• Benchmark
• Results
• Comparison

JPA Performance Benchmark (JPAB)

Comparison of DataNucleus with DB4O embedded vs EclipseLink 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 DB4O embedded	EclipseLink Derby embedded	DataNucleus DB4O embedded	EclipseLink Derby embedded	DataNucleus DB4O embedded	EclipseLink Derby embedded
Basic Person Test	0.059	5.5	1.3	3.7	0.69	4.6
Element Collection Test	0.045	3.0	1.0	1.7	0.52	2.4
Inheritance Test	0.051	5.3	0.85	3.5	0.45	4.4
Indexing Test	0.080	6.9	2.3	4.6	1.2	5.8
Graph (Binary Tree) Test	failed	2.0	failed	1.9	failed	1.9
Multithreading Test	failed	9.3	failed	4.1	failed	6.7
All Tests	0.059	5.3	1.4	3.2	0.71	4.3

DataNucleus with DB4O embedded has failed in 4 tests (see exceptions).

The results above show that in general EclipseLink with Derby embedded is much more efficient than DataNucleus with DB4O embedded in persisting JPA entity objects to the database. Comparing the normalized speed of DataNucleus with DB4O embedded database (0.71) to the normalized speed of EclipseLink with Derby embedded database (4.3) reveals that in these tests, EclipseLink with Derby embedded is 6.1 times faster than DataNucleus with DB4O embedded.

A huge performance gap has been detected when using class inheritance in the object model with small transaction size. Comparing the normalized speed of DataNucleus with DB4O embedded database (0.051) to the normalized speed of EclipseLink with Derby embedded database (5.3) reveals that in that case, EclipseLink with Derby embedded is 104 times faster than DataNucleus with DB4O embedded.

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

Retrieval Size =>	Few Entities		Many Entities		Average Score
	DataNucleus DB4O embedded	EclipseLink Derby embedded	DataNucleus DB4O embedded	EclipseLink Derby embedded	DataNucleus DB4O embedded	EclipseLink Derby embedded
Basic Person Test	0.0047	16.4	1.8	58.7	0.90	37.5
Element Collection Test	0.0038	7.5	1.5	13.3	0.77	10.4
Inheritance Test	0.0035	17.1	2.2	58.9	1.1	38.0
Indexing Test	0.0043	14.2	1.8	67.4	0.89	40.8
Graph (Binary Tree) Test	failed	2.8	failed	5.7	failed	4.2
Multithreading Test	failed	28.0	failed	48.7	failed	38.4
All Tests	0.0041	14.3	1.8	42.1	0.91	28.2

DataNucleus with DB4O embedded has failed in 4 tests (see exceptions).

The results above show that in general EclipseLink with Derby embedded is much more efficient than DataNucleus with DB4O embedded in retrieving JPA entity objects from the database. Comparing the normalized speed of DataNucleus with DB4O embedded database (0.91) to the normalized speed of EclipseLink with Derby embedded database (28.2) reveals that in these tests, EclipseLink with Derby embedded is 31.0 times faster than DataNucleus with DB4O embedded.

A huge performance gap has been detected when using class inheritance in the object model with small retrieval size. Comparing the normalized speed of DataNucleus with DB4O embedded database (0.0035) to the normalized speed of EclipseLink with Derby embedded database (17.1) reveals that in that case, EclipseLink with Derby embedded is 4,886 times faster than DataNucleus with DB4O embedded.

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

Retrieval Size =>	Few Entities		Many Entities		Average Score
	DataNucleus DB4O embedded	EclipseLink Derby embedded	DataNucleus DB4O embedded	EclipseLink Derby embedded	DataNucleus DB4O embedded	EclipseLink Derby embedded
Basic Person Test	failed	53.4	failed	13.8	failed	33.6
Element Collection Test	failed	37.9	failed	4.4	failed	21.2
Inheritance Test	failed	37.4	failed	15.0	failed	26.2
Indexing Test	failed	7.6	failed	49.3	failed	28.4
Multithreading Test	failed	41.2	failed	6.1	failed	23.6
All Tests	failed	35.5	failed	17.7	failed	26.6

DataNucleus with DB4O embedded 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
	DataNucleus DB4O embedded	EclipseLink Derby embedded	DataNucleus DB4O embedded	EclipseLink Derby embedded	DataNucleus DB4O embedded	EclipseLink Derby embedded
Basic Person Test	0.012	6.4	0.76	6.7	0.39	6.5
Element Collection Test	0.012	5.4	1.1	6.2	0.53	5.8
Inheritance Test	0.011	6.5	0.81	9.2	0.41	7.8
Indexing Test	0.011	7.3	1.3	11.1	0.64	9.2
Graph (Binary Tree) Test	failed	2.8	failed	1.7	failed	2.2
Multithreading Test	failed	21.3	failed	5.0	failed	13.1
All Tests	0.012	8.3	0.97	6.6	0.49	7.5

DataNucleus with DB4O embedded has failed in 4 tests (see exceptions).

The results above show that in general EclipseLink with Derby embedded is much more efficient than DataNucleus with DB4O embedded in updating JPA entity objects in the database. Comparing the normalized speed of DataNucleus with DB4O embedded database (0.49) to the normalized speed of EclipseLink with Derby embedded database (7.5) reveals that in these tests, EclipseLink with Derby embedded is 15.3 times faster than DataNucleus with DB4O embedded.

A huge performance gap has been detected when using database indexes with small transaction size. Comparing the normalized speed of DataNucleus with DB4O embedded database (0.011) to the normalized speed of EclipseLink with Derby embedded database (7.3) reveals that in that case, EclipseLink with Derby embedded is 664 times faster than DataNucleus with DB4O embedded.

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

Transaction Size =>	Few Entities		Many Entities		Average Score
	DataNucleus DB4O embedded	EclipseLink Derby embedded	DataNucleus DB4O embedded	EclipseLink Derby embedded	DataNucleus DB4O embedded	EclipseLink Derby embedded
Basic Person Test	0.011	4.6	1.5	4.7	0.74	4.6
Element Collection Test	0.010	2.5	1.6	1.4	0.80	1.9
Inheritance Test	0.0093	3.9	1.1	4.5	0.57	4.2
Indexing Test	0.017	6.2	2.1	3.8	1.1	5.0
Graph (Binary Tree) Test	failed	1.1	failed	1.0	failed	1.1
Multithreading Test	failed	6.2	failed	5.8	failed	6.0
All Tests	0.012	4.1	1.6	3.5	0.79	3.8

DataNucleus with DB4O embedded has failed in 4 tests (see exceptions).

The results above show that in general EclipseLink with Derby embedded is much more efficient than DataNucleus with DB4O embedded in deleting JPA entity objects from the database. Comparing the normalized speed of DataNucleus with DB4O embedded database (0.79) to the normalized speed of EclipseLink with Derby embedded database (3.8) reveals that in these tests, EclipseLink with Derby embedded is 4.8 times faster than DataNucleus with DB4O embedded.

A huge performance gap has been detected when using class inheritance in the object model with small transaction size. Comparing the normalized speed of DataNucleus with DB4O embedded database (0.0093) to the normalized speed of EclipseLink with Derby embedded database (3.9) reveals that in that case, EclipseLink with Derby embedded is 419 times faster than DataNucleus with DB4O embedded.

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

Transaction Size =>	Few Entities		Many Entities		Average Score
	DataNucleus DB4O embedded	EclipseLink Derby embedded	DataNucleus DB4O embedded	EclipseLink Derby embedded	DataNucleus DB4O embedded	EclipseLink Derby embedded
Basic Person Test	35.1	51.1	35.4	51.0	35.2	51.1
Element Collection Test	29.5	38.5	29.9	39.4	29.7	38.9
Inheritance Test	33.4	48.0	33.5	47.9	33.4	48.0
Indexing Test	41.0	51.6	41.9	52.3	41.5	52.0
Graph (Binary Tree) Test	failed	13.4	failed	11.7	failed	12.6
Multithreading Test	failed	52.6	failed	8.4	failed	30.5
All Tests	34.8	42.5	35.2	35.1	35.0	38.8

DataNucleus with DB4O embedded has failed in 4 tests (see exceptions).

The results above show that in general EclipseLink with Derby embedded is slightly more efficient than DataNucleus with DB4O 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 DB4O embedded	EclipseLink Derby embedded	DataNucleus DB4O embedded	EclipseLink Derby embedded	DataNucleus DB4O embedded	EclipseLink Derby embedded
Basic Person Test	0.022	17.3	1.3	17.5	0.68	17.4
Element Collection Test	0.018	11.3	1.3	5.4	0.66	8.3
Inheritance Test	0.019	14.0	1.2	18.2	0.63	16.1
Indexing Test	0.028	8.4	1.9	27.2	0.94	17.8
Graph (Binary Tree) Test	failed	2.2	failed	2.6	failed	2.4
Multithreading Test	failed	21.2	failed	13.9	failed	17.6
All Tests	0.022	12.7	1.4	14.5	0.73	13.6

The results above show that in general EclipseLink with Derby embedded is much more efficient than DataNucleus with DB4O embedded in performing JPA database operations. Comparing the normalized speed of DataNucleus with DB4O embedded database (0.73) to the normalized speed of EclipseLink with Derby embedded database (13.6) reveals that in these tests, EclipseLink with Derby embedded is 18.6 times faster than DataNucleus with DB4O 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 DB4O embedded database (0.022) to the normalized speed of EclipseLink with Derby embedded database (17.3) reveals that in that case, EclipseLink with Derby embedded is 786 times faster than DataNucleus with DB4O embedded.

Other Head to Head DBMS/JPA Comparisons

JPA Performance Benchmark - Copyright © 2010, 2011, 2012 ObjectDB Software Ltd. - All rights reserved.
The benchmark program is released as open source under the GPL license.