Semantic Query Transformations for Increased Parallelization in Distributed Knowledge Graph Query Processing

Ontologies have become an increasingly popular semantic layer for integrating multiple heterogeneous datasets. However, significant challenges remain with supporting efficient and scalable processing of queries with data linked with ontologies (ontological queries). Ontological query processing quer...

Full description

Saved in:
Bibliographic Details
Published in:SC19: International Conference for High Performance Computing, Networking, Storage and Analysis pp. 1 - 14
Main Authors: Kim, Hyeongsik, Bhattacharyya, Abhisha, Anyanwu, Kemafor
Format: Conference Proceeding
Language:English
Published: ACM 17.11.2019
Subjects:
algorithms
cloud workflow
data analytics and frameworks supporting data analytics
data and resource management including dynamic resource provisioning
data management
Dynamic scheduling
Graph and network algorithms
Heuristic algorithms
High performance computing
improved models
Knowledge graphs
metadata
namespaces
Ontologies
Parallel processing
performance or scalability of specific applications and respective software
Query processing
Resource management
Scalability
Scalable storage
Semantics
ISSN:2167-4337
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Ontologies have become an increasingly popular semantic layer for integrating multiple heterogeneous datasets. However, significant challenges remain with supporting efficient and scalable processing of queries with data linked with ontologies (ontological queries). Ontological query processing queries requires explicitly defined query patterns be expanded to capture implicit ones, based on available ontology inference axioms. However, in practice such as in the biomedical domain, the complexity of the ontological axioms results in significantly large query expansions which present day query processing infrastructure cannot support. In particular, it remains unclear how to effectively parallelize such queries. In this paper, we propose data and query transformations that enable inter-operator parallelism of ontological queries on Hadoop platforms. Our transformation techniques exploit ontological axioms, second order data types and operator rewritings to eliminate expensive query substructures for increased parallelizability. Comprehensive experiments conducted on benchmark datasets show up to 25X performance improvement over existing approaches.
ISSN:2167-4337
DOI:10.1145/3295500.3356212