I read the paper and it seemed like an apples to oranges comparison. It
sounded like comparison of
some proprietary API to JDBC ?
Also could not find details about how exactly the timings were measured
(JDBC PreparedStatement v/s some put(key,data) API ?).
Anyways, here is another paper from them comparing Berkeley DB with
Relational database systems in general.
http://www.oracle.com/database/docs/Berkeley-DB-v-Relational.pdf
This one seemed more relevant since Berkeley DB is essential touted as a
library and not a full-fledged database system.
Also should you get a chance try reading the licensing terms ( I got a
bit lost reading their FAQ):
http://www.oracle.com/technology/software/products/berkeley-db/htdocs/licensing.html
-Rajesh
Mike Matrigali wrote:
Just some observations from reading the paper:
o only one comparison is done with real durability? We usually just
up # of users or operations per transaction to show increased
throughput while not being I/O bound.
o great marketing on that first durability graph, I looked at it and
couldn't understand why we were half the speed then saw that we were
83 vs. 89 inserts/sec.
o I believe the insert numbers should be better for derby in 10.2.
o I wonder if our use of file sync during consistency points is hurting
us in the other write-sync tests?
Mike Matrigali wrote:
From a posting on the user list:
I actually just found a whitepaper
"Oracle Berkeley DB Java Edition vs Apache Derby: A Performance
Comparison"
http://www.oracle.com/technology/products/berkeley-db/pdf/je-derby-performance.pdf
Just copying there conclusion here:
"The Berkeley DB Java Edition Persistence API is a high performance,
complete solution for Java object persistence. Berkeley DB Java
Edition performance exceeds Derby performance in every test, by a
factor of 3 to 10, clearly demonstrating the superior performance of
Berkeley DB Java Edition"
Cheers
Kasper
