GFTAXTR has two objectives:

1. To reduce the quantity of input data so that GFTAVMA can be run iteratively using a DASD data set as input. Because the GFTAVMA analysis is an iterative process, using SMF records directly as input would be inefficient and costly in tape mounts.
2. To preprocess the SMF data by filling in missing record information and re-constructing the flow of the job that created each data set. GFTAXTR establishes all the inter-relationships between the data set and volume records, which enhances the performance of each GFTAVMA run.