I found a lot of comparisions here, but not this one; So, what is best in each one?
There's a full comparison at SQLite's site.
SQLite is much more restricted, as it only supports a small subset of SQL92, whereas Derby (now JavaDB) has full support of SQL92 and SQL99.
I execute a complex SQL which has more than 6000 rows 10000 times in my Websphere Server. Total net execution times are like that:
Derby (In Memory) Oracle(standard DB) SQLite (In Memory) HSQLDb (In Memory) nano sec. second nano sec. second nano sec. second nano sec. second 1. try 58000000 0,058 6149976000 6,1 1141988000 1,14 999403000 1,00 2. try 78560000 0,078 5268477000 5,2 1182621000 1,18 1338705000 1,34 3. try 58849000 0,058 5200898000 5,2 1133003000 1,13 2239527000 2,24 4. try 60901000 0,06 5435216000 5,4 1205442000 1,21 1370711000 1,37 5. try 58798000 0,058 6501929000 6,5 1186734000 1,19 1001800000 1,00 6. try 62928000 0,062 5913053000 5,9 1224470000 1,22 1066736000 1,07 7. try 71171000 0,071 5111207000 5,1 1200769000 1,20 1304524000 1,30 8. try 66913000 0,066 5517989000 5,5 1173495000 1,17 1299230000 1,30 9. try 58777000 0,058 7209555000 7,2 1179013000 1,18 1031795000 1,03 10. try 75299000 0,075 5356514000 5,3 1182715000 1,18 1368461000 1,37 average 65019600 0,064 5766481400 5,7 1181025000 1,18 1302089200 1,30
I obviously compare Derby, SQLite and HSQLDB. Oracle isn't a in memory db. But I put it's result to table because to show speed difference between a in memory db and normal db.
PS: In SQLite and HSQLDB results aren't stable. So I choosed 10 stable results in 100 tries. Sometimes HSQLDB is faster than SQLite. I think theirs performance are same.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With