This presentation will show a comparison, for multiple benchmarks, of different strategies when it comes to executing DistRDF tasks. We will see the difference in performance between non-optimized and optimized jitting. We will also see if it pays off to generate C++ code for the RDF graph on the fly in the DistRDF workers and compile it, as opposed to running the graph in Python.
Speakers:
Enric Tejedor Saavedra
(CERN),
Enrico Guiraud
(EP-SFT, CERN),
Vincenzo Eduardo Padulano
(Valencia Polytechnic University (ES))