Jeff Ramsdale wrote:
I work for IBM and I am listed on the to-do list
of the Apache/Derby website as being the contact for in-memory database.
4) Finally, I believe I saw a reference somewhere
indicating beta code existed to allow Derby (or
Cloudscape?) to run in-memory only. What's the status
of that code and if it's available how would one make
use of it?
I wrote some code to allow Derby to run in-memory. However, it is not
beta code. It is a prototype written to test the
org.apache.derby.io.StorageFactory interface and to show that the
interface could be used to write an in-memory database.
The prototype code implements entirely non-persistent dabases. The data
goes away when the JVM ends, the code does not automatically save the
data to a file at termination nor does it automatically read data from
a file when the database starts up. An application can use the normal
backup/restore facilities to save and restore to and from a file.
I have been told that I can contribute the code to the community,
though the mechanics of doing so and where it should go remain to be
worked out. Is there some interest in the code?