Large Report Formatting
Produce very large print files — 100,000-plus pages of statements, invoices, and batch reports — as PDF, PostScript, or AFP through parallel splitting, formatting, and joining.
Some print jobs do not fit in one pass: a monthly statement run, for example, where the input is 50,000 two-page invoices in a single XML file or spread across many. Formatting a document collection like that as one job is slow and memory-hungry, and the output file can exceed what a single engine run handles comfortably.
Split, format in parallel, join
RenderX solves this by breaking the pipeline into stages — transformation, formatting, and generation — and running the formatting stage in parallel. The intermediate RenderX XML format produced by the engine can be split, reordered, and joined, so a large run becomes many small, independent jobs that recombine into one output file of 100,000 or more pages, delivered as PDF, PostScript, or AFP.
VDPMill packages the whole pipeline without custom code. Its Splitter divides the input XML into chunks, its Formatter drives multiple IRen engines — including several threads at once for simultaneous multi-format output — and its Joiner tracks document order and assembles the final print file. For grid deployments behind a REST API, IRen InGrid distributes the same formatting work across engines with load balancing and failover.
One caveat from experience: these techniques suit document collections such as bills, statements, and benefit confirmations, where page numbering does not run across documents. They are not intended for a single huge manual or textbook that relies on PDF bookmarks, back-of-book indexes, and cross-references — those constructs require the whole document in memory.
The concepts are described in detail in our white paper, Large Report Formatting (PDF). Questions about your volumes? Contact sales.