README.md

Experiments

This folder contains the material needed to reproduce the experiments described in the extended version of the paper.

In particular, it contains SQL queries used in these experiments (and instructions on how to run them).

Data and required tables

The code assumes two tables node_big and edge_big as input (see create table statements below; for the data see https://github.com/amirpouya/tpath/tree/main/data/contact) and a table exp_sizes where the results of the experiments are store (see create table statement below).

CREATE TABLE node_big (id INT, type VARCHAR, prop1 VARCHAR, prop2 VARCHAR, prop3 VARCHAR, prop4 VARCHAR, ts INT, te INT);
CREATE TABLE edge_big (id INT, src INT, dst INT, label VARCHAR, prop1 VARCHAR, ts INT, te INT);
CREATE TABLE exp_sizes (t TIMESTAMP, query VARCHAR, points bigInt, v1 bigInt, v1_coalesced bigInt, v2 bigInt, v2_coalesced bigInt, k INT, x INT);

The bash script run.sh is the main entry point and can be used to run the experiments, with different durations x and domain scaling factors k

The result is a out.txt file, for logging purposes. The output size of the queries is stored in table exp_sizes, where:

• points is the result size for using time points,
• v1 and v1_coalesced is the result size of U^{[t]} before and after coalescing respectively, and
• v2 and v2_coalesced is the result size of U^{[d]} before and after coalescing respectively.