Comparison of OpenJPA with Derby server vs DataNucleus with MySQL server

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 Derby server	DataNucleus MySQL server	OpenJPA Derby server	DataNucleus MySQL server	OpenJPA Derby server	DataNucleus MySQL server
Basic Person Test	3.1	2.6	2.7	1.4	2.9	2.0
Element Collection Test	2.0	1.0	1.7	0.48	1.8	0.76
Inheritance Test	3.0	2.5	2.3	1.3	2.6	1.9
Indexing Test	4.7	4.0	4.1	2.5	4.4	3.3
Graph (Binary Tree) Test	0.71	1.1	0.64	0.99	0.67	1.1
Multithreading Test	6.0	6.2	3.8	2.5	4.9	4.4
All Tests	3.3	2.9	2.5	1.5	2.9	2.2

The results above show that in general OpenJPA with Derby server is more efficient than DataNucleus with MySQL server in persisting JPA entity objects to the database.

A large performance gap has been detected when using JPA element collections with large transaction size. Comparing the normalized speed of DataNucleus with MySQL database server (0.48) to the normalized speed of OpenJPA with Derby database server (1.7) reveals that in that case, OpenJPA with Derby server is 3.5 times faster than DataNucleus with MySQL server.

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

Retrieval Size =>	Few Entities		Many Entities		Average Score
	OpenJPA Derby server	DataNucleus MySQL server	OpenJPA Derby server	DataNucleus MySQL server	OpenJPA Derby server	DataNucleus MySQL server
Basic Person Test	0.78	0.96	11.6	9.3	6.2	5.1
Element Collection Test	0.0024	0.18	1.1	0.77	0.54	0.47
Inheritance Test	0.014	0.55	5.0	8.5	2.5	4.5
Indexing Test	0.62	0.82	13.0	9.7	6.8	5.3
Graph (Binary Tree) Test	0.065	2.1	0.39	3.5	0.23	2.8
Multithreading Test	1.6	0.27	19.0	7.8	10.3	4.0
All Tests	0.51	0.81	8.4	6.6	4.4	3.7

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

A large performance gap has been detected when using multithreading with small retrieval size. Comparing the normalized speed of DataNucleus with MySQL database server (0.27) to the normalized speed of OpenJPA with Derby database server (1.6) reveals that in that case, OpenJPA with Derby server is 5.9 times faster than DataNucleus with MySQL server.

On the other hand, OpenJPA with Derby server is slower, for instance, when using JPA element collections with small retrieval size. Comparing the normalized speed of OpenJPA with Derby database server (0.0024) to the normalized speed of DataNucleus with MySQL database server (0.18) reveals that in that case, OpenJPA with Derby server is 75.0 times slower than DataNucleus with MySQL server.

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

Retrieval Size =>	Few Entities		Many Entities		Average Score
	OpenJPA Derby server	DataNucleus MySQL server	OpenJPA Derby server	DataNucleus MySQL server	OpenJPA Derby server	DataNucleus MySQL server
Basic Person Test	29.2	7.2	5.5	2.2	17.3	4.7
Element Collection Test	2.2	5.8	1.4	0.27	1.8	3.1
Inheritance Test	2.2	2.9	1.0	1.7	1.6	2.3
Indexing Test	2.2	0.0084	11.7	2.5	7.0	1.2
Multithreading Test	35.5	failed	8.3	failed	21.9	failed
All Tests	14.3	4.0	5.6	1.7	9.9	2.8

DataNucleus with MySQL server has failed in 2 tests (see exceptions).

The results above show that in general OpenJPA with Derby server is much more efficient than DataNucleus with MySQL server in executing the tested JPA queries. Comparing the normalized speed of DataNucleus with MySQL database server (2.8) to the normalized speed of OpenJPA with Derby database server (9.9) reveals that in these tests, OpenJPA with Derby server is 3.5 times faster than DataNucleus with MySQL server.

A huge performance gap has been detected when using database indexes with small retrieval size. Comparing the normalized speed of DataNucleus with MySQL database server (0.0084) to the normalized speed of OpenJPA with Derby database server (2.2) reveals that in that case, OpenJPA with Derby server is 262 times faster than DataNucleus with MySQL server.

On the other hand, OpenJPA with Derby server is slower, for instance, when using JPA element collections with small retrieval size. Comparing the normalized speed of OpenJPA with Derby database server (2.2) to the normalized speed of DataNucleus with MySQL database server (5.8) reveals that in that case, OpenJPA with Derby server is 2.6 times slower than DataNucleus with MySQL server.

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

Transaction Size =>	Few Entities		Many Entities		Average Score
	OpenJPA Derby server	DataNucleus MySQL server	OpenJPA Derby server	DataNucleus MySQL server	OpenJPA Derby server	DataNucleus MySQL server
Basic Person Test	1.5	0.61	4.0	1.1	2.8	0.88
Element Collection Test	0.0081	0.31	1.9	0.44	0.94	0.38
Inheritance Test	0.043	0.51	3.8	1.8	1.9	1.1
Indexing Test	1.8	0.68	6.8	2.2	4.3	1.5
Graph (Binary Tree) Test	0.16	0.91	0.27	0.47	0.22	0.69
Multithreading Test	6.9	0.64	5.0	failed	5.9	0.64
All Tests	1.7	0.61	3.6	1.2	2.7	0.88

DataNucleus with MySQL server has failed in 1 tests (see exceptions).

The results above show that in general OpenJPA with Derby server is much more efficient than DataNucleus with MySQL server in updating JPA entity objects in the database. Comparing the normalized speed of DataNucleus with MySQL database server (0.88) to the normalized speed of OpenJPA with Derby database server (2.7) reveals that in these tests, OpenJPA with Derby server is 3.1 times faster than DataNucleus with MySQL server.

A huge performance gap has been detected when using multithreading with small transaction size. Comparing the normalized speed of DataNucleus with MySQL database server (0.64) to the normalized speed of OpenJPA with Derby database server (6.9) reveals that in that case, OpenJPA with Derby server is 10.8 times faster than DataNucleus with MySQL server.

On the other hand, OpenJPA with Derby server is slower, for instance, when using JPA element collections with small transaction size. Comparing the normalized speed of OpenJPA with Derby database server (0.0081) to the normalized speed of DataNucleus with MySQL database server (0.31) reveals that in that case, OpenJPA with Derby server is 38.3 times slower than DataNucleus with MySQL server.

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

Transaction Size =>	Few Entities		Many Entities		Average Score
	OpenJPA Derby server	DataNucleus MySQL server	OpenJPA Derby server	DataNucleus MySQL server	OpenJPA Derby server	DataNucleus MySQL server
Basic Person Test	1.4	0.70	3.7	1.3	2.5	1.0
Element Collection Test	0.0082	0.17	1.4	0.40	0.71	0.29
Inheritance Test	0.035	0.41	2.5	1.3	1.3	0.87
Indexing Test	2.3	1.7	3.5	1.9	2.9	1.8
Graph (Binary Tree) Test	0.13	0.49	0.35	0.46	0.24	0.47
Multithreading Test	3.6	failed	4.0	failed	3.8	failed
All Tests	1.3	0.69	2.6	1.1	1.9	0.89

DataNucleus with MySQL server has failed in 2 tests (see exceptions).

The results above show that in general OpenJPA with Derby server is more efficient than DataNucleus with MySQL server in deleting JPA entity objects from the database. Comparing the normalized speed of DataNucleus with MySQL database server (0.89) to the normalized speed of OpenJPA with Derby database server (1.9) reveals that in these tests, OpenJPA with Derby server is 2.1 times faster than DataNucleus with MySQL server.

A large performance gap has been detected when using JPA element collections with large transaction size. Comparing the normalized speed of DataNucleus with MySQL database server (0.40) to the normalized speed of OpenJPA with Derby database server (1.4) reveals that in that case, OpenJPA with Derby server is 3.5 times faster than DataNucleus with MySQL server.

On the other hand, OpenJPA with Derby server is slower, for instance, when using JPA element collections with small transaction size. Comparing the normalized speed of OpenJPA with Derby database server (0.0082) to the normalized speed of DataNucleus with MySQL database server (0.17) reveals that in that case, OpenJPA with Derby server is 20.7 times slower than DataNucleus with MySQL server.

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

Transaction/Retrieval Size	Few Entities		Many Entities		Average Score
	OpenJPA Derby server	DataNucleus MySQL server	OpenJPA Derby server	DataNucleus MySQL server	OpenJPA Derby server	DataNucleus MySQL server
Basic Person Test	7.2	2.4	5.5	3.1	6.4	2.7
Element Collection Test	0.84	1.5	1.5	0.47	1.2	0.99
Inheritance Test	1.1	1.4	2.9	2.9	2.0	2.1
Indexing Test	2.3	1.4	7.8	3.8	5.1	2.6
Graph (Binary Tree) Test	0.27	1.2	0.41	1.4	0.34	1.3
Multithreading Test	10.7	2.4	8.0	5.1	9.4	3.5
All Tests	3.9	1.7	4.5	2.6	4.2	2.1

The results above show that in general OpenJPA with Derby server is more efficient than DataNucleus with MySQL server in performing JPA database operations. Comparing the normalized speed of DataNucleus with MySQL database server (2.1) to the normalized speed of OpenJPA with Derby database server (4.2) reveals that in these tests, OpenJPA with Derby server is 2.0 times faster than DataNucleus with MySQL server.

A large performance gap has been detected when using multithreading with small transaction/retrieval size. Comparing the normalized speed of DataNucleus with MySQL database server (2.4) to the normalized speed of OpenJPA with Derby database server (10.7) reveals that in that case, OpenJPA with Derby server is 4.5 times faster than DataNucleus with MySQL server.

On the other hand, OpenJPA with Derby server is slower, for instance, when using graphs of objects with small transaction/retrieval size. Comparing the normalized speed of OpenJPA with Derby database server (0.27) to the normalized speed of DataNucleus with MySQL database server (1.2) reveals that in that case, OpenJPA with Derby server is 4.4 times slower than DataNucleus with MySQL server.

JPA Performance Benchmark (JPAB)