Live Analytics Across All Your Enterprise Data Powered by SAP HANA

[songlin620]

This blog post introduces a solution to expand SAP Analytics Cloud live analytics across 11 new data sources with SAP HANA Smart Data Access and SAP HANA Smart Data Integration.

Proposed Solution

To expand the SAP Analytics Cloud real-time analytics across heterogeneous sources, we leverage SAP HANA 2.0 on-premise to connect to on-premise and cloud systems. SAP Analytics Cloud provides live data connection to SAP HANA, and SAP HANA accesses remote sources using SAP HANA Smart Data Access and SAP HANA Smart Data Integration.

The above figure illustrates the proposed solution that utilizes SAP HANA as a data gateway to remote sources. Among the heterogenous sources, SAP Analytics Cloud live connectivity can be expanded to 11 new data sources: Microsoft SQL Server, Oracle, IBM DB2, Google BigQuery, and several other databases with SAP HANA Smart Data Access; SAP ERP and MySQL with SAP HANA Smart Data Integration.

The Advantages of this solution

When using SAP HANA as a single gateway access to data in remote sources, it offers the following advantages:

• Support your business decisions with real-time analytics


• No replication of authorization models and authorization model always up to date


• Adding hierarchical analysis to originally flat data

How to implement this solution

I have created a How-To Guide to comprehensively show the configuration steps with 2 examples. One example is to connect to Microsoft SQL Server using HANA Smart Data Access; the other is to connect to SAP ERP using HANA Smart Data Integration. It would be very similar to connect to the rest data sources.

Thus, if you want SAP Analytics Cloud live connect to any of these 11 data sources, and you have HANA 2.0 on-premise, this solution will be one of the best choices for you and your business to build real-time reporting.

[subash]

Hello Kevin,

 

What about Data transfer between SQL -> HANA is that a physical data load required?

 

Regards,

Subash

[songlin620]

Hi Subash,

Thank you for you question. There is no data replication between SQL and HANA. The live connection is end-to-end from SAC to SQL. Only result set is sent to HANA.

Best,

Kevin

[DebjitSingha]

Thanks songlin620  for sharing this with everyone. This defiantly add new dimension for customers looking for live connections to non SAP DS. 

[mattiagiovannini]

Thanks, Kevin, this is great!

A clarification about the security: SSO is only available for SAP Hana (not an option with a -eg- remote MS SQL/Oracle), correct?
And the case of  secondary credentials, how would you replicate / inherit this into SAC (origin>HANA>SAC)?

Thanks

[songlin620]

Hi Mattia,

Thank you for the questions. It is recommended to enable Kerberos SSO between HANA and remote sources, which allows the row-level security. Secondary credentials will also work, HANA will remember the associated secondary credentials for each user, and uses that to authenticate the user in remote sources. We don't replicate the secondary credentials into SAC, the live connection request is sent from user's browser to HANA, SAC host doesn't have a direct connection to HANA.

Best,

Kevin

[mattiagiovannini]

Hi Kevin,

 

Thanks for your answer. The document you pointed mentions that "All connections to remote database (SAP HANA remote databases only) are authenticated through Kerberos single sign-on (SSO)" , this is why I understood SSO was an option only for HANA databases.

Regarding the secondary credentials, I understand then that the solution would be to define them between the remote origin and HANA and then having a SSO between HANA and SAC to inherit the definitions given at HANA level, correct?

Thanks again

 

BR

[songlin620]

The verb 'inherit' is a bit ambiguous. SAC has a live connection to HANA using SAML SSO. On the other side, HANA accesses to remote sources using SDA authentications (Kerberos SSO, Secondary credentials, or user/password).These 2 authentication processes have no relationship, they are separate. If a SAC user logs into HANA and wants to use any remote sources, it will refer to the respective SDA authentication definition in HANA. I am not sure if this is the meaning of 'inherit'. Let me know if I am wrong.

[bjarkebb]

Hi Kevin,

Thanks for a very relevant webinar and this blog!

I have two questions I hope you can help with

1. My understanding is that we HANA Web IDE is deprecated and replaced by XS Advanced. However, in the how-to-guide it is mentioned that the virtual procedure can only be created in Web IDE and as far as I can see it is not possible from XSA to create virtual object on a remote procedure like it is for a table.
   So can you guide me to a guide for creating and utilizing virtual procedures in XSA for instance on an SQL database using the MSSqlLogReader adapter on the DPagent?


2. Is there a way HANA SDI/SDA can consume Odata endpoints created in SAP Open Connectors or SAP API Management? We have tried with the Odata adapter but it seems to fail, possibly because it is lacking the catalog as described in SNote 2539268

Thank you!

[songlin620]

Hi Bjarke,

Thank you for your questions:

1. You probably have to write some SQL to create the virtual procedure in SAP WEB IDE 2.0. https://help.sap.com/viewer/3823b0f33420468ba5f1cf7f59bd6bd9/2.0.05/en-US/6c8e6c1b62d54197bdaa786910...


2. I am not an expert in SDI, and haven't used OData Adapter either. You have to reach out to Product Support team for some investigation on that error.

Hope that helps.

Thanks,

Kevin

[bjarkebb]

Thank you Kevin for the suggestions!

[Sumit_Gupta]

Hello songlin620 ,

Thanks for this nice article.

I had a question if there is a connector available for Azure synapse Analytics. If not then if it is in roadmap.

Alternatively, could the MS SQL server connection can be used to connect to Synapse.

Please suggest.

Regards,

Sumit

[songlin620]

Hi Sumit,

There is not a dedicated SDI connector for Azure Synapse Analytics. You can use generic JDBC connector in SAP HANA SDI. SQL server connector doesn't support Azure Synapse Analytics.

Best,

Kevin