Language-aware Indexing for Conjunctive Path Queries

被引:1
|
作者
Sasaki, Yuya [1 ,2 ]
Fletcher, George [3 ]
Makoto, Onizuka [1 ]
机构
[1] Osaka Univ, Suita, Osaka, Japan
[2] JST PRESTO, Kawaguchi, Saitama, Japan
[3] Eindhoven Univ Technol, Eindhoven, Netherlands
来源
2022 IEEE 38TH INTERNATIONAL CONFERENCE ON DATA ENGINEERING (ICDE 2022) | 2022年
关键词
Graph databases; index; bisimulation; SUBGRAPH ISOMORPHISM; XML;
D O I
10.1109/ICDE53745.2022.00054
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Conjunctive path queries (CPQ) are one of the most frequently used queries for complex graph analysis. However, current graph indexes are not tailored to fully support the power of query languages to express CPQs. Consequently, current methods do not take advantage of significant pruning opportunities during CPQ evaluation, resulting in poor query processing performance. We propose the CPQ-aware path index CPQx, the first path index tailored to the expressivity of CPQ. CPQx is built on the partition of the set of source-target vertex pairs of paths in a graph based on the structural notion of path-bisimulation. Path-bisimulation is an equivalence relation on paths such that each partition block induced by the relation consists of paths in the graph indistinguishable with respect to CPQs. This language-aware partitioning of the graph can significantly reduce the cost of query evaluation. We present methods to support the full index life cycle: index construction, maintenance, and query processing with our index. We also develop interest-aware CPQx to reduce index size and index construction overhead while accelerating query evaluation for queries of interest. We demonstrate through extensive experiments on 14 real graphs that our methods accelerate query processing by up to multiple orders of magnitude over the state-of-the-art methods, with smaller index sizes. Our complete C++ codebase is available as open source for further research.
引用
收藏
页码:661 / 673
页数:13
相关论文
共 50 条
  • [1] Conjunctive Regular Path Queries with String Variables
    Schmid, Markus L.
    PODS'20: PROCEEDINGS OF THE 39TH ACM SIGMOD-SIGACT-SIGAI SYMPOSIUM ON PRINCIPLES OF DATABASE SYSTEMS, 2020, : 361 - 374
  • [2] Conjunctive Regular Path Queries with Capture Groups
    Schmid, Markus L.
    ACM TRANSACTIONS ON DATABASE SYSTEMS, 2022, 47 (02):
  • [3] Conjunctive Regular Path Queries under Injective Semantics
    Figueira, Diego
    Romero, Miguel
    PROCEEDINGS OF THE 42ND ACM SIGMOD-SIGACT-SIGAI SYMPOSIUM ON PRINCIPLES OF DATABASE SYSTEMS, PODS 2023, 2023, : 231 - 240
  • [4] Expressiveness and static analysis of extended conjunctive regular path queries
    Freydenberger, Dominik D.
    Schweikardt, Nicole
    JOURNAL OF COMPUTER AND SYSTEM SCIENCES, 2013, 79 (06) : 892 - 909
  • [5] SEMANTIC TREE-WIDTH AND PATH-WIDTH OF CONJUNCTIVE REGULAR PATH QUERIES
    Figueira, Diego
    Morvan, Remi
    LOGICAL METHODS IN COMPUTER SCIENCE, 2025, 21 (01) : 1 - 60
  • [6] BranchGuide: An Indexing Technique for Efficient, Lossless Processing of Branching Path Queries
    Brito, Talles
    Elias, Gledson
    30TH ANNUAL ACM SYMPOSIUM ON APPLIED COMPUTING, VOLS I AND II, 2015, : 1086 - 1092
  • [7] Conjunctive queries over trees
    Gottlob, Georg
    Koch, Christoph
    Schulz, Klaus U.
    JOURNAL OF THE ACM, 2006, 53 (02) : 238 - 272
  • [8] When Is Approximate Counting for Conjunctive Queries Tractable?
    Arenas, Marcelo
    Alberto Croquevielle, Luis
    Jayaram, Rajesh
    Riveros, Cristian
    STOC '21: PROCEEDINGS OF THE 53RD ANNUAL ACM SIGACT SYMPOSIUM ON THEORY OF COMPUTING, 2021, : 1015 - 1027
  • [9] MODULAR PATH QUERIES WITH ARITHMETIC
    Michaliszyn, Jakub
    Otop, Jan
    Wieczorek, Piotr
    LOGICAL METHODS IN COMPUTER SCIENCE, 2020, 17 (03) : 27:1 - 27:36
  • [10] Containment of Regular Path Queries Under Path Constraints
    Salvati, Sylvain
    Tison, Sophie
    27TH INTERNATIONAL CONFERENCE ON DATABASE THEORY, ICDT 2024, 2024, 290