Online Analytical Processing (OLAP), and cubes are other words for multi-dimensional sets of data that essentially serve as a staging space in which to analyze information. These special online analytic processing databases hold data not in tables but in OLAP cubes which are a mechanism used to store and query data in an organized, multi-dimensional, structure specifically optimized for analysis.
OLAP databases are designed to pre-calculate as many queries and combinations of data fields as possible in order to provide fast query response. However, while these solutions perform better than classical relational databases, their multi-dimensional structure makes them inflexible and unable to accommodate changes easily. In addition, storing large amounts of data in a cube causes a performance bottleneck. While OLAP databases are quite useful for basic use cases, large data sets require using capabilities from additional tools in tandem, which complicates analytical efforts and requires unique skills.
ROLAP an Enhanced Approach?
Snowflake is the only data warehouse built for the cloud. Snowflake’s high-performing cloud analytics database combines the power of data warehousing, the flexibility of big data platforms, the elasticity of the cloud and true data sharing at a fraction of the cost of traditional solutions.
Snowflake has the vision to make modern data warehousing effective, affordable and accessible to all data users.
One special thing about Snowflake is it’s ability to efficiently separate the storage capacity from computing capacity. This eliminates the limitation common to the cloud databases which does not allow growing the storage capacity without paying more for the computing capacity -- which you may not need at all.
Snowflake’s bing native cloud database has an important ability which is vital for data warehouses - namely elastic scalability. This means that Snowflake is able to dynamically scale up resources as needed in order to load the data into the database. This way, loading of data into the data warehouse does not negatively affect the execution of user queries.
What comes first? Move and model your data!
Another way to organize data for multi-dimensional querying is relational online analytic processing (ROLAP). ROLAP is a form of OLAP that performs multi-dimensional analysis of data stored in a relational database rather than in a multi-dimensional database, which is considered the OLAP standard.
Although ROLAP technology performs better than OLAP databases when processing large amounts of data, it cannot beat the speed and efficiency of OLAP on smaller amounts of data. ROLAP databases require a great deal of manual maintenance and are difficult for business users to operate so ROLAP is considered to be more inflexible than OLAP cubes. OLAP and ROLAP are both still popular today but neither technology can keep up with today’s demands for near real time data for analytics nor handle unstructured data.
Get to know Pipes - Your data movement solution
Learn more about the new technology that is replacing OLAP and ROLAP. Get your free eBook now.