Differences between SAP Smart Data Access,Smart Data Integration and Smart Data streaming ??

[former_member609878]

What exactly is the difference between -

-sap hana smart data access

-sap hana smart data integration

-sap hana smart data streaming

How they are used and scenarios in which they are used ??

[dvankempen · Accepted solution]

https://blogs.sap.com/2016/07/05/the-various-data-provisioning-options-for-hana/

[RobertWaywell]

Streaming analytics has a significantly different role from SDA and SDI. As others have mentioned, SDA supports the creation of virtual or remote table connections allowing tables in separate databases to be accessed and queried from a single connection and SDI is an EIM tool focused on loading data into HANA.

HANA streaming analytics was renamed from HANA smart data streaming back with the HANA 2.0 SPS 02 release because the "smart data" prefix had become associated with EIM tools and was confusing folks as to the real value of HANA streaming analytics. Streaming analytics is focused on processing event driven data, in the context of historical and reference data, as fast as it arrives. Typical processing time through a streaming analytics project is a few milliseconds. This technology originated in the financial services space to handle real time trading and risk management where there is a requirement to process millions of events per second. Today IoT is a significant growth area with streaming analytics being used everywhere from managing continuous manufacturing processes to monitoring utility networks for issues, using machine learning to diagnose the likely cause, deploying work crews to resolve the issue and notifying customers all from the streaming analytics project.

A major advantage of streaming analytics is the ability to process huge volumes of low value data *before* it ever gets to the database. If you consider an IoT scenario, in a healthy system most of the data generated will show that the system is perfectly fine. With streaming analytics you can constantly monitor the live data, discard or consolidate the "everything is OK" data, and react instantly when an issue arises. In one media usage monitoring scenario the processing load was reduced by 95% by moving from a database centered "store and query" design to a real time streaming analytics implementation using HANA streaming analytics.

[former_member600626]

Hello Robert,

Quick question, I have a scenarios where Any change in the database assuming ECC (on Hana or non HANA), would want that change to be sent to the target HANA system. Like for example I have an Employee data and he has updated his address and phone number and I want a change event and send the data to my target systems. I don't want to IDOC's (Change Pointers) or SLT . But I am trying to see if there are any ways that I can use HANA Smart Streaming Analytics for this scenario.

Please advice, Your inputs are much appreciated.

Regards,

KishoreJoga.

[RobertWaywell]

This was answered in another thread as well, but for completeness on this thread here is the response:

________________________________________________________________________________________________________

SLT is the recommended approach for HANA to HANA replication.

HANA streaming analytics provides a generic DB input adapter that can poll a database table. In order to detect changes, the table needs to have a "last updated" column that is maintained any time the record changes. The query from the DB input adapter can then select records where the last update time is newer than the last poll. You will also have to configure the project for guaranteed delivery.

In short, yes you can implement a basic replication process using HANA streaming analytics but you are better off using a tool specifically designed for data replication.

________________________________________________________________________________________________________

[daniele_tiles2]

Hi Aakanshka,

here are some hints on the differences between the 3 components:

- Smart Data Access: it's a system to reference in HANA external data without replicating the data inside HANA (like, pointing to and Oracle table, and so on), you create virtual tables:

https://help.sap.com/viewer/eb3777d5495d46c5b2fa773206bbfb46/2.0.01/en-US/710235120ce744c381176121d4...

- Smart Data Integration: it's part of the SAP EIM (Enterprise Information Management) and it's used to load external data into an HANA DB

https://blogs.sap.com/2014/11/27/hana-sp9-data-provisioning-overview/

- Smart Data Streaming: it's a feature of HANA for stream data capture inside HANA and stream analytics:

https://wiki.scn.sap.com/wiki/display/SAPHANA/SAP+HANA+Smart+Data+Streaming

HTH

Daniele Tiles