In this tutorial video from the SAP HANA Academy Fernando Velez provides a quick demonstration of some of the new features of Smart Data Integration and Smart Data Quality in SAP HANA SPS09. Check out Fernando’s video below:
(0:10 – 1:16) Overview of Information Management in SAP HANA and Outline of the Demonstration
This new Enterprise Information Management offering in SAP HANA provides a simplified landscape where users can author projects from a single modeling environment (either SAP HANA Studio or the Cloud version, SAP HANA Web IDE) while bringing in data from a variety of sources, either on-premise or cloud. It is equipped with several out of the box built-in adaptors that are extensible throughout documented SDK and because it works within SAP HANA, workloads are accelerated, resulting in lower latency for the availability of the consumable data.
Fernando will demonstrate Information management in SAP HANA through a working demonstration on a business case involving auditing purchasing transactions. Fernando desires to identify how much business he is doing with vendors that have been favorably rated.
For the purpose of the demo Fernando is assume that his external information has already been brought into SAP HANA and is already present in a table. Fernando will be replicating two tables from a SAP production system that is holding purchasing information. The production system is a remote source that can be accessed in the SAP HANA Studio cockpit.
(1:16 – 3:50) Using SAP Web IDE to Create Virtual Tables
Fernando will be using the DB2ECC adaptor and will have one agent running on his remote sources where the SAP production is stored. All of the tables will be brought to the RF schema.
First Fernando needs to create a replication task to insert a table into the RF schema. Currently replication tasks must be defined using the SAP Web IDE so Fernando logs on to his SAP Web IDE. Fernando opens the package of his replication flow and then right clicks and selects New and File. After entering a file name he clicks create.
Next Fernando chooses the remote source, identifies the target schema, chooses to create a virtual table schema by entering the same RF value and then also gives his virtual tables a prefix. Then he clicks on Add Objects and browses through the SAP remote source and selects the appropriate table.
So now all of the column names from table LFA1 (holds the all of the providers from SAP) has been added. Next Fernando follows the same steps to add the purchasing table. After Fernando saves the transformation he logs out.
Back in SAP HANA Studio Fernando’s RF schema is now populated with virtual real tables. The real tables are empty but the virtual tables have data.
(3:50 – 4:50) Using a Stored Procedure to Execute Replication
To effectively execute replication Fernando clicks on the START_REPLICATION_DEMO folder under his Procedures folder. The procedure was generated when the replication flow was saved Web IDE. It defines remote subscription objects at the remote source that will read DB2’s re-do logs to bring over only transactional consistent data. The procedure then preforms the initial loading on both tables and will listen and distribute changes over to SAP HANA afterwards.
Fernando highlights the SQL script to call the stored procedure and hits the execute button. Now in the command prompt window for the remote site you can see there is some activity as the agent is receiving the order from SAP HANA and is starting to load the changes.
After 44 seconds (length of procedure) Fernando checks to see if the data has been loaded by running select count(*) from SQL statements on both of his real tables. Now he has confirmed that his purchasing and providing tables have many rows of data and thus he has successfully replicated them in SAP HANA.
(4:50 – 7:20) Using Flows and Cleanse Transformation to Identify Transactions from First-Rate Vendors
Now Fernando can write interesting transformations to solve his business problem. So Fernando starts by restricting his providers to Germany, restricting his transactions to client 800 and preforms a join on two tables by using provider number and client number. Fernando wants to see the number of transactions in Germany cities so forms an aggregate to run after the two tables have been joined and then stores the results in a new table.
Once the flow has been executed he runs a select * and a select sum(countNbr) SQL query statement on his VENDOR_TR_CITY table to see the breakdown by city of his over 4,000 transactions.
Now Fernando uses another flow to determine how much business goes through the top-rated vendors. This flow looks very similar to one he used earlier but now he is joining his independent set of ranked vendors with his LAF1 SAP table using the names of the firms. Once the second flow executes he has a new table.
Running the same pair of SQL queries on his new table Fernando discovers that his transaction from first rated German Vendors is only 824 out of the over 4,000 transaction he had before.
To Fernando it seems very low that less that 25% of his business is going through first-rate vendors. However, Fernando compares the names of the firms with the Name1 column in the LAF1 table and discovers that many of the rows contain additional and unrecognized names and abbreviations for the same firms.
So Fernando uses a cleanse transform come from the Data Quality software that isgreat at cleansing party data such as firm names. Now Fernando applies the cleanse transform to the LAF1 table to cleanse the firm names and then join the cleansed result with the first-rated vendors using a strict column equality.
After executing this new flow Fernando examines the newly created third table to see that now nearly half of his transactions are from first-rated vendors and they originate from many more cities. Thus he has used Smart Data quality to solve his business problem.
SAP HANA Academy over 800 free tutorial videos on using SAP HANA.