A bit late, the answer might still be valid for others. 100K records in one go is performance wise a potential problem. Next to this there might be a time out problem if an Excel migration job is used (can't remember if that functionality was fully implemented for IFS7.5).
Open your spreadsheet, include the wanted HS code per inventory part, save the file as csv/txt.
Build migration job A which will load the data. Best to push the csv/txt file to connected folder/directory that can be used in migration job A. Run job A in the background by using RMB to schedule it.
Build migration job B that will read the loaded file and update the inventory part with the new HS code.
If you have multiple sites, you might prepare a replication job and have the new HS code automatically copied to the other site/inventory combinations.
Start with a small set (10 rows). First use the migration jobs interactively. If they work (debugging is easier if they run interactive), schedule it. If that works, step up to use the large file.