Workshop: Smart-Data Collection and Processing for Real-Time Sensor Networks

April 28, 9:00am - 12:00pm
Mānoa Campus, 103 Keller Hall or Online via Zoom Add to Calendar

The recent rise of inexpensive devices in the Internet of Things (IoT) as well as the number of instruments and sensors to observe and measure everything has led to a deluge of data and demand for support related to storing, processing, and analyzing time-series data. Many science use cases based on monitoring require ongoing data processing or special processing/modeling and notification of anomalous or special events that can be identified by processing and computing the data as it arrives. Systems that have been designed for sensors are often aimed at industry, and either are complex to deploy and maintain or, if hosted, are expensive. This leaves a gap for hosted academic streaming data solutions capable of supporting data event-driven computational workflows. To help address this cyberinfrastructure need to support streaming data for science, the Texas Advanced Computing Center (TACC) and the University of Hawaii (UH) have developed an open-source unified middleware API infrastructure platform, Tapis, with collaborative features to fill gaps in the existing streaming time-series data landscape.

This tutorial will focus on providing attendees with an introduction to the Tapis framework, an NSF-funded project, which is an open-source, scalable API platform that enables researchers to perform distributed computational experiments securely and achieve faster scientific results with increased reproducibility. Attendees will learn about the Tapis Streams API, which focuses on supporting scientific use cases that require working with real-time sensor data, and Abaco. Abaco is Tapis’ functions-as-a-service (FaaS), based on the actor model of concurrent computation and Docker. Attendees will be able to run Jupyter Notebooks leveraging the tapipy Python library which will aid in interacting with the Tapis API services within an interactive Jupyter notebook. The concepts of containers and Actors will also be introduced at a high level. Using resources in the Streams API, attendees will learn to build a hierarchy of organizational objects to simulate a real-world example of the sensor network and create webhook notifications when sensor data exceeds predefined thresholds. Attendees will also learn how to set up and use a serverless interactive science gateway that runs entirely on GitHub pages, by utilizing the Tapis UI project.

Presenters: Dr. Joe Stubbs - Research Associate, Texas Advanced Computing Center (TACC)

Dr. Anagha Jamthe - Research Engineering Scientist, TACC

Dr. Rich Cardone - Research Associate, TACC

Dr. Steve Black - Engineering Scientist, TACC

Dr. Sean Cleveland - Cyberinfrastructure Research Scientist - Information Technology Services, University of Hawaiʻi System

Register at: http://go.hawaii.edu/x4X


Ticket Information

Event Sponsor
Hawaiʻi Data Science Institute, University of Hawaii

More Information
8089563503,
datasci@hawaii.edu

Share by email