# Types of OLAP Cubes and Operations There are three different types of cubes. the first one is MOLAP, ROLAP, HOLAP. We also called M-OLAP MOLAP; this stands for multi-dimensional online analytic processing. ROLAP stands for relational online analytical processing and HOLAP stand for hybrid online analytical processing.

OLAP cubes

OLAP cubes where your data will be stored. Now the analysis that will do the kind of queries that you run there will be OLAP queries, and they will all be on the multidimensional data. So, the data that is going to be stored inside your cube is going to be multidimensional data. But the other multi-dimensional data where they want to get stored.

There are three different types of places where you can sort your multi-dimensional data. So, there are three different types of OLAP cubes.

MOLAP

So, the first type is a multidimensional OLAP cube. When we say multidimensional OLAP then this is your default type of OLAP cube. So, here MOLAP is the form of OLAP that processes and stores data directly into a multidimensional database. So, you have a multidimensional cube and then your data will be stored inside that database. The advantage here is that it will give you good performance and it can perform complex calculations, but the problem is an only a limited amount of data can be handled in your MOLAP. But then there is a difference between MOLAP and ROLAP.  Now that is where ROLAP scores MOLAP.

ROLAP

ROLAP stands for relational online analytical processing. Now ROLAP is the form of OLAP that performs dynamic multi-dimensional analysis of data stored in relational database rather than in a multidimensional database. Now what this means is that in your MOLAP you have your multidimensional data you will be storing inside a multidimensional database this will be your OLAP cubes. You have your data that we store inside of multidimensional database. Thus, making it a multidimensional data and that is one type of OLAP that is multidimensional OLAP. But relational OLAP is getting that multidimensional data converting it into relational data and then stored inside relational database. So that is what it says here is the process of storing data in a relational database so in your ROLAP you will have your multidimensional data stored inside your relational database and then you will be running multidimensional analysis and multidimensional queries on a relational database.

So, queries that will be running your data is basically going to be multidimensional. Your queries that will be running will be the same OLAP queries. But the difference is the place where it’s stored. The data in the case of a ROLAP is stored in a relational database. But whereas is in the case for multidimensional OLAP it is stored inside of multidimensional database itself.

Why ROLAP can be used instead of MOLAP

Because a greater amount of data can be processed in this case. But in the case of MOLAP, only a limited amount of data can be handled at any point in time. But the problem is it requires more processing time and a lot of disc space. Now it needs more processing time because you are going to convert your multidimensional data into a relational data and once you convert that you must then store it inside a relational database. Now, this is certainly more time consuming than you’re a multidimensional OLAP.

So, that is one disadvantage with your relational OLAP and your amount of disc space that will be occupied because of all these processes going to be greater. Now that is the disadvantage but of course that comes with the benefit, so you’re getting something out of using this. So that is the difference between multidimensional OLAP and relational OLAP and then these are the two basic differences.

HOLAP

Then you have a third one that is called a hybrid OLAP. So, your hybrid OLAP is basically a combination of both are MOLAP and ROLAP. So, the positives and your advantages of a fourth is using your hybrid overlap. So, the advantage of HOLAP, it can drill through from the cube into the underlying relational data. So, what it means is you will have your cube here and you have the underline relational data. So, using HOLAP you can drill through into the relational data using your cube. so that’s the thing about the HOLAP which makes use of the best features of both your multi-dimensional OLAP and your relational OLAP. These are the three different types of overlap cubes.

OLAP Operations

Roll up, drill down, slice, dice, and Pivot. These are the five different overlap operations that we can do on our dimensional data.

Roll Up:

This is to know what kind of operations that you can do on your warehousing and things that you can do on your database. So, we have will happen since data stored in such a multi-dimensional fashion these kinds of options can be performed. Roll up is something that forms aggregation or data cube by either climbing up a concept hierarchy for a dimension or for dimension reduction. It basically means in a particular dimension.

Drilldown:

Drill down is something that is just the reverse of rolled up. So, what we did in Roll up was we aggregated a set of attributes. So, let us break down the entire attribute into smaller attributes. So, we can do that by stepping down a concept hierarchy for a dimension and by also introducing a new dimension.

Slice:

Slice operation provides a new sub-cube from one dimension in each cube. Now what this means is, in any cube of ours will have three different dimensions. So, we have a Z-axis, we have Y-axis and all X axes. So, three different dimensions and what the slice operation means is with help three, in the three different dimensions we can use one of the dimensions and break it down into a two-dimensional cube.

Dice:

Dice operation provides and a new sub-cube from two or more dimensions in each cube. In the earlier example we saw slice, slice what it does is give us a new sub-cube from one by using one dimension in each cube. But Dice here gives us new sub-cubes from two or more dimensions in each cube.

Pivot:

Then finally we have one operation called pivot operation. So, the pivot operation is also known as the rotation operation. It basically transposes both the axis whether it is the X&Y axis transposes them to provide an alternative presentation of data.

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Social profiles