Maximizing throughput

Output can process large XML data files that contain many records. You can instruct the Output service to create an output file for each record or a single output file that contains all records. (See Processing batch data to create multiple documents.)

Throughput limitations are not discreet and vary depending of the complexity of the form design, available memory, options chosen, and other activity. Errors that are generated when such a limitation is reached can identify lack of memory as being the problem. Be aware of this limitation, and use the Output service within number guidelines for safe page generation.

For a medium complexity form on a typical entry-level business server, the following approximate number guidelines for safe page generation were observed.

Format

Batch records/many small documents

Big record/one big document

PDF (not tagged) and PDF/a-1b

10000 pages

500 pages

PDF (tagged)

2500 pages

250 pages

PostScript

10000 pages

500 pages

PCL (using TrueType or printer-resident fonts)

10000 pages

500 pages

Select tagging for PDF only if it is required; tagged files are substantially larger than non-tagged files and take longer to render. For PCL printing, the use of Microsoft OpenType® fonts in form designs formats properly but decreases throughput, generates a larger output file, and does not deliver optimal performance. For increased performance, use or map properly licensed TrueType fonts or use printer-resident fonts.

When processing large data files or operating on a busy server, increase the Output service time-out; the default time-out is 180 seconds. To change the time-out value, ensure that hardware servers have adequate memory and the memory is available to the Java application server heap. For information about changing the time-out value, see LiveCycle Administration Console Help.

When processing a large record or one large document case, throughput is maximized when the XML data is structured so that a recordLevel run-time option of 2 can be used. For example, instead of structuring your data file in this manner:

<datafile> 
    <field>123</field>… 
</datafile>

Structure it this way, adding another level:

<datafile> 
    <record> 
        <field>123</field> 
        … 
    </record> 
</datafile>

// Ethnio survey code removed