Aklakan commented on PR #3480: URL: https://github.com/apache/jena/pull/3480#issuecomment-3361459739
So benchmark results with this PR are pretty much in the same magnitude. The benchmark adds the "type Triples" that are mandatory with jena-5.5.0. But when omitting those, the current version still returns the usual results whereas jena-5.5.0 won't. ``` Benchmark (p1_numGeometryMixes) (p2_queryId) (p3_inferenceMode) (p4_jenaVersion) Mode Cnt Score Error Units BenchmarkGeoQueries.run 3000 q1 off current avgt 5 1.134 ± 0.199 s/op BenchmarkGeoQueries.run 3000 q1 off 5.5.0 avgt 5 0.964 ± 0.252 s/op BenchmarkGeoQueries.run 3000 q1 virtual current avgt 5 1.380 ± 0.192 s/op BenchmarkGeoQueries.run 3000 q1 virtual 5.5.0 avgt 5 1.482 ± 0.494 s/op BenchmarkGeoQueries.run 3000 q1 materialized current avgt 5 0.759 ± 0.922 s/op BenchmarkGeoQueries.run 3000 q1 materialized 5.5.0 avgt 5 0.897 ± 0.508 s/op BenchmarkGeoQueries.run 3000 q2 off current avgt 5 0.027 ± 0.004 s/op BenchmarkGeoQueries.run 3000 q2 off 5.5.0 avgt 5 0.029 ± 0.006 s/op BenchmarkGeoQueries.run 3000 q2 virtual current avgt 5 0.034 ± 0.012 s/op BenchmarkGeoQueries.run 3000 q2 virtual 5.5.0 avgt 5 0.087 ± 0.075 s/op BenchmarkGeoQueries.run 3000 q2 materialized current avgt 5 0.018 ± 0.006 s/op BenchmarkGeoQueries.run 3000 q2 materialized 5.5.0 avgt 5 0.021 ± 0.028 s/op ``` Unrelated fixes: * `DatasetGraphRDFS.find()` did not return inferences. * Credits to @SimonBin for spotting the `forEachRemaining` resource leaks - interestingly `forEach` does resource handling :exploding_head: -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
