Multidimensional vs Tabular SSAS models: we support both

MULTIDIMENSIONAL

When analyzing really large data sets, we use Microsoft SQL Server Analysis Services. OLAP cubes allow coping with much bigger data volumes than relational databases. However, lots of users claimed that multidimensional cubes were hard to understand, especially when designing the model.

That was one of the reasons why SSAS Tabular Model was introduced in 2012. It has become widely popular since then. Basically, tabular model is something in between relational databases and multidimensional cubes. Similar to databases, tabular model supports tables with relations. Similar to cubes, the model supports measures and key performance indicators (KPIs).

Which model to choose?

As far as choosing the model goes, it’s better to conduct a detailed study based on project requirements.

      Why
tabular solution
    can be chosen:
  • easier for understanding and creating the model
  • works quicker than multidimensional cubes for queries based on columns
  • hardware, such as disks, is not important. However, tabular is a memory dependent solution and more memory will ensure better performance
  • more efficient data compression about one-tenth of the size, whereas the compression of multidimensional is a third of the size of the original database
      Why
multidimensional solution
    can be chosen:
  • works better with a large amount of data – when we are talking about terabytes, it’s better to go with the multidimensional database. If your database requires more than 5 terabytes, multidimensional is the only option
  • performs better in terms of scalability
  • some features, such as aggregations or actions, are supported in the multidimensional model only

Which model is supported in Flexmonster?

Our development team believes that both tabular and multidimensional solutions are highly useful. That’s why Flexmonster pivot table component supports both models. You can view data from tabular OLAP cubes as well as from multidimensional ones. We hope you will enjoy your reporting experience with OLAP cubes!