How connect HANA to BI ?

[maxim_molochko2]

Hello . We have HANA 1.0 sp 2 r16 and BI 7.01 .

How get DATA from table HANA to BI.

[Vitaliy-R · Accepted solution]

Maxim.

What BI tool you want to use? Do you want to see data from HANA's SQL tables or from analytics/calculation view?

If you are using SBO Analysis - you need to configure ODBC connection to your HDB. If you are using Universe or Explorer based analytics - you need to deploy HANA's JDBC drivers and configure the connection using SBO IDT.

Udachi,

-Vitaliy

[Former Member]

Connecting Hana to IDT(BI)

1.Install HANA Client on BI 4.0 Server

2.Locate the JDBC driver

The default HANA client install folder locations (in Windows):

64-bit Client install: /program files/sap/hdbclient

32-bit Client install: /program files (x86)/sap/hdbclient

Locate the JDBC driver file: ngdbc.jar

3.Setting at BI4 connection server

Copy ngdbc.jar from previous step to:

BI4_INSTALL_PATH\SAP BusinessObjects Enterprise XI 4.0\dataAccess\connectionServer\jdbc\drivers\newdb

(create newdb folder if required)

Restart the BI4 Adaptive Processing Server in CMC

4.Setting at BI4 Explorer server

Stop Tomcat server where BI4 Explorer is deployed.

Copy the same ngdbc.jar and replace the jar in: BI4_INSTALL_PATH\Explorer14.0\plugins\com.sap.ngdbc_1.0.0\lib

Restart Tomcat server and create connections in IDT

[Former Member]

Use Data services to extract data from BW and then load to HANA db, that is the appropriate way.

you can create a job & schedule both for initial load and delta loads on data services.

[former_member184944]

Hi Max,

I read through your above post, and i come to you that you have worked on scenarios connecting PI to HANA.

I really need help here, In my current project we are trying to connect to HANA from SAP PI .

Since this is new , i would like to know how is this possible . Your inputs/Help would be great .

In case you have some document about the same , please do share .

Thanks in advance.

[maxim_molochko2]

Hello.

My scenario SAP-PI-HANA synchronously.

1. i install jdbc driver for hana in PI. I get it from Hana studio.

2. i made mapping from Source system to hana - I use SQL_QUERY

3. And i made data type for response from hana.

All work good. But if you have big array data from hana and many parallel users - be warning - it many PI memory use.

Ask if need.

Regards

Maxim

[former_member259744]

Hi,

I'm on the same page as Anumeha trying to connect PI to HANA.

When monitoring the flow I get a "SAPClassNotFoundException: com.sap.db.jdbc.Driver".

I tried to install the ngdbc.jar file as a JDBC-Resource (just named it 'NGDBC') within the NetWeaver administrator, but the driver does not appear at the position /usr/sap/POI/J00/j2ee/cluster/server0/./temp/dbpool/_NGDBC/ngdbc.jar (where it is pointed by NetWeaver).

How/where did you install the driver for PI?

Thanks,

Martin

[maxim_molochko2]

Hi, Martin

JDBC driver need install from jspm.

See links.

http://www.sdn.sap.com/irj/scn/go/portal/prtroot/docs/library/uuid/f04ce027-934d-2a10-5a8f-fa0b1ed4d...

http://www.sdn.sap.com/irj/scn/go/portal/prtroot/docs/library/uuid/8000cffc-6b92-2d10-3493-f2ac13992...

Regards

Maxim

[former_member259744]

Hi Maxim,

thanks for the quick answer - will take a closer look.

Regards, Martin

---

Edit:

I'm on version NW 7.31 - the documents seem to be not accurate in this case. (I opened another thread for)

[maxim_molochko2]

Thanks for reply Vitaliy.

But your advice is not suitable.

I need load data to BW infocube. There is an idea shift heavy extractor from R/3 to HANA.

I see three variants

1. through SBO->Universe->Webservice->BW

2. XI JDBC Adapter -> RFC Adapter ->BW

3. DB Connect in BW

Numbers of records are not so much - about some hundreds, from millons.

[Vitaliy-R]

If I see "BI" I think about BusinessObjects BI, not about BW, so I was confused by your question.

To your question - non-technical advise first: check if your license with SAP covers "SAP HANA Data Distribution Rights" to get data out of HANA to downstream systems.

On the technical side: ideally, you should consider BW-on-HANA which is going into Ramp-up on Nov 7th (not sure why SAP picked the anniversary of October Revolution for that ;-). That way you can replicate your ERP data into HANA db into separate schema, and then take data from there into BW schema for loading into PSAs and further up BW processing chain.

Other "clean" solutions should be (but you need to think how to manage deltas as well):

1/ Using BO DataServices (but then you need to upgrade your DS license from HANA's one to include BW as a target) - ODBC connector on HANA side, BW's Source System on the other side.

2/ Using UDConnect (DBConnect does not support HANA db as a asource, at least to my knowledge) thought JDBC.

Thing you need to take into consideration is the performance of what you are trying to do. Columnar databases, like HANA db, are suited for analytics solutions, and not for data distribution. You need to test what is the performance of converting your volumes (100s should not be that bad) of complete lines from columnar storage back the row-based format.

Udachi,

-Vitaliy

Edited by: Vitaliy Rudnytskiy on Sep 23, 2011 4:48 AM

[maxim_molochko2]

Spasibo Vitaliy.

We connect to HANA from SAP PI by JBDC adapter. It work and enough.

Regards

Maxim.

[Vitaliy-R]

Interesting. Thank you for sharing. -Vitaliy