Comparison of OpenJPA with PostgreSQL server vs DataNucleus with H2 embedded

Each of the following tables focuses on a specific database operation, where the last table presents average results comparison.

Speed comparison of JPA database persistence operations (normalized score, higher is better)

Transaction Size =>	Few Entities		Many Entities		Average Score
	OpenJPA PostgreSQL server	DataNucleus H2 embedded	OpenJPA PostgreSQL server	DataNucleus H2 embedded	OpenJPA PostgreSQL server	DataNucleus H2 embedded
Basic Person Test	5.2	19.2	4.8	8.0	5.0	13.6
Element Collection Test	2.8	6.6	2.2	2.8	2.5	4.7
Inheritance Test	4.6	15.6	3.2	4.4	3.9	10.0
Indexing Test	7.4	24.4	6.9	11.8	7.1	18.1
Graph (Binary Tree) Test	0.98	9.2	0.81	6.2	0.89	7.7
Multithreading Test	23.4	34.9	9.1	7.1	16.2	21.0
All Tests	7.4	18.3	4.5	6.7	5.9	12.5

The results above show that in general DataNucleus with H2 embedded is more efficient than OpenJPA with PostgreSQL server in persisting JPA entity objects to the database. Comparing the normalized speed of OpenJPA with PostgreSQL database server (5.9) to the normalized speed of DataNucleus with H2 embedded database (12.5) reveals that in these tests, DataNucleus with H2 embedded is 2.1 times faster than OpenJPA with PostgreSQL server.

A large performance gap has been detected when using graphs of objects with small transaction size. Comparing the normalized speed of OpenJPA with PostgreSQL database server (0.98) to the normalized speed of DataNucleus with H2 embedded database (9.2) reveals that in that case, DataNucleus with H2 embedded is 9.4 times faster than OpenJPA with PostgreSQL server.

Speed comparison of JPA database retrieval operations (normalized score, higher is better)

Retrieval Size =>	Few Entities		Many Entities		Average Score
	OpenJPA PostgreSQL server	DataNucleus H2 embedded	OpenJPA PostgreSQL server	DataNucleus H2 embedded	OpenJPA PostgreSQL server	DataNucleus H2 embedded
Basic Person Test	5.8	13.2	16.1	13.5	10.9	13.3
Element Collection Test	0.013	6.4	8.7	8.9	4.4	7.7
Inheritance Test	5.2	12.2	17.6	12.7	11.4	12.4
Indexing Test	4.8	8.1	18.7	16.5	11.8	12.3
Graph (Binary Tree) Test	0.27	10.6	0.35	20.9	0.31	15.8
Multithreading Test	11.9	13.3	23.3	9.9	17.6	11.6
All Tests	4.7	10.6	14.1	13.7	9.4	12.2

The results above show that in general DataNucleus with H2 embedded is slightly more efficient than OpenJPA with PostgreSQL server in retrieving JPA entity objects from the database.

A huge performance gap has been detected when using JPA element collections with small retrieval size. Comparing the normalized speed of OpenJPA with PostgreSQL database server (0.013) to the normalized speed of DataNucleus with H2 embedded database (6.4) reveals that in that case, DataNucleus with H2 embedded is 492 times faster than OpenJPA with PostgreSQL server.

On the other hand, DataNucleus with H2 embedded is slower, for instance, when using multithreading with large retrieval size. Comparing the normalized speed of DataNucleus with H2 embedded database (9.9) to the normalized speed of OpenJPA with PostgreSQL database server (23.3) reveals that in that case, DataNucleus with H2 embedded is 2.4 times slower than OpenJPA with PostgreSQL server.

Speed comparison of JPA database query operations (normalized score, higher is better)

Retrieval Size =>	Few Entities		Many Entities		Average Score
	OpenJPA PostgreSQL server	DataNucleus H2 embedded	OpenJPA PostgreSQL server	DataNucleus H2 embedded	OpenJPA PostgreSQL server	DataNucleus H2 embedded
Basic Person Test	failed	4.7	failed	1.4	failed	3.1
Element Collection Test	failed	5.0	failed	2.4	failed	3.7
Inheritance Test	failed	3.8	failed	1.6	failed	2.7
Indexing Test	failed	0.0041	failed	1.8	failed	0.91
Multithreading Test	failed	1.2	failed	0.79	failed	1.0
All Tests	failed	3.0	failed	1.6	failed	2.3

OpenJPA with PostgreSQL server has failed in 10 tests (see exceptions).

Speed comparison of JPA database update operations (normalized score, higher is better)

Transaction Size =>	Few Entities		Many Entities		Average Score
	OpenJPA PostgreSQL server	DataNucleus H2 embedded	OpenJPA PostgreSQL server	DataNucleus H2 embedded	OpenJPA PostgreSQL server	DataNucleus H2 embedded
Basic Person Test	4.9	10.0	3.9	3.9	4.4	6.9
Element Collection Test	0.033	7.6	4.6	3.9	2.3	5.8
Inheritance Test	4.8	9.4	4.9	4.4	4.9	6.9
Indexing Test	5.1	10.6	6.0	5.9	5.6	8.2
Graph (Binary Tree) Test	0.57	5.9	0.31	3.7	0.44	4.8
Multithreading Test	22.2	failed	7.4	failed	14.8	failed
All Tests	6.3	8.7	4.5	4.4	5.4	6.5

DataNucleus with H2 embedded has failed in 2 tests (see exceptions).

The results above show that in general DataNucleus with H2 embedded is slightly more efficient than OpenJPA with PostgreSQL server in updating JPA entity objects in the database.

A huge performance gap has been detected when using JPA element collections with small transaction size. Comparing the normalized speed of OpenJPA with PostgreSQL database server (0.033) to the normalized speed of DataNucleus with H2 embedded database (7.6) reveals that in that case, DataNucleus with H2 embedded is 230 times faster than OpenJPA with PostgreSQL server.

Speed comparison of JPA database removal operations (normalized score, higher is better)

Transaction Size =>	Few Entities		Many Entities		Average Score
	OpenJPA PostgreSQL server	DataNucleus H2 embedded	OpenJPA PostgreSQL server	DataNucleus H2 embedded	OpenJPA PostgreSQL server	DataNucleus H2 embedded
Basic Person Test	4.6	4.2	5.3	1.3	5.0	2.8
Element Collection Test	0.032	4.9	2.8	1.6	1.4	3.2
Inheritance Test	4.3	4.0	4.6	1.3	4.4	2.7
Indexing Test	8.0	7.5	7.1	1.9	7.5	4.7
Graph (Binary Tree) Test	0.53	5.6	0.55	4.1	0.54	4.8
Multithreading Test	17.4	failed	9.1	failed	13.3	failed
All Tests	5.8	5.2	4.9	2.0	5.4	3.6

DataNucleus with H2 embedded has failed in 2 tests (see exceptions).

The results above show that in general OpenJPA with PostgreSQL server is more efficient than DataNucleus with H2 embedded in deleting JPA entity objects from the database.

A large performance gap has been detected when using simple basic entities with large transaction size. Comparing the normalized speed of DataNucleus with H2 embedded database (1.3) to the normalized speed of OpenJPA with PostgreSQL database server (5.3) reveals that in that case, OpenJPA with PostgreSQL server is 4.1 times faster than DataNucleus with H2 embedded.

On the other hand, OpenJPA with PostgreSQL server is slower, for instance, when using JPA element collections with small transaction size. Comparing the normalized speed of OpenJPA with PostgreSQL database server (0.032) to the normalized speed of DataNucleus with H2 embedded database (4.9) reveals that in that case, OpenJPA with PostgreSQL server is 153 times slower than DataNucleus with H2 embedded.

Comparison of JPA/Database speed - the averages (normalized score, higher is better)

Transaction/Retrieval Size	Few Entities		Many Entities		Average Score
	OpenJPA PostgreSQL server	DataNucleus H2 embedded	OpenJPA PostgreSQL server	DataNucleus H2 embedded	OpenJPA PostgreSQL server	DataNucleus H2 embedded
Basic Person Test	5.1	10.3	7.5	5.6	6.3	7.9
Element Collection Test	0.72	6.1	4.6	3.9	2.6	5.0
Inheritance Test	4.7	9.0	7.6	4.9	6.2	6.9
Indexing Test	6.3	10.1	9.7	7.6	8.0	8.9
Graph (Binary Tree) Test	0.59	7.8	0.50	8.7	0.55	8.3
Multithreading Test	18.7	16.5	12.2	5.9	15.5	11.2
All Tests	6.0	9.6	7.0	6.0	6.5	7.8

The results above show that in general DataNucleus with H2 embedded is slightly more efficient than OpenJPA with PostgreSQL server in performing JPA database operations.

A huge performance gap has been detected when using graphs of objects with large transaction/retrieval size. Comparing the normalized speed of OpenJPA with PostgreSQL database server (0.50) to the normalized speed of DataNucleus with H2 embedded database (8.7) reveals that in that case, DataNucleus with H2 embedded is 17.4 times faster than OpenJPA with PostgreSQL server.

On the other hand, DataNucleus with H2 embedded is slower, for instance, when using multithreading with large transaction/retrieval size. Comparing the normalized speed of DataNucleus with H2 embedded database (5.9) to the normalized speed of OpenJPA with PostgreSQL database server (12.2) reveals that in that case, DataNucleus with H2 embedded is 2.1 times slower than OpenJPA with PostgreSQL server.

JPA Performance Benchmark (JPAB)