Queries created couple of years ago could be more slow today just because the size of the tables queried increased, those tables could require an index?
Before jumping in the detail of the SAS Indexes, an overview of it is way of working will be described:

• Where are stored the indexes and what do they contains?
• Once created, are they used by default?
• What are the different kind of indexes?

We will then see how we can create or delete them using Proc SQL or Data step or Proc Datasets:

• Create simple index
• Create composite index
• Create unique index

Finally, we will see different options allowing you to use them. Via proc sql, it is quite straightforward but with datastep it is not that obvious and this will force us to review the different types of data access: sequential, direct (POINT and MODIFY) or direct via index (KEY and MODIFY). Part of this tips and tricks will be spent on match merge technics in datastep using the _IORC_, _SOK and _SDENOM.