2012Unpublished venueRequires access

Benchmarking Traversal Operations over Graph Databases

Marek Ciglan, Alex Averbuch, Ladislav Hluchý

Open publisher page 74 citations

Abstract

A significant number of graph database systems has emerged in the past few years. Most aim at the management of the property graph data structure: where graph elements can be assigned with properties. In this paper, we address the need to compare the performance of different graph databases, and discuss the challenges of developing fair benchmarking methodologies. We believe that, compared to other database systems, the ability to efficiently traverse over the graph topology is unique to graph databases. As such, we focus our attention on the benchmarking of traversal operations. We describe the design of the graph traversal benchmark and present its results. The benchmark provides the means to compare the performance of different data management systems and gives us insight into the abilities and limitations of modern graph databases.

About this research paper

What this paper is about

A significant number of graph database systems has emerged in the past few years. Most aim at the management of the property graph data structure: where graph elements can be assigned with properties. In this paper, we address the need to compare the performance of different graph databases, and discuss the challenges of developing fair benchmarking methodologies. We believe that, compared to other database systems, the ability to efficiently traverse over the graph topology is unique to graph databases. As such, we focus our attention on the benchmarking of traversal operations. We describe the design of the graph traversal benchmark and present its results. The benchmark provides the means to compare the performance of different data management systems and gives us insight into the abilities and limitations of modern graph databases.

Why it matters

OpenAlex reports 74 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

A significant number of graph database systems has emerged in the past few years. Most aim at the management of the property graph data structure: where graph elements can be assigned with properties. In this paper, we address the need to compare the performance of different graph databases, and discuss the challenges of developing fair benchmarking methodologies. We believe that, compared to other database systems, the ability to efficiently traverse over the graph topology is unique to graph databases. As such, we focus our attention on the benchmarking of traversal operations. We describe the design of the graph traversal benchmark and present its results. The benchmark provides the means to compare the performance of different data management systems and gives us insight into the abilities and limitations of modern graph databases.

Key concepts: Graph database, Benchmarking, Computer science, Tree traversal, Traverse, Graph traversal, Wait-for graph, Graph

Related papers

Back to paper searchBrowse research topicsOriginal source
Benchmarking Traversal Operations over Graph Databases — Research Paper | ScholarLens