FOP0.94 and FOP0.95beta use more memory than FOP0.20.5
Hi all, I tested FOP-0.94, FOP-0.95beta and FOP-0.20.5 using pdfoutline.fo. I added more fo:block(simple text) to the pdfoutline.fo file until the size of this file is reached 7.74M. Both FOP-0.94 and FOP-0.95beta require at least 896M heap size to run and produce 1580 pages PDF document. However, FOP-0.20.5 only require 64M to run. That's a huge different. I tested with Sun JDK1.4.2 and Sun JDK1.5_06. FOP-Devepers: can you please look into the FOP process and see why new re-design takes much more memory than old FOP-0.20.5. -- View this message in context: http://www.nabble.com/FOP0.94-and-FOP0.95beta-use-more-memory-than-FOP0.20.5-tp17188811p17188811.html Sent from the FOP - Users mailing list archive at Nabble.com. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: FOP0.94 and FOP0.95beta use more memory than FOP0.20.5
On May 12, 2008, at 16:37, LN wrote: Hi I tested FOP-0.94, FOP-0.95beta and FOP-0.20.5 using pdfoutline.fo. I added more fo:block(simple text) to the pdfoutline.fo file until the size of this file is reached 7.74M. Both FOP-0.94 and FOP-0.95beta require at least 896M heap size to run and produce 1580 pages PDF document. However, FOP-0.20.5 only require 64M to run. That's a huge different. I tested with Sun JDK1.4.2 and Sun JDK1.5_06. FOP-Devepers: can you please look into the FOP process and see why new re-design takes much more memory than old FOP-0.20.5. Simple text-layout has become a lot more memory-consuming than the previous version, but should also lead to much better-looking results. How exactly did you alter the document? By putting a lot of text in a single fo:block? By putting all of the fo:blocks in one page-sequence? FOP 0.9x has a much more complex layout-algorithm that works at optimum efficiency if you use relatively small to middle-sized fo:blocks (corresponding to paragraphs), and relatively small to medium-sized page-sequences (corresponding to chapters in your document) Cheers Andreas - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Memory issue
Jean-François El Fouly a écrit : So I hired the team member who's competent in profiler usage next week but I must say at the moment I'm still stuck :-( The sysadmins made a tarball from the staging server and copied everything to a similar server that has full profiling instrumentation (with JProfiler). And obviously there the application works and memory usage is quite normal. Completely different behaviour. Very, very strange. I'll try to understand what's happening (well, I badly need to) but it's probably not going to be easy :-(
Re: FOP0.94 and FOP0.95beta use more memory than FOP0.20.5
Andreas, I added multiple fo:blocks, each fo:block contains the same text (Exact text: Rather than performing transformation with an XSLT before invoking FOP, it is possible, if you use XT as your XSLT engine, to just call FOP and have it call XT for you. To do this, run the class org.apache.fop.apps.CommandLine with the source XML file name, XSL file name and PDF file name as arguments. You will need to include FOP, SAX, your SAX Parser and XT in your classpath and so you might invoke). All these fo:blocks are in the same page-sequence. Andreas Delmelle-2 wrote: On May 12, 2008, at 16:37, LN wrote: Hi I tested FOP-0.94, FOP-0.95beta and FOP-0.20.5 using pdfoutline.fo. I added more fo:block(simple text) to the pdfoutline.fo file until the size of this file is reached 7.74M. Both FOP-0.94 and FOP-0.95beta require at least 896M heap size to run and produce 1580 pages PDF document. However, FOP-0.20.5 only require 64M to run. That's a huge different. I tested with Sun JDK1.4.2 and Sun JDK1.5_06. FOP-Developers: can you please look into the FOP process and see why new re-design takes much more memory than old FOP-0.20.5. Simple text-layout has become a lot more memory-consuming than the previous version, but should also lead to much better-looking results. How exactly did you alter the document? By putting a lot of text in a single fo:block? By putting all of the fo:blocks in one page-sequence? FOP 0.9x has a much more complex layout-algorithm that works at optimum efficiency if you use relatively small to middle-sized fo:blocks (corresponding to paragraphs), and relatively small to medium-sized page-sequences (corresponding to chapters in your document) Cheers Andreas - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] -- View this message in context: http://www.nabble.com/FOP0.94-and-FOP0.95beta-use-more-memory-than-FOP0.20.5-tp17188811p17189264.html Sent from the FOP - Users mailing list archive at Nabble.com. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: FOP0.94 and FOP0.95beta use more memory than FOP0.20.5
On May 12, 2008, at 16:57, LN wrote: snip / All these fo:blocks are in the same page-sequence. That explains a lot. This is a known scalability issue. FOP has problems with very large page-sequences, as it is currently based on a total-fit approach for the entire page-sequence. If you divide the content over, say, 16 page-sequences of +/-100 pages, you should already notice a difference Something else that helps is using strategically placed forced breaks (break-before=page). HTH! Andreas - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
