Skip to content

Welcome to Beehive

Beehive is Java based framework for parallel programming of graph applications using a transactional model of task execution. Many data analytics applications for large scale graph data require parallel processing utilizing a cluster computing environment. Parallelism in many graph problems tends to be fine-grained and irreguluar, and it is not easy to extract parallelism through static analysis and data partitioning. This is called amorphous paralleism.

Graph problems with amorphous paralleism cannot easily be partitioned for programming using the MapReduce model. The Beehive framework addresses this problem based on transactional model of parallelism programming. In Beehive, vertex-centric computation tasks for a problem are executed as serializable transactions using an optimistic model for concurrency control.

This Guidebook is based on the latest version of Beehive, Beehive 4.0


This work was supported by NSF Award 1319333. Computing resources were provided by NSF Award 1512877 and Minnesota Supercomputing Institute.

The Beehive system is based on the contributions by several people over a period of five years. The names of the main contributors are listed in the chronological order of their association with the project: Anand Tripathi, Vinit Padhye, Tara Sasank Sunkara, Jeremy Tucker, Indajith Premanath, BhagavathiDhass Thirunavukarasu, Varun Pandey, Charandeep Parisineti, Rahul Sharma, Tanmay Mehta, Manu Khandelwal, Henry Hoang


If you use this software for your work, please use the following citation:

author={A. Tripathi and V. Padhye and T. S. Sunkara and J. Tucker and B. Thirunavukarasu and V. Pandey and R. R. Sharma},
booktitle={2017 IEEE 10th International Conference on Cloud Computing (CLOUD)},
title={A Transactional Model for Parallel Programming of Graph Applications on Computing Clusters},
keywords={graph colouring;message passing;parallel programming;workstation clusters;parallel programming;graph coloring;single-source shortest path computation;transactional programming model;cluster computing systems;unstructured parallelism;vertex-centric computations;message passing;k-nearest neighbors;Computational modeling;Parallel processing;Parallel programming;Adaptation models;Message systems;Data models;Parallel computing;Distributed Systems;Cluster computing;Transaction models;Concurrency control;Graph problems},

GPL License

Beehive Parallel Programming Framework
© Regents of the University of Minnesota.
This software is licensed under GPL v3.0
( open source license.
For an alternative license please contact the Office for Technology
Commercialization at the University of Minnesota.