The MicroStrategy BI Architecture comprises of the following components,
– Source Systems
– ETL Process
– Metadata Database
– MicroStrategy applications
The first 3 components has been explained in yesterdays post itself. Lets talk about the remaining components today.
To say short, they are typically databases or mainframes systems that store transaction processing data (OLTP)
In short source systems are designed and optimized for transaction processing where as DWH systems are designed & optimized for analytical processing (OLAP). Main thing to note in DWH applications is Frequent read happens ( Select SQL most & few inserts/updates/deletes ). This everyone should make a note of it & should not coin a project or solution to be as DWH system without knowing what DWH is. There are people who used to perform Dataload once in 30 minutes time interval, but then they coin the name of the product/project to be as DWH. Ideally it is not. DWH application is for analysis purpose and not for transactional processing or viewing the transactional data in near real time.
Transfer of data from source systems to DWH using ETL process ( Extract /Transform / Load )
The metadata database contains the information that facilitates the retrieval of data from the DWH when using MSTR applications. It stores the MicroStrategy Object definitions and information about the DWH in the proprietary format and maps the MicroStrategy objects to the DWH structures and content. In short to say –> MicroStrategy Metadata means — the logical informations of the Database is stored inside MicroStrategy as MD, which inturn helps the MicroStrategy engine to communicate to the database to fetch data using the MSTR Metadata informations ( Mainly To generate Optimized SQLs )
MSTR Applns use the Metadata (MD) database to translate user requests into SQL queries and to translate the result of those SQL queries back to MicroStrategy objects like Reports/Dashboards 🙂 Which the end users will be interested to see.
So the metadata stores all the objects definitions ( Facts / Attributes / Metrics / Filters / Prompts / etc ). It acts as a central repository for all the object definitions.
The MicroStrategy applications will alow us to interact with the Business intelligence system. Providing the ability to look the data from different perspective ( forecasting / trend growth / etc ). Grid or Graph representations helps the end users to be available for superior presentations. Documents can be built – which combines multiple datasets together and showing them to be graphical & flashy. Hence the users look and feel is addressed using flashy widgets usage & interactive Grid/Graphs in it.
Also MicroStrategy offers web UI customisations as well but it involves SDK customisations ( which is quite easy with the help of MicroStrategy Developer Library – it has several examples that the end user wants like to have a custom logo & color changes etc. )
Note: But you should keep a point that doing too much of UI customisation will cause hindrance during the upgrade of versions when MicroStrategy releases. So keep it simple and easy always 🙂 !!!
Hope the above content provide you an overview about MicroStrategy BI Architecture!!
Going forward, will start on MicroStrategy Desktop stuffs 🙂 — The Key Developer Interface in MicroStrategy.
Have a Good Day!!