Researchers at the TSSG of Waterford Institute of Technology have received $50,000 funding from CISCO Research Centre to investigate how the internet connected devices around us can be harnessed to improve how Internet services are consumed.
Currently Internet services are delivered to consumers from the ‘cloud’, over the Internet and to our devices. Take for example Netflix, our streaming content is being delivered from the Netflix platform running on the Amazon Cloud. As more and more devices with computing capabilities become connected to the Internet, these devices can also potentially participate in the delivery of Internet services. In essence, the computing resources are moving beyond the ‘cloud’ and into the ‘fog’. This new paradigm shift can bring many advantages if used effectively, such as lower cost, faster service delivery and improved data protection to name but a few. However this paradigm shift requires a fundamental rethink as to how Internet services are deployed and operated.
The team at TSSG made up of Dr. Alan Davy and Mohit Taneja, will focus on how to design, develop and deploy part of these services on the devices connected to the Internet and on the cloud, aiming to achieve a fruitful interplay between fog and cloud. This requires investigating how the services are currently handled and how they now need to be re-designed to be deployed on these devices. The project seeks to develop a framework to achieve that by means of ‘decomposition of computing programs’ to generate partial computing units that can be hosted on the Internet connected devices without affecting their natural working.
The idea is that if the device cannot hold the full service, then only put the part of the service that can be handled by it. The project specifically focuses on decomposing and re-thinking the design of data analytics services that consume the data generated from the devices connected to the Internet to generate insights into the system.
The project aims to achieve the following technical outputs from this research :
- The use of machine learning based application classification for dynamic service sharing between fog and cloud resources for efficient resource utilization.
- The ability to provide data analytics capability on fog resources without paralyzing the network core by means of decomposition of analytic programs. To support the future of a tech-driven smart wold, the vision here is to provide data driven intelligence capability on the network edge to support the Internet of Things (IoT) oriented big data analytics.
- A hybrid microservices based approach to develop the whole framework in fog enabled scenarios with cloud support. This is essential from a practical standpoint of tying the above two together, and is a challenge very few teams in the research community have combined and looked into so far.
Once validated and further developed beyond this work, this research will contribute to sustainable computing solutions for the tech-driven world. The support from industry leaders such as Cisco helps progress this research closer towards market update.
M. Taneja, N. Jalodia and A. Davy, “Distributed Decomposed Data Analytics in Fog Enabled IoT Deployments,” in IEEE Access, vol. 7, pp. 40969-40981, 2019.