Publication Date

8-2017

Date of Final Oral Examination (Defense)

5-11-2017

Type of Culminating Activity

Thesis

Degree Title

Master of Science in Computer Science

Department

Computer Science

Supervisory Committee Chair

Francesca Spezzano, Ph.D.

Supervisory Committee Member

Vijay Dialani, Ph.D.

Supervisory Committee Member

Amit Jain, Ph.D.

Supervisory Committee Member

Sole Pera, Ph.D.

Abstract

Graph is a commonly used data structure for modeling complex data such as chemical molecules, images, social networks, and XML documents. This complex data is stored using a set of graphs, known as graph database D. To speed up query answering on graph databases, indexes are commonly used. State-of-the-art graph database indexes do not adapt or scale well to dynamic graph database use; they are static, and their ability to prune possible search responses to meet user needs worsens over time as databases change and grow. Users can re-mine indexes to gain some improvement, but it is time consuming. Users must also tune numerous parameters on an ongoing basis to optimize performance and can inadvertently worsen the query response time if they do not choose parameters wisely. Recently, a one-pass algorithm has been developed to enhance the performance of these indexes in part by using the algorithm to update them regularly. However, there are some drawbacks, most notably the need to make updates as the query workload changes.

We propose a new index based on graph-coarsening to speed up query answering time in dynamic graph databases. Our index is parameter-free, query-independent, scalable, small enough to store in the main memory, and is simpler and less costly to maintain for database updates.

We conducted an extensive sets of experiments on two types of databases, i.e., chemical and social network databases, to compare our graph-coarsening based index vs. hybrid-indexes as follows. First, we considered no database updates or query workload changes (static graph databases) and compared the indexes according to query vi answering time and index size for different minSup values. Second, we compared the indexes in the case of dynamic graph databases, i.e. when graphs are added to or removed from the database. Third, we compared the indexes with regard to query workload changes. Fourth, we studied the scalability of our index vs. hybrid-indexes.

Experimental results show that our index outperforms hybrid-indexes (i.e. indexes updated with one-pass) for query answering time in the case of social network databases, and is comparable with these indexes for frequent and infrequent queries on chemical databases. Our graph-coarsening index can be updated up to 60 times faster in comparison to one-pass on dynamic graph databases. Moreover, our index is independent of the query workload for index update and is up to 15 times better after hybrid indexes are attuned to query workload for social network databases.

This work is also published in 26th ACM International Conference on Information and Knowledge Management (CIKM) held in Singapore[18].

DOI

https://doi.org/10.18122/B2042R

Download

Share

COinS