Minimum Property-Cut (MPC) is a vertex-centric RDF graph partitioning approach, where the objective function is to minimize the cuts to the number of distinct crossing properties. This approach can be used to avoid inter-partition joins in a wider set of SPARQL workloads in the context of distributed SPARQL query evaluation.
System Requirement: 64-bit linux server with GCC. We have tested on linux server with CentOS 7.4 x86_64 where the version of GCC is 5.4.0.
You can compile mpc in one command. Just run
g++ main.cpp -std=c++11 -o mpc
to compile the code and build executable "mpc".
mpc is used to partition an RDF graph.
RUN
./mpc rdf_triple_file_name output separator number_of_parts
Now, mpc only support RDF datasets in NT format. Each line is a triple of subject, property and object, where the tab "\t" or the space " " are used as the separator. If separator in the above command equals to "1", the separator is the space " "; if separator in the above command equals to "2", the separator is the tab "\t".
The partitioning results are written in the file named #output#InternalPoints.txt, where each line in the file consists of two parts: the first part is the URI of an resouce and the second part is the identifier of the partition that the resouce belongs to.
The benchmark queries used in our experimental evaluation exists in #queries# folder.
If you encounter any problems, please send emails to me (email address: hnu16pp@hnu.edu.cn).
The code is also contributed by Cen Yan (email address: yan_cen@hnu.edu.cn) and Chengjun Liu (email address: lcj2021@hnu.edu.cn).