![](assets/img/participants/amit-samanta.jpg)
I am a Ph.D. student in the Computer Science Department at the University of Utah, USA. I am advised by Prof. Ryan Stutsman. My research interests include serverless, edge, and large-scale software systems. I have received my M.S. degree from the Department of Computer Science and Engineering, Indian Institute of Technology Kharagpur, India. I was a visiting research scholar at the Max Planck Institute for Software Systems, Saarbrücken, Germany, and an exchange scholar with the Department of Electronic Engineering at Tsinghua University, Beijing, China.
The explosion of the Internet of Things (IoT) devices within cyber-physical systems (CPS) has ushered in a new era of intelligent and interconnected infrastructure. From smart factories and self-driving cars to connected healthcare systems and intelligent grids, CPS transforms how we interact with the physical world. However, the ability of these systems to function effectively hinges on their ability to process vast amounts of data in real time, often at the network edge where devices and sensors reside. This poses a significant challenge due to the resource-constrained nature of edge devices, which often lack the processing power and memory necessary for complex computations. To address this challenge, my research delves into two key approaches: (a) Edge-Cloud Orchestration. The optimal balance between processing data at the edge and sending it to the cloud depends on various factors, such as real-time requirements and resource availability. Edge-cloud orchestration plays a crucial role in dynamically managing this distribution. (1) Resource Scheduling: I designed RIASE, a distributed resource allocation and offloading scheme; it deals with which tasks should be executed at the edge and which should be offloaded to the cloud based on their resource demands and latency requirements. (2) Incentive Mechanism: I proposed mISO, a novel incentive mechanism for offloading computational tasks in edge environments for CPS. It addresses the issue of incentivizing mobile IoT devices to participate in the offloading process, ensuring efficient resource utilization and improved performance. (b) Serverless Computing for Agile and Efficient CPS. Edge devices in CPS environments often operate with limited processing power and memory. Serverless functions alleviate this burden by eliminating the need for local servers. Developers can design lightweight functions that execute specific tasks efficiently, minimizing resource consumption on edge devices (i.e., sensors or embedded controllers). (1) Resource Management: I designed Anubis, which allocates resources fairly and efficiently among serverless functions and fulfills the SLA requirements of participating tenants. It helps eliminate SLA miss rates for response-time-sensitive serverless workloads, even when multiple concurrent workloads compete for multiple resources (CPU, memory, network). (2) Data Management: I am also designing Mosaic, an elastic storage engine for ephemeral data management of serverless applications.