In an increasingly connected world, relationships have become more and more significant. However, current analytic tools struggle to offer insight into the web of relationships found in data.
How can HANA Graph help?
HANA Graph; a new capability within the SAP HANA Platform, provides an innovative way of interrogating relational data. HANA Graph can help businesses solve complex problems and make recommendations for customers, answering questions such as:
- Which products are frequently bought together?
- Which restaurants do my friends go to?
- Which products tend to be purchased by the same customer and other members of their household on a regular basis?
- What other factors could dictate how people make purchase decisions?
How does it work?
HANA Graph works on data stored in the native SAP HANA Platform, providing a means of modeling data within the graph workspace and processing it using the graph engine. The ability of the HANA Graph engine to be consumed via HANA’s native calculation views enable it to be exposed to the enterprise via the wider capabilities of the SAP HANA Platform. The added benefit is that HANA Graph provides organisations with a tool for performing complex graph operations on data without the need for a ‘sidecar’ graph database and the overhead that comes with it.
What are the benefits?
HANA Graph allows for complex pattern recognition within data, either via the use of proprietary code or a native UI. The example below illustrates how graph can be used to identify similar products bought in the same transaction:
Graph’s utilisation of in-memory processing and connecting entities with their relationships in a central store enables queries to be executed far quicker than other database technologies allow. In addition, the graph structure upon which the data model is built highly flexible, overcoming the limitations of traditional database structures.
HANA Graph: An Example
The example below looks at the relationships between sales orders and products with the aim of answering the question ‘Which products are frequently bought together?’ Here, a customer wishes to purchase a Hammer and HANA Graph shows which products have been purchased with it in the past, linked by their sales order:
From this, a recommendation trigger can be set up to suggest a set of products which the customer might also be interested in purchasing. It is clear from the visualisation that the strongest relationship lies between the purchase of a Hammer and Nails. However, for much larger datasets these conclusions are not easily deduced visually and the output will need further investigation.
Are you making the most of your investment in SAP HANA? Do you wish to learn more about the relationships within your data? AgilityWorks can help you harness the power of SAP HANA and help you maximize the return on your investment. For more information, please do not hesitate to get in contact.