This benchmark test compares the performance of MongoDB and JSONB when inserting, updating, reading and deleting documents.
The test was performed on a machine with the following specifications:
- CPU: intel core i3 9100f
- RAM: 16GB DDR4
- Storage: 256GB SSD
- OS: Fedora 37
The test results are shown in the table below:
| Test Name | Number of Documents | Time (ns/op) | Memory (B/op) | Allocs/op |
|---|---|---|---|---|
| BenchmarkMongo1 | 1 | 0.03156 ns/op | 0 B/op | 0 |
| BenchmarkMongo100 | 100 | 0.4614 ns/op | 0 B/op | 0 |
| BenchmarkMongo200 | 200 | 0.9224 ns/op | 0 B/op | 0 |
| BenchmarkMongo500 | 500 | 2076665187 ns/op | 15726880 B/op | 251912 |
| BenchmarkJSONB1 | 1 | 0.01976 ns/op | 0 B/op | 0 |
| BenchmarkJSONB100 | 100 | 204185849 ns/op | 136963 B/op | 2044 |
| BenchmarkJSONB200 | 200 | 2617058157 ns/op | 1292336 B/op | 20323 |
| BenchmarkJSONB500 | 500 | 6457528499 ns/op | 3113936 B/op | 50623 |
- MongoDB tests took about 0.03 seconds for 1 document, 0.46 seconds for 100 documents, and 0.92 seconds for 200 documents.
- JSONB tests took about 0.019 seconds for 1 document, 0.204 seconds for 100 documents, and 2.61 seconds for 200 documents.
- This is just one way to measure the performance of these technologies, and the results may vary depending on the specific use case and hardware being used. Other factors such as the complexity of the queries, the amount of data being stored and retrieved, and the number of concurrent requests being made can also affect performance.
- Maybe JSONB is more performant when working with big data, but MongoDB is more performant when dealing with small data. It depends on your use case and the amount of data that you need to manage.
hard 136 mb mem 335 mb
hard 80 mem 262
BenchmarkPostgres50-4 1000000000 0.3320 ns/op 0 B/op 0 allocs/op BenchmarkPostgres100-4 6128 172249 ns/op 23 B/op 0 allocs/op BenchmarkPostgres200-4 1 1545513570 ns/op 256448 B/op 6906 allocs/op
BenchmarkPostgres50-4 1000000000 0.3794 ns/op 0 B/op 0 allocs/op BenchmarkPostgres100-4 21 51429318 ns/op 6835 B/op 171 allocs/op BenchmarkPostgres200-4 1 1406782021 ns/op 256752 B/op 6909 allocs/op
mem after 276
BenchmarkPostgres50-4 1000000000 0.3135 ns/op 0 B/op 0 allocs/op BenchmarkPostgres100-4 4622 220736 ns/op 24 B/op 0 allocs/op BenchmarkPostgres200-4 1 1264586084 ns/op 213152 B/op 6875 allocs/op
BenchmarkPostgres50-4 1000000000 0.3150 ns/op 0 B/op 0 allocs/op BenchmarkPostgres100-4 271 3899507 ns/op 422 B/op 12 allocs/op BenchmarkPostgres200-4 1 1576092967 ns/op 212048 B/op 6866 allocs/op
mem after 276