User Tools

Site Tools


This is an old revision of the document!

7b.041.SBU - Medical Insurance Claim Prediction

Project - Summary

The overall goal of the project is to develop partial Replication Strategies for Managing Data Stream Processing & Analytics on Edge Devices. Specifically, the team will

  • Deploy a live smart building platform in collaboration with Tieto at UL Lafayette
  • Investigate the communication protocols, data collection, buffering and processing methods, and impact of failures to actual gateway devices
  • Investigate partial data-replication strategies for distributing data processing and analytics between the edge and the cloud

Experiment with partial-replication strategies for data streams. These methods will be studied in the context of smart buildings.

Project - Team

Team Member Role Email Phone Number Academic Site/IAB
Raju Gottumukkala PI (337) 482-0632 UL Lafayette
Magdy Bayoumi Co-PI (337) 482-5365 UL Lafayette
Peng Yin Researcher (337) 482-6519 UL Lafayette
Terrence Chambers Researcher (337) 482-6731 UL Lafayette
Gretchen Vanicor Researcher (337) 482-0053 UL Lafayette

Project - Novelty of Approach

Key Idea: Partial replication strategies for fault tolerance on edge devices to guarantee Quality of Service (QOS) for streaming jobs

Project - Deliverables

1 Deploy Smart Building testbed
2 Techniques for fault tolerance
3 Proof-of-concept implementation
4 Submit paper demonstrating results from the approach

Project - Benefits to IAB

  1. Saves communication & computation cost – due to scalable and reliable real-time processing on the edge (instead of cloud)
  2. Improve performance of smart building control management - in terms of availability
  3. Knowledge of what type of fault tolerance strategies work for edge computing

Project - Presentation Video

Project - Documents

projects/year7/7b.041.sbu.1565726272.txt.gz · Last modified: 2019/08/13 14:57 by matt.delcambre