gtpi1m6c | System Installation Support Reference |
For each ID from which data is to be retrieved, a hash value is first computed by dividing the desired record ID by the size of the hash table and saving the remainder. By multiplying this remainder by 4, the correct pointer into the second part of the RIAT table is accessed.
Once the second part of the RIAT table is accessed, the desired ID is compared against the ID corresponding to the RIAT entry currently being accessed. If these IDs match, the correct entry has been found. If the IDs do not match, the synonym chain pointer is used to access another RIAT entry. By continuing to follow this scheme, the correct entry is eventually located.