IoT edge processing with Apache NiFi, Apache MiniFi, and multiple deep learning libraries
Timothy Spann leads a hands-on deep dive into using Apache MiniFi with Apache MXNet and other deep learning libraries on edge devices.
Talk Title | IoT edge processing with Apache NiFi, Apache MiniFi, and multiple deep learning libraries |
Speakers | TIMOTHY SPANN (Cloudera) |
Conference | Strata Data Conference |
Conf Tag | Make Data Work |
Location | New York, New York |
Date | September 11-13, 2018 |
URL | Talk Page |
Slides | Talk Slides |
Video | |
Timothy Spann leads a hands-on deep dive into using Apache MiniFi with Apache MXNet and other deep learning libraries on edge devices, such as Raspberry Pis with Movidius and the NVIDIA Jetson TX1. You’ll learn how to run deep learning models on edge devices and send images, GPS data, sensor data, and deep learning results if values exceed norms. Using S2S, data is sent to NiFi for further processing, additional TensorFlow processing, and data augmentation with weather and geolocation. A stream of data is landed as ORC files in HDFS with Hive tables on top. Processed data in the Avro format with a schema stored in Schema Registry is sent to Streaming Analytics Manager via Kafka, and additional processing and rules are processed to Druid endpoints. Visualization is shown in Zeppelin and Superset. Potential use cases for this solution include security camera monitoring, utility asset anomaly detection, and temperature and humdity filtering for devices.