Hi -

Is it reasonable to have up to several hundred
DBI database handles open and connected in one
process? I am running a multi-threaded perl (5.8)
server and would like thread-by-thread access to
severl databases; the current threading module
does not allow complex structures (like dbhs)
to be declared as shared variables, so I am thinking
of connecting/disconnecting at thread start/end time
to save the overhead of connecting/disconnecting on
each use (like Apache::DBI).

But my thread pool can grow to several hundred
(configuration parameter). My server is robust -
I think it can handle the overhead - but are there other
considerations?

Aloha => Beau;


-- 
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to