The metadata files are host / os independent and are the same for everyone. Since some task's metadata take longer to process than others (some can take +30min to complete), and since the content is the same for everyone, it would nice to use Amazon's S3 to host these pre-processed metadata files to improve user's quality of life by speeding up the process.

The user should also have the option to process the metadata if he/she wishes via an API input arg.