I was wondering which is better/faster:

  1. Having a separate collection of documents that just contain the image saved as binary data, and possibly some metadata.
  2. Or using GridFS to store the images.
Edit
Report