Berkeley DB about hash Access Methods

Question

When I use Berkeley DB with hash Access Methods, and the recored is very large, I want to know:
for examle when i insert two recored
(key1, data1) and (key2, data2) into DB, the hash of key1 and hash of key2 is same, but key1 and key2 is different, how the Berkeley DB to do this?
is the Berkeley DB cover the key1,data1 with key2,date2 ?, if so, i think it is a terrible things.
thanks for any reply, waiting......

Answer 1

You are very welcome, thanks again,
but now i find the Berkeley DB working are very slowly when the Database size grow up to more than 2GB, below is the list of my computer information:
OS: windows server 2003
API: C language
memory: 1GB
CPU: intel 820
database Access Methods: hash
key and dada: key is always char[32], data is always int32
when I open the database in the env, I didn't set the catch and page size because I don't know how to set it, and i also find there is anthor method to improve the performance by set the "Page fill factor", and "DB->set_h_ffactor", but can you give me some advice how can set it to make DB more fast,
by the way I find when the Berkeley DB's database size is larger than memory, it works very slowly, why?