Hi Akshey,
the first thing’s to say is: Sridharan is quiet right.
So for the better understanding, let’s specify what’s what and really means.
OLAP
Known as the short form of Online-Analytical-Processing. As You mentioned right, things like slice&dice, cubes, pivoting, dill-up&drill-down.
In other words, this only describes what You want to do - in a first descibtion-step.
The seccond is, that OLAP also means, that the analyst uses his own business-language for his query- and anlalytical job. So the “naming” of those enteties he’s “looking” on is quiet different from those used within the “technical”-view of relational databases.
The main-aim of OLAP is the multi-dimensional view on data, basing on whatever DB.
ROLAP
That means Relational OLAP. Within a ROLAP inviroment the DB helds its data not in a multidimensional form, but in relational DBs. The Sence of multidimendionality is a simulation. So there are direct SQL-requests against relational DBs. Those DBs regulary use special data schemes, like Star- or Snow-Flake-Schemes.
MOLAP
Multidimensional OLAP. This is the “thing” which really stores data in multidimensional manner. That means that data are physically stored as multidimensional cubes.
HOLAP
This is the so named Hybrid OLAP.
The reason for this was the empirial knowlegde, that sometimes a sheer MOLAP is not very prolific for all business questions. So You get a combination of real MOLAP as one part and a real ROLAP as the other.
DOLAP
This only means Desktop OLAP. This most of the time describes a client (Desktop) which get’s it’s data from the database, most of the time relational DB’s, keeps those data offline and forms a OLAP-cube, with the a.m. features like drill… and so on.
I hope, that, at this point, it becomes clear, what I mentioned on top of my posting…
Sridharan is quiet right. 
MJGrass
(BOB member since 2004-09-09)