DeepAI AI Chat
Log In Sign Up

Experimental Analysis of Distributed Graph Systems

by   Khaled Ammar, et al.

This paper evaluates eight parallel graph processing systems: Hadoop, HaLoop, Vertica, Giraph, GraphLab (PowerGraph), Blogel, Flink Gelly, and GraphX (SPARK) over four very large datasets (Twitter, World Road Network, UK 200705, and ClueWeb) using four workloads (PageRank, WCC, SSSP and K-hop). The main objective is to perform an independent scale-out study by experimentally analyzing the performance, usability, and scalability (using up to 128 machines) of these systems. In addition to performance results, we discuss our experiences in using these systems and suggest some system tuning heuristics that lead to better performance.


page 9

page 10

page 11


A Comparison of Parallel Graph Processing Implementations

The rapidly growing number of large network analysis problems has led to...

TurboGraph++: A Scalable and Fast Graph Analytics System

Existing distributed graph analytics systems are categorized into two ma...

COST of Graph Processing Using Actors

Graph processing is an increasingly important domain of computer science...

Graph3S: A Simple, Speedy and Scalable Distributed Graph Processing System

Graph is a ubiquitous structure in many domains. The rapidly increasing ...

IOPathTune: Adaptive Online Parameter Tuning for Parallel File System I/O Path

Parallel file systems contain complicated I/O paths from clients to stor...