It is no wonder that Hadoop is recognized as a new star in the Big Data universe. With the aid of the open source framework, data volumes of various structures can be quickly adjusted at will, while at the same time cost-effectively administered, used and evaluated. In many cases, however, Hadoop alone is insufficient in meeting the demands of Big Data analytics.
The evaluation of semi- or unstructured data in combination with the most current business data, lends itself to in-memory processing with modern analytic techniques.
SAP HANA and Hadoop are a strong team: the combination of high-performance databases with a solid platform. Opening new channels for real-time analytics with massive cost savings and the possibility of a multitude of application scenarios.
Application scenarios for Hadoop and SAP HANA:
- Analyses of data from various data sources
- Prediction of customer behavior/improvement of customer retention
- Monitoring of machinery/devices
- Risk minimization
- Atmosphere and trend analysis
- Product and sales success predictions
- Increase of the efficiency of operating processes
- Monitoring/optimization of IT
Cutting costs through a successful combination of SAP HANA and Hadoop.
A large German chemical company is an example of how real time evaluation of large data volumes can be applied for proactive maintenance and the minimization of risk. One of the company’s manufacturing facilities generates 5 terabytes of data hourly. Provision of hardware and licenses alone for this volume of data in SAP HANA would cost upwards of 250.000 Euro per month.
In a Hadoop cluster, the mass data that is produced could be cost-effectively stored through data streaming. Data processing is then carried out using SAP HANA. The in-memory platform correlates various measurements, such as temperature, flow and pressure/pressure loss, and calculates what probability a part (for example, a pump) will malfunction. Before a threshold value is reached, a maintenance request is automatically produced in order to prevent an unplanned stoppage at the manufacturing facility.