Temporary File Storage

When used to sort, presort, double-sort, or match documents, Enrichment copies entire input print streams to temporary storage to enable random retrieval after the sort. Enrichment uses extended memory, Hiperspace memory, and one or more entry-sequenced data set (ESDS) VSAM files for temporary storage.

Note: Enrichment only caches input print streams if you are sorting, presorting, or double-sorting documents. In all other configurations, Enrichment processes documents one at a time and does not use Hiperspace, VSAM, or disk.

To reduce disk I/O, Enrichment uses memory (above the 16MB) to temporarily store inputs for processing. If Enrichment exceeds a user-specified amount of memory, it then uses Hiperspace memory to hold the inputs, when available. If Hiperspace is not available, is exceeded, or if a user-specified amount of Hiperspace is exceeded, Enrichment uses ESDS VSAM files for temporary storage. In many cases, you will not need to use VSAM or disk storage with Enrichment. However, if an input is very large, you can specify that it be stored exclusively in Hiperspace or VSAM or disk. Set the <IO> tag to HIPERSPACE to use Hiperspace exclusively, to DISK to use VSAM or disk exclusively, or to TRANSFER to use Hiperspace then VSAM.

Note: Make sure to set the <SINGLEBUFFER> tag when doing large document processing. <SINGLEBUFFER> allows you to define the initial size and growth factor for the single document buffer.

Use the Input group <TEMPDISK> tag to identify VSAM or disk files to Enrichment. If it is possible that your input data could exceed the capacity of a single ESDS VSAM file, you can use multiple <TEMPDISK> tags to specify multiple files.

Insert pages can be stored in memory, Hiperspace, or disk. Or, if uniquely set for each document, the insert pages may not be stored at all by using <IO> NONE.