Subject: RE: New XSL Optimization From: S Manimaran <mani@xxxxxxxxxxxxxxxxx> Date: Wed, 23 Jun 1999 22:19:48 +0530 (IST) |
Hi, There is also another problem which slows down considerably. I have used an Earley parser which is written in C. It makes heavy use of pointers and the memory space required to store the Parse items is quadratic in the size of input. That means to parse a 10KB XML file you need 100 MB of memory space! Thus using Earley parser to find all the matching patterns at all nodes is trading too much with memory to gain performance in time!? But by the virtue of it needing too much memory, only slows down the running of it. -S.Manimaran, On Wed, 23 Jun 1999, Earl Bingham wrote: > > If you use IBM's C engine for this operation it will 100 times faster if > your concerned about speed. Otherwise use flat-file to database > conversion to speed up the procedures. > > - Earl > ------------------------------------------------------------------------------ URL::http://www2.csa.iisc.ernet.in/~mani/ Indian Internet Chess Server IICS::telnet 144.16.67.53 5000 Join frenchlist by sending a mail to french@xxxxxxxxxxxxxxxxx f16 Fingerprint16 = EC FF 4D 6C 3A 09 84 30 AA B5 F0 A2 38 A0 4C A4 XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list
Current Thread |
---|
|
<- Previous | Index | Next -> |
---|---|---|
RE: New XSL Optimization, Kay Michael | Thread | RE: New XSL Optimization, S Manimaran |
Re: Oracle XML/XSLT Technology "Bla, Christopher Conway | Date | Using a path expression stored in t, Jean-Christophe |
Month |