At the recent SAP TechED && d-code event, we were excited to see what SAP is doing in terms of their major initiatives and how SAP (and MapR) will be able to help organizations around the world achieve simplicity while embracing the new trends shaping our industry: cloud, mobility, big data, and the Internet of Things. Apache Hadoop is a key part of SAP’s overall big data strategy, and we believe we’re very much aligned, both in terms of technology and strategy, with SAP’s key initiatives.
In the demo you can see how to bring packet capture data and JSON today together to track down a failing sensor in an IP network. Typically, bringing packet capture data (exported by Wireshark) and sensor data (in this case, JSON) together might not be an easy task—it can require transformations that take a lot of time. By using Drill to run SQL queries directly on packet captures and sensor logs, you can quickly turn that pile of data into something actionable:
If you know a little bit of SQL, it’s possible to get to the data you want while filtering out the noise. In this example, we filter IP conversations on a sensor network down to the set that interests us, and look at which sensors are seeing a lot of TCP retransmissions. The beauty of the solution is that by using Drill, you can access JSON and other data without putting it into a database or needing to apply schema first. The ability to run JSON queries within SAP Lumira is a real advantage that can save organizations a lot of time.
In addition, we’ve recently teamed with SAP to integrate the real-time deep analytics of SAP Hana with the MapR Distribution. UDFs (user-defined functions) have been added to SAP HANA in order to invoke MapReduce jobs directly in the MapR Distribution. This means that you’ll have the ultimate flexibility in creating next-generation, big data, and IoT applications.
At MapR, we’re looking forward to strengthening our partnership with SAP to benefit you. You can also read about this on the SAP IoT blog.