Supplementary material for the paper: "Native Execution of GraphQL Queries over RDF Graphs Using Multi-way Joins"
- Tentris (graphql endpoint): https://github.com/dice-group/tentris/tree/graphql-endpoint
- Hypertrie (implementation of the multi-way left join algorithm): https://github.com/dice-group/hypertrie/tree/left-join
To set up the experiments we provide an Ansible playbook.
Installation: https://docs.ansible.com/ansible/latest/installation_guide/intro_installation.html
- Replace
<ip_of_host>inansible_playbook/inventory.yamlwith the IP of the managed node (i.e., the server that will run the experiments). - Replace
<target_dir>inansible_playbook/roles/base/defaults/main.yamlwith the absolute path to the directory of the managed node that will store the experiments' required files. - Replace
<user>inansible_playbook/roles/base/defaults/main.yamlwith the username that will be used to login with to the managed node. - Install docker in the server that will run the experiments (https://docs.docker.com/engine/install/debian/) (Note: The user must be added to the docker group)
You can execute the playbook by issuing the following command:
ansible-playbook -kKi inventory.yaml playbook.yaml
Before running the benchmarks:
-
increase ulimit
ulimit -n 64000 -
set swappiness to 0
sudo sysctl vm.swappiness=0 -
prepare neo4j databases
cd <target_dir> sudo ./run_load_.sh
To run the benchmark execute the script run_graphql.sh with root privileges.
cd <target_dir>
sudo ./run_graphql.sh
The results of the benchmark (IGUANA result files) are stored in the directory <target_dir>/iguana_results/graphql
The datasets used in the benchmark, their respective queries and the GraphQL schemata are available for download here.
The results reported in the paper and the scripts to generate the plots are available for download here.
The pre-built binary of our implementation is available for download here.