Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

New Quarkus version #2

Open
franz1981 opened this issue Mar 2, 2024 · 4 comments
Open

New Quarkus version #2

franz1981 opened this issue Mar 2, 2024 · 4 comments

Comments

@franz1981
Copy link

Hi @novoj

It is some time I am not running your benchmark with latest quarkus version, let me know if you can give it a shot, so I can verify if I found anything strange

@novoj
Copy link
Contributor

novoj commented Mar 3, 2024

I will give it a shot

novoj added a commit that referenced this issue Mar 3, 2024
@novoj
Copy link
Contributor

novoj commented Mar 3, 2024

Original results based on Quarkus 3.0.4.Final (baseline):

Benchmark                                                      Mode  Cnt      Score      Error  Units
ServersThroughputBenchmark.graphQLApiEchoQuery_QuarkusServer  thrpt    5  21830.819 ± 1492.236  ops/s
ServersLatencyBenchmark.graphQLApiEchoQuery_QuarkusServer      avgt    5    281.107 ±   20.137  us/op

Native Quarkus:

Benchmark                                                      Mode  Cnt      Score     Error  Units
ServersThroughputBenchmark.graphQLApiEchoQuery_QuarkusServer  thrpt    5  15720.155 ± 799.392  ops/s
ServersLatencyBenchmark.graphQLApiEchoQuery_QuarkusServer      avgt    5    392.781 ±  38.727  us/op

New results with version 3.8.1:

Benchmark                                                      Mode  Cnt      Score      Error  Units
ServersThroughputBenchmark.graphQLApiEchoQuery_QuarkusServer  thrpt    5  22147.961 ± 1692.798  ops/s
ServersLatencyBenchmark.graphQLApiEchoQuery_QuarkusServer      avgt    5    281.611 ±   19.583  us/op

Native Quarkus:

Benchmark                                                      Mode  Cnt      Score      Error  Units
ServersThroughputBenchmark.graphQLApiEchoQuery_QuarkusServer  thrpt    5  14901.192 ± 2034.417  ops/s
ServersLatencyBenchmark.graphQLApiEchoQuery_QuarkusServer      avgt    5    412.826 ±   29.148  us/op

I think the difference is negligible.

@novoj
Copy link
Contributor

novoj commented Mar 3, 2024

Tests were run as follows:

java -jar performance_tests/target/benchmarks.jar .*Quarkus.* -wi 1 -i 5 -f 1

@franz1981
Copy link
Author

Yep, thanks! I need to check what's going on with the graphQL lib at this point; the fact that native is not half of the performance of JVM mode is sign of reflective operations which are likely better behaving in native mode (which is expected) - but it still means there are lot of hanging fruits there

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants