Because we encounter a lot of random lock errors when running large datasets over many iterations. Currently we modified our codes to move some datasets to a user-defined library with the filelockwait option, but if we can have the option to the work library, we don't need to change many of our codes.
... View more
