I completely rewrote the cache in 4.0 it is now called PrefetchCache and it no longer uses all the funky data structures. It can also merge Prefetched data back into the main cache when the transaction ends, instead of throwing it away. This rocks for Commit Option A but does not help B or C. Does anyone want to back port this to 3.2? It is fairly simple. If not, I'll do it. I don't think I should back port it to 3.0 as it has not been heavily tested.

-dain

On Friday, November 22, 2002, at 07:26 AM, Stephen Coy wrote:

I'll try to get it done in the next day or so.

On Friday, November 22, 2002, at 10:50  PM, Christian Riege wrote:

hi stepehn et. al,

As far as the Finders are concerned, I suspect that the issue may be
in the ReadAheadCache code.
yeah ReadAheadCache is currently the HotSpot. I think Dain did some
optimizations for that in 3.2 which haven't made it into 3.0 just yet?
if you port the fix that you applied to 3.0 (the GlobalTxEntityMap one)
to 3.2 i can re-run through OptimizeIt and see if its faster.

regards,
christian


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
JBoss-user mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/jboss-user


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
JBoss-user mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/jboss-user

Reply via email to