GraphSC is a parallel secure computation framework that supports graph-parallel programming abstractions resembling GraphLab. GraphSC is suitable for both multi-core and cluster-based computing architectures. Link for the paper.

git clone https://github.com/kartik1507/GraphSC.git

cd GraphSC/bin/

./compile.sh

./runOne.py

e.g. ./runOne.py pr.PageRank 16 2

The above example will run the PageRank example using 2 garblers and 2 evaluators on the same machine. The configuration for running garblers and evaluators on a cluster can be found in machine_spec/. The input files are stored in in/.

In case of any queries, please contact Kartik Nayak (kartik@cs.umd.edu)