EFFICIENT TRAVERSAL OF HIERARCHICAL DATASETS

In one embodiment, a computer implemented method comprises receiving, at a computing device, a request for a particular user identification (ID) to perform a particular operation on a particular data object. An entitlement cache is generated that associates each operation that the particular user ID is entitled to perform with a first encoding of a tuple of a plurality of tuples, the first encoding comprising a vector of one or more value pairs, each value pair of the one or more value pairs including a left value and right value, each of the one or more value pairs corresponding to the left value and the right value of a node of a plurality of nodes from hierarchical data. An object mapping cache is generated that associates each tuple of the plurality of tuples with a second encoding of each respective tuple of the plurality of tuples, the second encoding comprising a vector of one or more left values, each of the one or more left values corresponding to a left value of a node of the plurality of nodes from the hierarchical data. An object mapping is used to determine a first tuple based on the particular data object. The object mapping cache is used to determine a first vector of one of more left values based on the first tuple. The entitlement cache is used to determine a second vector of one or more value pairs based on the particular operation. A match is identified between the first vector and the second vector. In response to identifying the match between the first vector and the second vector, the computing device grants the particular user ID access to the particular data object.