Towards a property graph generator for benchmarking

Abstract

The use of synthetic graph generators is a common practice among graph-oriented benchmark designers, as it allows obtaining graphs with the required scale and characteristics. However, finding a graph generator that accurately fits the needs of a given benchmark is very difficult, thus practitioners end up creating ad-hoc ones. Such a task is usually timeconsuming, and often leads to reinventing the wheel. In this paper, we introduce the conceptual design of DataSynth, a framework for property graphs generation with customizable schemas and characteristics. The goal of DataSynth is to assist benchmark designers in generating graphs efficiently and at scale, saving from implementing their own generators. Additionally, DataSynth introduces novel features barely explored so far, such as modeling the correlation between properties and the structure of the graph. This is achieved by a novel property-to-node matching algorithm for which we present preliminary promising results.

Power-Law Distributed Graph Generation With MapReduce

“…DataSynth [2] is a framework for property graph generation with customizable schemas and characteristics. Data-Synth is based on a property-to-node matching algorithm that allows to resemble real-life characteristics, like correlation between properties and the structure of the graph.…”

“…G RAPHS are a recognized abstraction model as they can be used to represent structured and semi-structured data occurring in many application domains [1]. A graph generator is a tool which allows to generate graphs resembling (ideally) characteristics found in real world networks, such as a power-law distribution, a large clustering coefficient, a small diameter or a community structure [2]. There are several methods and tools for graph generation [3], [4], most of them designed to run in a single computer.…”

Automated generation of consistent, diverse and structurally realistic graph models

“…As noted in [84], the cumulative distribution function of an object property contains more representative information than simply the mean or the average. Moreover, [45,57,62,90] also use cumulative distributions for comparing various graph properties.…”

R3MAT: A Rapid and Robust Graph Generator

“…In order to cover the above restriction, we develop a hash-based implementation of an array whose java code is shown in Algorithm 6. Specifically, we create the java class LargeArray which uses an object of type HashMap<Long,Long> to implement an array as a hash table 5 . Hence, a key-value pair (i, } } this implementation is that the memory space is allocated on demand, and the maximum number of elements is extended to the maximum value of Long, i.e.…”

“…One of the main problems when developing graph-based applications is the availability of large and representative datasets, because data is a very valuable resource for organizations. The lack of real graphs has motivated the development of methods and tools for generating synthetic graphs [5], most of them based on random procedures [6].…”