In simple words... when you work in JDE, your requests will provide quick output. e.g. if your table search in databrowser is based on the index within a table, you will find the data appears quickly.
e.g. a table when created initially, it is created with primary key constraints and one or many indexes (Logical files), so that application or reports runs faster..
You can do reindexing from front-end using JDE application or back-end using RDBMS.
While reindexing, indices are being dropped and created afresh...
There should not be any locks on objects or should not be in use while performing this activity ...
Hope this clears you....
