Compare the two non-sequential file structure models: the random (hashed) file and the index file. What advantages does the first one have over the second and what advantages does the second have over the first? What would be your criteria for choosing one over the other for different applications?