Memory usage with large data driven test
Hi - I am running a data driven test case with about 100 rows of data. It calls 2 other test cases during each loop. It also uses the methods of an external Groovy class, but they are all declared ...
Forum Discussion
Hi again
following on, after cutting my dataset down to a measly 5k rows now
it appears that the java heap error occurs at some point after the data loop (last step in the test case) and before and data is actually written out to the datasink destination file
is there any setting that will allow batch/serial writes to the datasink output file during the data loop instead of caching the data till the end, I understand that this may have implications on the time it will take to complete the test case but I'm willing to take that hit (within reason of course)
if I drop the dataset down a bit more the test case finishes but the output file exceeds 100MB, correct me if I'm wrong but I can only assume that by trying to create this file in one go with the larger dataset is a major factor in the memory problem
regards
Mike
following on, after cutting my dataset down to a measly 5k rows now
it appears that the java heap error occurs at some point after the data loop (last step in the test case) and before and data is actually written out to the datasink destination file
is there any setting that will allow batch/serial writes to the datasink output file during the data loop instead of caching the data till the end, I understand that this may have implications on the time it will take to complete the test case but I'm willing to take that hit (within reason of course)
if I drop the dataset down a bit more the test case finishes but the output file exceeds 100MB, correct me if I'm wrong but I can only assume that by trying to create this file in one go with the larger dataset is a major factor in the memory problem
regards
Mike