Acquiring, Preparing, and Modeling Data

Create connections to source systems and acquire data from them:

• SAP Datasphere supports a wide range of connections. See Connection Types Delivered by SAP.

• You can import semantically rich data from SAP S/4HANA or SAP BW and choose whether to federate or replicate it. See Importing Objects with Semantics from SAP Systems.

• By default, you use remote tables to federate data, but you can improve performance by enabling replication. You can make a one-off snapshot replication, schedule regular replication or, for some connections, enable real-time replication. See Import Remote Tables.

• You can use replication flows to replicate data from multiple source objects into local tables with support for delta loading. See Creating a Replication Flow.

• You can use data flows to perform a number of ETL actions before loading data into a local table. See Creating a Data Flow

• You can schedule remote tables, replication flows, and data flows to occur regularly. See Scheduling Data Integration Tasks.

• Task chains allow you to run and schedule multiple data integration tasks together. See Creating a Task Chain.

Import data from CSV files to create or fill a local table with support for basic transformations, such as filtering, splitting, replacing, and changing case. See Creating a Local Table from a CSV File.

Use your preferred ETL tools to write data to an Open SQL schema and then access it as a local table in your space. See Integrating Data via Database Users/Open SQL Schemas.

Associate HDI containers to your space and access their data as local tables. See Exchanging Data with SAP HANA for SQL Data Warehousing HDI Containers.

If your SAP Datasphere is part of an SAP Business Data Cloud formation, you can load data via data products. See Working with SAP Datasphere in SAP Business Data Cloud.

All data is loaded into spaces and only users who have access to a space can see the data in that space. See Working in SAP Datasphere Spaces.

SAP Datasphere supports row-level security via data access controls, so that each user can access only those records she is entitled to see. See Securing Data with Data Access Controls.

You can share data from one space to other named spaces. See Sharing Entities and Task Chains to Other Spaces.

Only the data you choose to expose can be accessed outside SAP Datasphere. See Exposing Data For Consumption.

Graphical views can contain Join and Union operators to combine data in a graphical interface and output it as a view. See Create a Join in a Graphical View and Create a Union in a Graphical View.

If you are more comfortable with writing SQL can use a traditional join clause in an SQL view. See Creating an SQL View.

You can combine data and output it to a table by using an appropriate transform in a transformation flow. See Creating a Transformation Flow or Creating a Transformation Flow in a File Space.

For situations where join data is incomplete or unreliable, you can use an intelligent lookup to define rules and iteratively match records, reviewing and processing the results. See Creating an Intelligent Lookup.

Data flows can also perform joins and unions. See Create a Join in a Data Flow and Create a Union in a Data Flow.

You should federate or replicate all the data you could need so that SAP Datasphere can serve as a single source of truth for any form of analytics or other modeling, but you shouldn't load data that will never be of use:

• You can optimize performance of remote tables by removing unnecessary columns and setting filters. See Restrict Remote Table Data Loads.

• You can filter data and remove unneeded columns in replication flows. See Define Filter Expressions in a Replication Flow Object Projection and Define Mappings in a Replication Flow.

• Data flows also support filtering and removing columns. See Create a Projection in a Data Flow.

• Ensure that you understand and are using appropriate column data types. See Column Data Types.

In each processing step, try to minimize the data you are working with and consider persisting data where necessary to optimize performance:

• Graphical views provide the Filter operator in a graphical view. See Filter Data in a Graphical View.

• If you prefer to write SQL code, use an appropriate WHERE clause in your SQL views. See Creating an SQL View.

• Define parameters to allow users to choose only the data they need. See Create an Input Parameter in a Graphical View.

• You can filter data by using an appropriate transform in a transformation flow. See Creating a Transformation Flow or Creating a Transformation Flow in a File Space.

When exposing data for consumption, you should optimize performance by exposing only the data necessary to answer that business question:

• Set a global filter in your analytic model and ensure you exclude unneeded measures and attributes. See Create a Global Filter in an Analytic Model.

• Business Builder consumption models also support filtering and create lightweight perspectives. See Create a Consumption Model.

Graphical views provide the Aggregation node. See Aggregate Data in a Graphical View.

Include a GROUP BY clause in your SQL view. See Creating an SQL View.

You can aggregate data by using an appropriate transform in a transformation flow. See Creating a Transformation Flow or Creating a Transformation Flow in a File Space.

You can perform a wide range of aggregations and exception aggregations on the measures in your analytic model. See Create a Measure in an Analytic Model.

Business Builder consumption models also support aggregations of your measures. See Creating a Consumption Model.

You can provide translations of master data such as product names and time and geographical data. See Create a Text Entity for Attribute Translation.

Enable translation in your space and use the Translation Dashboard to ensure that the names of your columns are available in appropriate languages. See Translating Metadata for SAP Analytics Cloud.

Convert currencies and units to ensure that totals makes sense. See Converting Currency and Unit Values.

Select a Semantic Usage of Fact to indicate that your entity contains numerical measures that can be analyzed. See Create a Fact to Contain Measurable Data.

Select a Semantic Usage of Dimension to indicate that your entity contains attributes that can be used to analyze and categorize measures defined in other entities. See Create a Dimension to Categorize Data.

Select a Semantic Usage of Hierarchy to indicate that your entity contains parent-child relationships for members in a dimension. See Create an External Hierarchy for Drill-Down.

Select a Semantic Usage of Hierarchy with Directory to indicate that your entity contains one or more parent-child hierarchies and has an association to a directory dimension containing a list of the hierarchies. Create a Hierarchy with Directory.

Select a Semantic Usage of Text to indicate that your entity contains strings with language identifiers to translate text attributes in other entities. See Create a Text Entity for Attribute Translation

Create an analytic model to consume a fact and its associated dimensions and then filter and enrich its data as necessary for a particular analytic question. See Creating an Analytic Model.

The Business Builder provides an alternative method for consuming your facts and dimensions. See Modeling Data in the Business Builder.