	Diff (%)
Total Size (MB)	124.1213	138.0315	11.21%
Reflection stats	---	---	---
Classes (Reflection)	10118	10597	4.73%
Methods (Reflection)	53379	58439	9.48%
Fields (Reflection)	11850	13283	12.09%
Reachability stats	---	---	---
Classes (Reachability)	20436	20992	2.72%
Methods (Reachability)	169566	179854	6.07%
Fields (Reachability)	41992	44702	6.45%
JNI stats	---	---	---
Classes (JNI)	62	62	0.00%
Methods (JNI)	55	55	0.00%
Fields (JNI)	67	67	0.00%
Classpath stats	---	---	---
Classes (classpath)	26733	26067	-2.49%
Methods (classpath)	274662	286739	4.40%
Fields (classpath)	55286	58010	4.93%
Build Perf stats	---	---	---
Total Build Time (s)	223.4698	245.1337	9.69%
GC Time (s)	25.1700	27.0010	7.27%
CPU Load	3.4226	3.3848	-1.10%
Peak RSS (GB)	4.4929	4.9672	10.56%
Available RAM (GB)	15.6065	15.6065	0.00%

cc @metacosm @manusa

Expected behavior

No response

Actual behavior

No response

How to Reproduce?

Build openshift-client integration test with both versions.

Output of uname -a or ver

No response

Output of java -version

No response

Mandrel or GraalVM version (if different from Java)

No response

Quarkus version or git rev

No response

Build tool (ie. output of mvnw --version or gradlew --version)

No response

Additional information

No response

[quarkus-bot]

/cc @Karm (native-image), @galderz (native-image), @geoand (kubernetes), @iocanel (kubernetes)

[geoand]

cc @manusa @metacosm

[manusa]

The Fabric8 Kubernetes Client v7.0 major version bump included among other things updated versions of the OpenShift model types.

I suspect that the deltas in size (MiB), classes, methods, and fields, are due to these updated types and not from any other sort of regression or other problem.

I'm not sure if you're able to share the break-down report on module sizes, but I'd say that probably kubernetes-model-core should have either remained stable or event shrunk while some of the openshift-model-xxx might have expanded.

[zakkak]

I'm not sure if you're able to share the break-down report on module sizes, but I'd say that probably kubernetes-model-core should have either remained stable or event shrunk while some of the openshift-model-xxx might have expanded.

Unfortunately we don't keep this kind of detailed info. One would need to run the test with both versions to confirm this.