Hi Bruno.
I upgraded to Version 06.01.3.3127-pthreads linux and some queries
that ran on 4 seconds or less are now taking 45 seconds.
Also, the commercial version Version 06.02.3129-pthreads have
similar problems.
One query that got a lot worse was:
sparql define input:inference <http://semantica.globo.com/ruleset>
select distinct ?s ?label
{
{graph <http://semantica.globo.com/esportes/>
{
?s a <http://semantica.globo.com/esportes/
Jogo> ; ?p ?o .
?o bif:contains "'22/05/2011*'" .
?s rdfs:label ?label .
} }
UNION
{ graph <http://semantica.globo.com/> {
?s a <http://semantica.globo.com/base/
Evento> ; ?p ?o .
?o bif:contains "'22/05/2011*'" .
?s rdfs:label ?label .
} }
UNION
{ graph <http://semantica.globo.com/esportes/> {
?s a <http://semantica.globo.com/esportes/
Prova> ; ?p ?o .
?o bif:contains "'22/05/2011*'" .
?s rdfs:label ?label .
} }
UNION
{ graph <http://semantica.globo.com/esportes/> {
?s a <http://semantica.globo.com/esportes/Edicao_de_campeonato
> ; ?p ?o .
?o bif:contains "'22/05/2011*'" .
?s rdfs:label ?label .
} }
}
ORDER BY (?label)
LIMIT 10;
Is anyone experiencing the same problems ?
We are already in contact with Roosevelt on this query.
Please check with him for solution.
Patrick