If the codebase is relatively large, for example, exceeding 20GB, how much memory and disk space is the parsing process expected to consume?