Creation of a scaleable and efficient Analytics Gateway to collect anonymous usage data was achieved leveraging a "Serverless" Architecture. The "Serverless" Framework allows the construction auto-scaling, event driven components deployable to many cloud platforms such as Amazon Web Services, Azure and Google Cloud Platform or as the business case requires, deploying as easily to your own private cloud (Intel Private Cloud). Currently servicing sensor and usage data to over 20,000 live connected IoT devices deployed across the U.S., we are able to collect real-time, anonymous statistics for only pennies per device over the service life of the device.
Extremely Low latency (<100ms) and minimal downtime from connectivity disruption is achieved from this service model.
Exploration of usage of Analytics data with leading Machine Learning platforms: Caffe2 (on AWS), Apache Spark BigDL, Cloud AI (Google) & Watson (IBM).
Topics explored: * Serverless - https://serverless.com/ * Caffe2 - https://caffe2.ai/ * Google Cloud AI - https://cloud.google.com/products/machine-learning/ * Apache Spark BigDL - https://software.intel.com/en-us/articles/bigdl-distributed-deep-learning-on-apache-spark * IBM Watson Machine Learning - https://datascience.ibm.com/features#machinelearning * Intel Private Cloud - https://www.intel.com/content/www/us/en/cloud-computing/private-cloud-solutions.html