The reverse key index (in Oracle) is simply a B-tree index with inverted keys (1234 - 4321). This can prevent unbalanced indexes if you add incrementing keys. It also makes it impossible to scan the range, so you should know what you are doing when using this.
The bitmap index is completely different from the B-tree index. You can think of it as a long bitmap for each key value, with one entry for each string that is set to true if the string has this value, and false if not. This works better (than B-tree indexes) for columns with several separate values (e.g. MALE, FEMALE). You can compress these bitmaps and then they will become very compact and fast for scanning.
The main problem with raster indexes is that it takes a lot of work to update them, so they are more suitable for storage scenarios where data is loaded into a night package and then only requested (and not changed) during the day.
Wikipedia has a page about raster indexes .
Thilo
source share