Comparison of EclipseLink with H2 server vs DataNucleus with SQLite 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
	EclipseLink H2 server	DataNucleus SQLite embedded	EclipseLink H2 server	DataNucleus SQLite embedded	EclipseLink H2 server	DataNucleus SQLite embedded
Basic Person Test	12.6		9.5		11.1
Element Collection Test	12.5		6.6		9.5
Inheritance Test	6.5		10.5		8.5
Indexing Test	12.2		12.3		12.3
Graph (Binary Tree) Test	8.3		7.3		7.8
Multithreading Test	11.5		0.0000		5.8
All Tests	10.6		7.7		9.2

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

Retrieval Size =>	Few Entities		Many Entities		Average Score
	EclipseLink H2 server	DataNucleus SQLite embedded	EclipseLink H2 server	DataNucleus SQLite embedded	EclipseLink H2 server	DataNucleus SQLite embedded
Basic Person Test	2.8	0.43	6.6		4.7	0.43
Element Collection Test	1.5		1.8		1.7
Inheritance Test	2.1		7.9		5.0
Indexing Test	2.4		8.0		5.2
Graph (Binary Tree) Test	0.38		0.81		0.60
Multithreading Test	4.7		11.2		8.0
All Tests	2.3	0.43	6.0		4.2	0.43

The results above show that in general EclipseLink with H2 server is much more efficient than DataNucleus with SQLite embedded in retrieving JPA entity objects from the database. Comparing the normalized speed of DataNucleus with SQLite embedded database (0.43) to the normalized speed of EclipseLink with H2 database server (4.2) reveals that in these tests, EclipseLink with H2 server is 9.8 times faster than DataNucleus with SQLite embedded.

A large performance gap has been detected when using simple basic entities with small retrieval size. Comparing the normalized speed of DataNucleus with SQLite embedded database (0.43) to the normalized speed of EclipseLink with H2 database server (2.8) reveals that in that case, EclipseLink with H2 server is 6.5 times faster than DataNucleus with SQLite embedded.

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

Retrieval Size =>	Few Entities		Many Entities		Average Score
	EclipseLink H2 server	DataNucleus SQLite embedded	EclipseLink H2 server	DataNucleus SQLite embedded	EclipseLink H2 server	DataNucleus SQLite embedded
Basic Person Test	1.2	failed	3.1		2.2	failed
Element Collection Test	1.8		1.9		1.9
Inheritance Test	0.75		3.3		2.0
Indexing Test	2.1		3.5		2.8
Multithreading Test	3.6		6.7		5.1
All Tests	1.9	failed	3.7		2.8	failed

DataNucleus with SQLite embedded has failed in 1 tests (see exceptions).

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

Transaction Size =>	Few Entities		Many Entities		Average Score
	EclipseLink H2 server	DataNucleus SQLite embedded	EclipseLink H2 server	DataNucleus SQLite embedded	EclipseLink H2 server	DataNucleus SQLite embedded
Basic Person Test	2.1	0.025	2.8		2.4	0.025
Element Collection Test	4.8		1.9		3.4
Inheritance Test	2.1		3.0		2.6
Indexing Test	5.5		4.4		5.0
Graph (Binary Tree) Test	1.0		0.73		0.89
Multithreading Test	2.7		1.4		2.0
All Tests	3.0	0.025	2.4		2.7	0.025

The results above show that in general EclipseLink with H2 server is much more efficient than DataNucleus with SQLite embedded in updating JPA entity objects in the database. Comparing the normalized speed of DataNucleus with SQLite embedded database (0.025) to the normalized speed of EclipseLink with H2 database server (2.7) reveals that in these tests, EclipseLink with H2 server is 108 times faster than DataNucleus with SQLite embedded.

A huge performance gap has been detected when using simple basic entities with small transaction size. Comparing the normalized speed of DataNucleus with SQLite embedded database (0.025) to the normalized speed of EclipseLink with H2 database server (2.1) reveals that in that case, EclipseLink with H2 server is 84.0 times faster than DataNucleus with SQLite embedded.

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

Transaction Size =>	Few Entities		Many Entities		Average Score
	EclipseLink H2 server	DataNucleus SQLite embedded	EclipseLink H2 server	DataNucleus SQLite embedded	EclipseLink H2 server	DataNucleus SQLite embedded
Basic Person Test	12.1	0.17	9.6		10.9	0.17
Element Collection Test	8.3		1.2		4.8
Inheritance Test	13.7		7.4		10.6
Indexing Test	23.0		7.0		15.0
Graph (Binary Tree) Test	7.5		10.9		9.2
Multithreading Test	7.0		0.0000		3.5
All Tests	11.9	0.17	6.0		9.0	0.17

The results above show that in general EclipseLink with H2 server is much more efficient than DataNucleus with SQLite embedded in deleting JPA entity objects from the database. Comparing the normalized speed of DataNucleus with SQLite embedded database (0.17) to the normalized speed of EclipseLink with H2 database server (9.0) reveals that in these tests, EclipseLink with H2 server is 52.9 times faster than DataNucleus with SQLite embedded.

A huge performance gap has been detected when using simple basic entities with small transaction size. Comparing the normalized speed of DataNucleus with SQLite embedded database (0.17) to the normalized speed of EclipseLink with H2 database server (12.1) reveals that in that case, EclipseLink with H2 server is 71.2 times faster than DataNucleus with SQLite embedded.

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

Transaction/Retrieval Size	Few Entities		Many Entities		Average Score
	EclipseLink H2 server	DataNucleus SQLite embedded	EclipseLink H2 server	DataNucleus SQLite embedded	EclipseLink H2 server	DataNucleus SQLite embedded
Basic Person Test	6.2	0.21	6.3		6.2	0.21
Element Collection Test	5.8		2.7		4.2
Inheritance Test	5.0		6.4		5.7
Indexing Test	9.0		7.0		8.0
Graph (Binary Tree) Test	4.3		4.9		4.6
Multithreading Test	5.9		3.9		4.9
All Tests	6.1	0.21	5.2		5.7	0.21

The results above show that in general EclipseLink with H2 server is much more efficient than DataNucleus with SQLite embedded in performing JPA database operations. Comparing the normalized speed of DataNucleus with SQLite embedded database (0.21) to the normalized speed of EclipseLink with H2 database server (5.7) reveals that in these tests, EclipseLink with H2 server is 27.1 times faster than DataNucleus with SQLite embedded.

A huge performance gap has been detected when using simple basic entities with small transaction/retrieval size. Comparing the normalized speed of DataNucleus with SQLite embedded database (0.21) to the normalized speed of EclipseLink with H2 database server (6.2) reveals that in that case, EclipseLink with H2 server is 29.5 times faster than DataNucleus with SQLite embedded.

JPA Performance Benchmark (JPAB)