Abstract:Quantum computing is a popular topic in computer science, which has recently attracted many studies in various areas such as machine learning and network. However, the topic of quantum data structures seems neglected. There is an open problem in the database area: Can we improve existing data structures by quantum techniques? Consider a dataset of key-record pairs. Given an interval as a query range, a classical B+ tree can report all the records with keys within this interval, which is called a range query, in O(log N + k) time, where N is the total number of records and k is the output size. It is asymptotically optimal in a classical computer but not efficient enough in a quantum computer, because it is expected that the execution time and the output size are linear in a quantum computer. In this paper, we propose the quantum range query problem. Different from the classical range queries, a quantum range query returns the results in quantum bits, which has broad potential applications due to the foreseeable advance of quantum computers and quantum algorithms. To the best of our knowledge, we design the first tree-like quantum data structure called the quantum B+ tree. Based on this data structure, we propose a hybrid quantum-classical algorithm to do the range search. It answers a static quantum range query in O(log_B N) time, which is asymptotically optimal in quantum computers. Since the execution time does not depend on the output size (i.e., k, which could be as large as O(N)), it is significantly faster than the classical data structure. Moreover, we extend our quantum B+ tree to answer the dynamic and d-dimensional quantum range queries efficiently in O(log^2_B N) and O(log^d_B N) time, respectively. Our experimental results show that our proposed quantum data structures achieve up to 1000x improvement in the number of memory accesses compared to their classical competitors.

Biased Range Trees

Instance-level Worst-Case Query Bounds on R-trees

Randomized Binary and Tree Search under Pressure

GDCRT: In-Memory 2D Geographical Dynamic Cascading Range Tree

Two trees are better than one

On Rotation Distance of Rank Bounded Trees

Independent Range Sampling on Interval Data (Longer Version)

Near-optimal Linear Decision Trees for k-SUM and Related Problems

Branches: A Fast Dynamic Programming and Branch & Bound Algorithm for Optimal Decision Trees

Scalable $k$-d trees for distributed data

PriPL-Tree: Accurate Range Query for Arbitrary Distribution under Local Differential Privacy

Path Query Data Structures in Practice

The height of record-biased trees

Binary Space Partitioning Forests

Multidimensional binary search trees used for associative searching

Orthogonal Range Searching on the RAM, Revisited

Space Efficient Multi-Dimensional Range Reporting

PK-Tree: A Spatial Index Structure for High Dimensional Point Data

On Range Summary Queries

First Tree-like Quantum Data Structure: Quantum B+ Tree