Partitioning *any* table is generally possible.

The more important question is why, and this is generally a combination of business requirements and/or technical requirements.

For example, there is nothing wrong with a 180GB table that is not partitioned...*unless* there is something about it that makes a case for *needing* partitioning.

This could be:

a) it is hard to manage an object that size
b) you might to manage different parts of the data in different way (eg compress old data, or have some of it read only)
c) you might get some performance benefits from a partitioning scheme.

Generally its (c) that tends to motivate people toward partitioning, and this is also what governs the *how* you partition. If you partition on a YEAR column, but every query never provides that column in queries, then its unlikely you'll any performance benefit for those queries.

So focus on the *business needs* and your current *pain points*. They are the things that will determine *if* you need to partition, and if so, *how* to partition the data.