Parallel Point-to-Point Shortest Paths and Batch Queries

2025-06-19

AI Summary

To address the inefficiency of point-to-point shortest path (PPSP) and batch PPSP queries on large-scale graphs, this paper proposes Orionet, a parallel framework. Methodologically, Orionet introduces: (1) the first query-graph-based abstraction for batch PPSP, unifying multi-source–multi-target query dependencies; (2) the systematic parallelization of classic algorithms—including bidirectional search and bidirectional A*—integrated with single-source shortest paths (SSSP) extension, heuristic pruning, and early termination; and (3) a lightweight, scalable batch-processing architecture. Experiments across 14 real-world graph datasets demonstrate that Orionet’s parallel bidirectional search achieves 2.9× and 6.8× speedup over GraphIt and MBQ, respectively, while its bidirectional A* variant attains 4.4× and 6.2× speedup. Moreover, Orionet significantly outperforms naive parallel baselines in batch query throughput and latency.

Technology Category

• Search and Optimization: Distributed Search
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Data Mining & Knowledge Management: Scalability, Parallel & Distributed Systems

Application Category

• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Social Networks and Social Media: Social mining and social search on the Web

Abstract

We propose Orionet, efficient parallel implementations of Point-to-Point Shortest Paths (PPSP) queries using bidirectional search (BiDS) and other heuristics, with an additional focus on batch PPSP queries. We present a framework for parallel PPSP built on existing single-source shortest paths (SSSP) frameworks by incorporating pruning conditions. As a result, we develop efficient parallel PPSP algorithms based on early termination, bidirectional search, A$^*$ search, and bidirectional A$^*$ all with simple and efficient implementations. We extend our idea to batch PPSP queries, which are widely used in real-world scenarios. We first design a simple and flexible abstraction to represent the batch so PPSP can leverage the shared information of the batch. Orionet formalizes the batch as a query graph represented by edges between queried sources and targets. In this way, we directly extended our PPSP framework to batched queries in a simple and efficient way. We evaluate Orionet on both single and batch PPSP queries using various graph types and distance percentiles of queried pairs, and compare it against two baselines, GraphIt and MBQ. Both of them support parallel single PPSP and A$^*$ using unidirectional search. On 14 graphs we tested, on average, our bidirectional search is 2.9$ imes$ faster than GraphIt, and 6.8$ imes$ faster than MBQ. Our bidirectional A$^*$ is 4.4$ imes$ and 6.2$ imes$ faster than the A$^*$ in GraphIt and MBQ, respectively. For batched PPSP queries, we also provide in-depth experimental evaluation, and show that Orionet provides strong performance compared to the plain solutions.

Problem

• Efficient parallel Point-to-Point Shortest Paths (PPSP) queries
• Batch PPSP queries leveraging shared information
• Performance improvement over existing parallel PPSP solutions

Innovation

• Parallel PPSP with bidirectional search and heuristics
• Framework extends SSSP with pruning conditions
• Batch PPSP via query graph abstraction