Distributed join processing on social network data

Structure

This project consists of 3 folders: /data: includes data files: facebook.data twitter.dat dblp.dat data.txt(made manually) /src: includes all header and source code files /test_command: includes arguments for executing programs

In the root directory, a Makefile is provided to compile and execute programs.

Compiling

For compiling programs, run: make All. Alternatively, for compiling each algorithm individually, run: make Seq_join make Distr_join make Multi_join make Hypercube

Testing

For testing the performance, we provides Makefile items for each algorithm and each data file.

Testing sequential algorithm for 4 data files:

make seq_join_test make seq_join_fb

Testing distributed join algorithm with single machine

make distr_join_test_loc make distr_join_fb_loc make distr_join_tw_loc make distr_join_db_loc

Testing distributed join algorithm with machine cluster

make distr_join_fb make distr_join_tw make distr_join_db

Testing multi way join algorithm with single machine

make multi_join_fb_loc make multi_join_tw_loc make multi_join_db_loc

Testing multi way join algorithm with machine cluster

make multi_join_fb make multi_join_tw make multi_join_db

Testing hypercube join algorithm with machine cluster

make hypercube_fb make hypercube_tw make hypercube_db

Testing searching complete subgraph of 4 vertices

make hypercube_fb4 make hypercube_db4

Delete all compiled files

make clean