Hello, we're running a sedna database with about 1.500.000 documents with dtSearch support.
We have tested it on different configuration and server (32 and 64bit) but the result is almost the same. We have a performance problem with full-text indexes. Each query takes a lots of seconds if we use a very common word like "legge" (or other words that have high frequence in body text). In detail we have used this query to make and extract data from a fulltext index: declare namespace s="http://www.schemata.it/lml/1.0";; declare namespace l="http://www.schemata.it/lml/1.0/linker";; create full-text index "ft_body" on document("schemata")/schemata/s:lml/s:*/s:body type "string-value" -------------------------- declare namespace s="http://www.schemata.it/lml/1.0";; declare namespace l="http://www.schemata.it/lml/1.0/linker";; count(ftindex-scan("ft_body","legge")) (21.034 sec) How we can do to make runs all faster ? We can try the sedna native full-text query ? (How ?) Thanks, Giocondo Sticca ------------------------------------------------------------------------------ What You Don't Know About Data Connectivity CAN Hurt You This paper provides an overview of data connectivity, details its effect on application quality, and explores various alternative solutions. http://p.sf.net/sfu/progress-d2d _______________________________________________ Sedna-discussion mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/sedna-discussion
