To understand the space requirements of QuiP
let me explain its architecture.
Quip is a standalone client software, which does
not require a Tamino installation.
Quip can retrieve data from Tamino by way of the ‘collection’ function.
Quip then
tries to see, which parts of a query followed by
a call to the collection function
can be made within the Tamino server.
All the data retrieved from Tamino, will then be
kept in the main memory.
I.e:
collection(“theater”)/PLAY[TITLE=“Cymberline”]
will retrieve just those documents from Tamino, matching the requested
title. The complete query is passed on to Tamino, wheras
let $x collection(“theater”)/PLAY
where $x/TITLE = “Cymberline”
return $x
will retrieve all PLAY documents from Tamino and filter them manually itself.
There is not very much that can be passed on directly to Tamino
(well Quip could do some smart query rewrite, but is does not…)
This is the case,
because Tamino does not yet know
about the let- and where-expressions, because they
are not part of XPath 1.0.
As you see, it is not easy and only in limited
cases possible to make a real data-base
applications with Quip. You have to make
queries in the way that the amount of data
processed by Quip is small and most
of the work can be done within the Tamino server.
When in doubt you can consult the log.txt file written by QuiP.
On a big machine you can try to encrease the
maximal heap size of the quip executor to
a very large value. This can be done in the
options dialoge within the QuiP gui.
Sven Eric
#Tamino#API-Management#webMethods