Resources of "Declarative RDF Graph Construction: Materialization or Virtualization?"

Resources for the paper "Declarative RDF Graph Construction: Materialization or Virtualization?", currently under double-blind review, thus all information about the authors is redacted.

• GTFS-MADRID-BENCH.tar.xz: GTFS Madrid Bench with scales 1, 25, 50
• BSBM.tar.xz: BSBM benchmark with scales 5000, 10000, 15000
• SANTA:
  • PARALLEL-LOADING.tar.xz: parallel loading parameter data with 1, 8, 16, 24, 32 cores.
  • DUPLICATES.tar.xz: duplicate values parameter data with 0%, 25%, 50%, 75% and 100% of the data containing duplicates.
  • EMPTY-VALUES.tar.xz: empty values parameter data with 0%, 25%, 50%, 75% and 100% of the data containing empty values.
  • RAW-DATA.tar.xz: scaling among data records & properties parameter data with 1000, 5000, 25000, 125000 records and 1, 5, 10, 15 properties.
  • JOIN-1_1.tar.xz: Join 1-1 data.
  • JOIN-1-N.tar.xz: Join 1-N data with relationship N: 3, 5, 10, 15.
  • JOIN-N-1.tar.xz: Join N-1 data with relationship N: 3, 5, 10, 15.
  • JOIN-N-M.tar.xz: Join N-M data with relationships N-M: 3-3, 3-5, 5-3, 10-5, 5-10.
  • JOIN-DUPLICATES.tar.xz: Join duplicates data with 0%, 25%, 50%, 75%, 100% of the joins generating duplicates.
  • MAPPINGS.tar.xz: Triples Maps & Predicate Object Maps mapping rules impact parameter data with 1 TM + 15 POMs, 15 TMs + 1 POM, 3 TMs + 5 POMs, 5 TMs + 3 POMs.
• Graphs
  • GTFS-graphs.pdf: GTFS-Madrid-Bench query execution time graph in full resolution
  • BSBM-graphs.pdf: BSBM query execution time graph in full resolution

Note: These archives are heavily compressed, make sure you have enough disk space to unpack them.

Our bench-executor tool can execute these automatically for you, you need to set the --root parameter to the root directory of the unpacked archive and execute them with the run command or list them all with the list command. Extensive instructions on how to use the tool are available in the README of the linked repository.