From: Ghislain A. <ghi...@gm...> - 2017-03-20 12:22:19
|
Hi again, I am trying to use this framework [1] to make a small benchmark on a dataset from the publication office [a]. I really don't understand why I am getting some very long time for results when they are fast in the workbench. So, I think I am doing something wrong. Probably I was not able to set the 1- The loading time was extremely long (almost 3 days to load 727 Mio dataset) 2- During each warm up, the results are taking more and more time to execute. 3- Based on the indication here [2] to get in touch with the team, I really need to understand what I am missing. Best, Ghislain [a] http://publications.europa.eu/ [1] https://sourceforge.net/p/sparql-query-bm/wiki/CLI/ [2] https://wiki.blazegraph.com/wiki/index.php/SPARQL_Benchmarks -- ------- "Love all, trust a few, do wrong to none" (W. Shakespeare) Web: http://atemezing.org |