Comparison of DataNucleus with HSQLDB embedded vs Hibernate with Derby 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
	DataNucleus HSQLDB embedded	Hibernate Derby embedded	DataNucleus HSQLDB embedded	Hibernate Derby embedded	DataNucleus HSQLDB embedded	Hibernate Derby embedded
Basic Person Test	14.2	5.9	4.8	3.8	9.5	4.9
Element Collection Test	6.6	2.9	2.1	1.8	4.3	2.3
Inheritance Test	13.6	5.3	4.6	3.7	9.1	4.5
Indexing Test	24.6	7.0	11.1	4.8	17.8	5.9
Graph (Binary Tree) Test	6.8	1.9	4.5	1.7	5.7	1.8
Multithreading Test	38.5	10.5	8.5	3.7	23.5	7.1
All Tests	17.4	5.6	5.9	3.3	11.7	4.4

The results above show that in general DataNucleus with HSQLDB embedded is more efficient than Hibernate with Derby embedded in persisting JPA entity objects to the database. Comparing the normalized speed of Hibernate with Derby embedded database (4.4) to the normalized speed of DataNucleus with HSQLDB embedded database (11.7) reveals that in these tests, DataNucleus with HSQLDB embedded is 2.7 times faster than Hibernate with Derby embedded.

A large performance gap has been detected when using multithreading with small transaction size. Comparing the normalized speed of Hibernate with Derby embedded database (10.5) to the normalized speed of DataNucleus with HSQLDB embedded database (38.5) reveals that in that case, DataNucleus with HSQLDB embedded is 3.7 times faster than Hibernate with Derby embedded.

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

Retrieval Size =>	Few Entities		Many Entities		Average Score
	DataNucleus HSQLDB embedded	Hibernate Derby embedded	DataNucleus HSQLDB embedded	Hibernate Derby embedded	DataNucleus HSQLDB embedded	Hibernate Derby embedded
Basic Person Test	6.6	14.2	10.2	16.4	8.4	15.3
Element Collection Test	3.6	4.4	8.1	4.9	5.8	4.6
Inheritance Test	5.7	10.3	8.8	18.6	7.2	14.4
Indexing Test	5.1	10.0	9.5	16.4	7.3	13.2
Graph (Binary Tree) Test	7.2	2.6	9.3	3.0	8.3	2.8
Multithreading Test	8.1	19.3	9.5	19.3	8.8	19.3
All Tests	6.0	10.1	9.2	13.1	7.6	11.6

The results above show that in general Hibernate with Derby embedded is more efficient than DataNucleus with HSQLDB embedded 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 HSQLDB embedded database (8.1) to the normalized speed of Hibernate with Derby embedded database (19.3) reveals that in that case, Hibernate with Derby embedded is 2.4 times faster than DataNucleus with HSQLDB embedded.

On the other hand, Hibernate with Derby embedded is slower, for instance, when using graphs of objects with large retrieval size. Comparing the normalized speed of Hibernate with Derby embedded database (3.0) to the normalized speed of DataNucleus with HSQLDB embedded database (9.3) reveals that in that case, Hibernate with Derby embedded is 3.1 times slower than DataNucleus with HSQLDB embedded.

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

Retrieval Size =>	Few Entities		Many Entities		Average Score
	DataNucleus HSQLDB embedded	Hibernate Derby embedded	DataNucleus HSQLDB embedded	Hibernate Derby embedded	DataNucleus HSQLDB embedded	Hibernate Derby embedded
Basic Person Test	1.4	49.3	0.48	6.8	0.95	28.1
Element Collection Test	1.3	37.6	0.67	2.7	1.0	20.1
Inheritance Test	1.6	24.1	0.61	7.6	1.1	15.9
Indexing Test	0.0016	5.3	0.63	14.8	0.32	10.1
Multithreading Test	0.40	39.7	failed	6.1	0.40	22.9
All Tests	0.96	31.2	0.60	7.6	0.80	19.4

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

The results above show that in general Hibernate with Derby embedded is much more efficient than DataNucleus with HSQLDB embedded in executing the tested JPA queries. Comparing the normalized speed of DataNucleus with HSQLDB embedded database (0.80) to the normalized speed of Hibernate with Derby embedded database (19.4) reveals that in these tests, Hibernate with Derby embedded is 24.2 times faster than DataNucleus with HSQLDB embedded.

A huge performance gap has been detected when using database indexes with small retrieval size. Comparing the normalized speed of DataNucleus with HSQLDB embedded database (0.0016) to the normalized speed of Hibernate with Derby embedded database (5.3) reveals that in that case, Hibernate with Derby embedded is 3,312 times faster than DataNucleus with HSQLDB embedded.

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

Transaction Size =>	Few Entities		Many Entities		Average Score
	DataNucleus HSQLDB embedded	Hibernate Derby embedded	DataNucleus HSQLDB embedded	Hibernate Derby embedded	DataNucleus HSQLDB embedded	Hibernate Derby embedded
Basic Person Test	11.5	4.5	2.7	3.2	7.1	3.8
Element Collection Test	6.9	3.9	2.7	3.1	4.8	3.5
Inheritance Test	10.3	4.6	3.6	4.5	6.9	4.5
Indexing Test	11.4	4.6	4.4	5.5	7.9	5.1
Graph (Binary Tree) Test	3.5	2.7	1.5	1.2	2.5	2.0
Multithreading Test	failed	11.6	failed	2.6	failed	7.1
All Tests	8.7	5.3	3.0	3.3	5.8	4.3

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

The results above show that in general DataNucleus with HSQLDB embedded is more efficient than Hibernate with Derby embedded in updating JPA entity objects in the database.

A large performance gap has been detected when using simple basic entities with small transaction size. Comparing the normalized speed of Hibernate with Derby embedded database (4.5) to the normalized speed of DataNucleus with HSQLDB embedded database (11.5) reveals that in that case, DataNucleus with HSQLDB embedded is 2.6 times faster than Hibernate with Derby embedded.

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

Transaction Size =>	Few Entities		Many Entities		Average Score
	DataNucleus HSQLDB embedded	Hibernate Derby embedded	DataNucleus HSQLDB embedded	Hibernate Derby embedded	DataNucleus HSQLDB embedded	Hibernate Derby embedded
Basic Person Test	4.2	3.8	1.3	4.2	2.8	4.0
Element Collection Test	4.9	2.3	1.6	1.5	3.2	1.9
Inheritance Test	4.0	4.0	1.3	4.3	2.7	4.2
Indexing Test	7.5	5.5	1.9	3.6	4.7	4.6
Graph (Binary Tree) Test	2.9	0.74	2.0	0.96	2.5	0.85
Multithreading Test	40.1	5.6	13.7	4.9	26.9	5.3
All Tests	10.6	3.7	3.7	3.3	7.1	3.5

The results above show that in general DataNucleus with HSQLDB embedded is more efficient than Hibernate with Derby embedded in deleting JPA entity objects from the database. Comparing the normalized speed of Hibernate with Derby embedded database (3.5) to the normalized speed of DataNucleus with HSQLDB embedded database (7.1) reveals that in these tests, DataNucleus with HSQLDB embedded is 2.0 times faster than Hibernate with Derby embedded.

A large performance gap has been detected when using multithreading with small transaction size. Comparing the normalized speed of Hibernate with Derby embedded database (5.6) to the normalized speed of DataNucleus with HSQLDB embedded database (40.1) reveals that in that case, DataNucleus with HSQLDB embedded is 7.2 times faster than Hibernate with Derby embedded.

On the other hand, DataNucleus with HSQLDB embedded is slower, for instance, when using class inheritance in the object model with large transaction size. Comparing the normalized speed of DataNucleus with HSQLDB embedded database (1.3) to the normalized speed of Hibernate with Derby embedded database (4.3) reveals that in that case, DataNucleus with HSQLDB embedded is 3.3 times slower than Hibernate with Derby embedded.

Comparison of database storage efficiency (normalized score, higher is better)

Transaction Size =>	Few Entities		Many Entities		Average Score
	DataNucleus HSQLDB embedded	Hibernate Derby embedded	DataNucleus HSQLDB embedded	Hibernate Derby embedded	DataNucleus HSQLDB embedded	Hibernate Derby embedded
Basic Person Test	41.4	52.7	41.4	52.7	41.4	52.7
Element Collection Test	23.4	42.3	23.4	42.3	23.4	42.3
Inheritance Test	41.4	49.2	41.4	49.2	41.4	49.2
Indexing Test	48.4	53.0	49.2	53.9	48.8	53.4
Graph (Binary Tree) Test	12.5	13.1	9.4	9.8	10.9	11.5
Multithreading Test	51.9	49.3	10.3	8.8	31.1	29.0
All Tests	36.5	43.3	29.2	36.1	32.9	39.7

The results above show that in general Hibernate with Derby embedded is slightly more efficient than DataNucleus with HSQLDB embedded in using disk space.

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

Transaction/Retrieval Size	Few Entities		Many Entities		Average Score
	DataNucleus HSQLDB embedded	Hibernate Derby embedded	DataNucleus HSQLDB embedded	Hibernate Derby embedded	DataNucleus HSQLDB embedded	Hibernate Derby embedded
Basic Person Test	7.6	15.5	3.9	6.9	5.7	11.2
Element Collection Test	4.7	10.2	3.0	2.8	3.8	6.5
Inheritance Test	7.1	9.7	3.8	7.7	5.4	8.7
Indexing Test	9.7	6.5	5.5	9.0	7.6	7.8
Graph (Binary Tree) Test	5.1	2.0	4.3	1.7	4.7	1.8
Multithreading Test	21.8	17.3	10.6	7.3	17.0	12.3
All Tests	9.0	10.5	4.8	6.1	7.0	8.3

The results above show that in general Hibernate with Derby embedded is slightly more efficient than DataNucleus with HSQLDB embedded in performing JPA database operations.

A large performance gap has been detected when using JPA element collections with small transaction/retrieval size. Comparing the normalized speed of DataNucleus with HSQLDB embedded database (4.7) to the normalized speed of Hibernate with Derby embedded database (10.2) reveals that in that case, Hibernate with Derby embedded is 2.2 times faster than DataNucleus with HSQLDB embedded.

On the other hand, Hibernate with Derby embedded is slower, for instance, when using graphs of objects with small transaction/retrieval size. Comparing the normalized speed of Hibernate with Derby embedded database (2.0) to the normalized speed of DataNucleus with HSQLDB embedded database (5.1) reveals that in that case, Hibernate with Derby embedded is 2.6 times slower than DataNucleus with HSQLDB embedded.

JPA Performance Benchmark (JPAB)