Difference between molap, rolap and holap in ssas

Difference between molap, rolap and holap in ssas

• 1.
  Difference between MOLAP,ROLAP and HOLAP in SSAS MOLAP ROLAP HOLAP MOLAP stands for ROLAP stands for Relational HOLAP stands for Hybrid Multidimensional Online Online Analytical Processing Online Analytical Processing Analytical Processing The MOLAP storage mode The ROLAP storage mode The HOLAP storage mode causes the aggregations of the causes the aggregations of combines attributes of both partition and a copy of its the partition to be stored in MOLAP and ROLAP. Like source data to be stored in a indexed views in the MOLAP, HOLAP causes the multidimensional structure in relational database that was aggregations of the partition Analysis Services when the specified in the partition’s to be stored in a partition is processed. data source. multidimensional structure in an SQL Server Analysis Services instance. This MOLAP structure is Unlike the MOLAP storage HOLAP does not cause a highly optimized to maximize mode, ROLAP does not copy of the source data to be query performance. The cause a copy of the source stored. For queries that storage location can be on the data to be stored in the access only summary data in computer where the partition Analysis Services data the aggregations of a is defined or on another folders. Instead, when results partition, HOLAP is the computer running Analysis cannot be derived from the equivalent of MOLAP. Services. Because a copy of query cache, the indexed the source data resides in the views in the data source are multidimensional structure, accessed to answer queries. queries can be resolved without accessing the partition’s source data. Query response times can be Query response is generally Queries that access source decreased substantially by slower with ROLAP storage data—for example, if you using aggregations. The data than with the MOLAP or want to drill down to an in the partition’s MOLAP HOLAP storage modes. atomic cube cell for which structure is only as current as Processing time is also there is no aggregation data the most recent processing of typically slower with —must retrieve data from the partition. ROLAP. However, ROLAP the relational database and enables users to view data in will not be as fast as they real time and can save would be if the source data storage space when you are were stored in the MOLAP working with large datasets structure. With HOLAP that are infrequently queried, storage mode, users will such as purely historical typically experience data. substantial differences in query times depending upon whether the query can be resolved from cache or aggregations versus from the source data itself.
• 2.
  Pros Pros Pros • Provides maximum • Ability to view the • HOLAP balances the query performance, data in near real-time. disk space because all the • Since ROLAP does requirement, as it required data (a copy not make another only stores the of the detail data and copy of data as in aggregate data on the calculated aggregate case of MOLAP, it OLAP server and the data) are stored in the has less storage detail data remains in OLAP server itself requirements. This is the relational and there is no need to very advantageous database. So no refer to the underlying for large datasets duplicate copy of the relational database. which are queried detail data is • All the calculations infrequently such as maintained. are pre-generated historical data. • Since HOLAP does when the cube is • In ROLAP mode, the not store detail data processed and stored detail data is stored on the OLAP server, locally on the OLAP on the underlying the cube and server hence even the relational database, so partitions would be complex calculations, there is no limitation smaller in size than as a part the query on data size that MOLAP cubes and result, will be ROLAP can support partitions. performed quickly. or limited by the data • Performance is better • MOLAP uses size of relational than ROLAP as in compression to store database. In nutshell, HOLAP the summary the data on the OLAP it can even handle data are stored on the server and so has less huge volumes of data. OLAP server and storage requirements queries can be than relational satisfied from this databases for same summary data. amount of data. • HOLAP would be • MOLAP does not optimal in the need to have a scenario where query permanent connection response is required to the underlying and query results are relational database based on (only at the time of aggregations on large processing) as it stores volumes of data. the detail and aggregate data in the OLAP server so the data can be viewed even when there is connection to the relational database. Cons Cons Cons • With MOLAP mode, • Compared to • Query performance you need frequent MOLAP or HOLAP (response time) processing to pull the query response is degrades if it has to refreshed data after generally slower drill through the last processing because everything is detail data from
• 3.
  resulting in drainon stored on relational relational data store, system resources. database and not in this case HOLAP • Latency; just after the locally on the OLAP performs very much processing if there is server. like ROLAP. any changes in the • A permanent relational database it connection to the will not be reflected underlying database on the OLAP server must be maintained to unless re-processing is view the cube data. performed. • MOLAP stores a copy of the relational data at OLAP server and so requires additional investment for storage. • If the data volume is high, the cube processing can take longer, though you can use incremental processing to overcome this. And, further updates on difference between questions and answers, please visit my blog @ http://onlydifferencefaqs.blogspot.in/