The Melbourne shuffle algorithm is a sequence of actions intended to obscure the patterns by which cloud-based data is accessed. The goal is to make it difficult for unauthorized parties to draw conclusions about what type of data is being stored in the cloud by observing patterns that emerge as the data is accessed.
Even when data is encrypted, details about how often the data is accessed or what action is taken after the data has been accessed can be revealing. By analyzing digital footprints, an outsider can predict such things as who is likely to own a particular data set or what business announcement is likely to correlate with a particular access pattern.
As with a deck of cards, a data shuffle rearranges the array to achieve a random permutation of its elements. The Melbourne shuffle moves small amounts of data from the cloud server to the user's local memory, where it is rearranged before being returned to the server. Even when the same user repeatedly accesses the same data, shuffling ensures the access path will not be consistent.
The algorithm, which obfuscates access patterns by making them look quite random, was written by computer scientists at Brown University in 2014. It is named for another kind of shuffle -- a popular dance move in Australia during the 1990s.