Marklogic : Query response time is very high -
Marklogic : Query response time is very high -
i have around 15000 records in xml format in uri, say:"documents/products/specs/*.xml
". each xml of around 25 kilo bytes in size. connected marklogic server uisng remote apache tomcat server has xcc client
(java), tries execute adhocquery
resembles this:
let $a := cts:uri-match('documents/products/specs/*.xml') $xml in $a homecoming fn:doc($a)
(for loop implemented in java).
this works fine. records larger count, 15000 record, takes 60 minutes, when server , net speed good. (total size of document in uri around 20 mb should not take more 20 minutes).
is there workaround ?
what doing requesting total body of documents. not typical query, rather db dump. query show buffer data, send through tomcat 1 time again buffers info , send you. big dataset sending in 1 request.
what intent of query ? if want documents should either dump them out using programme mlcp or fetch them in smaller batches first collecting uri's fetching documents. can speeded substantially doing document fetching in parallel. can see examples of java source in xmlsh shows how fetch documents in xcc in parallel
http://xmlsh.svn.sourceforge.net/viewvc/xmlsh/extensions/marklogic/src/org/xmlsh/marklogic/get.java?revision=792&view=markup
my guess (correct me if wrong) experimenting , dont need docs. in case more realistic query should tried.
marklogic
Comments
Post a Comment