I have found that the TorqueDataDump and even more so the TorqueDataSQL
tasks are HUGE memory hogs when run against a medium sized data list - my
current tests are 3 tables of about 8000 rows.

It appears that these tasks load everything into memory before writing out -
in the case of DataSQL the whole xml file is loaded into an AppData object
then run through velocity, which seems to keep quite  bit in memory also.

Is anyone already working on a way to reduce this memory usage? (trying to
prevent duplication of effort on something that will take me a while with no
velocity exprience)

Thanks
Russell Smyth

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

Reply via email to