Creating an application that uses a partitioned dataset

To create an application based on a partitioned dataset, you need to perform the following high-level tasks:

• Evaluate the needs of your application and whether your application is suitable for partitioned datasets.
• Prepare your dataset and determine the number of partitions that you require.
• Ensure that you have the necessary hardware to support the partitions of your dataset.

• Design a logical data model of your dataset, similar to designing a project schema for MicroStrategy. The logical model is a representation of your dataset as experienced by your users. A logical data model depicts the flow and structure of your data, providing a way of organizing data so that it can be analyzed from different perspectives.

  For steps and considerations for creating a logical data model, see the Project Design Help.

• Select the attribute that will divide your dataset (the partition attribute). All the tables that contain the partition attribute are split into partitions. The tables are analyzed in parallel by the processor cores of your Intelligence Server machine.
• Create the dataset using Web.
• Use simple base metrics when you create the dataset, and then create derived metrics for more complex metrics calculations after you import the dataset. For steps to create derived metrics, see the MicroStrategy Web Help.
• Design your documents and dashboards using MicroStrategy Web. For steps to design documents and dashboards, see the MicroStrategy Web Help.
• Maintain and update your dataset.