Open Systems Laboratory at Illinois

Atul Sandur

I am a PhD candidate in CS @ Illinois, working with Prof. Gul Agha in Open systems laboratory. My research interest is broadly in distributed systems, with my current focus being on Internet-of-things (IoT) applications that interact with their physical environment using sensors and actuators. Such systems operate in a dynamic environment and their interactions with other collocated applications aren''t well understood. Further, they are deployed on heterogeneous and resource-constrained devices (such as smartphones, smart sensors and other edge devices), making it challenging to perform complex computations on them. I am currently working on ways to provide performance, energy-consumption and privacy guarantees for such systems.

Previously, I was working with cloud monitoring and diagnostics group at Microsoft Azure, Redmond. I got to work on a wide variety of problems ranging from data modeling for cloud service analytics, logging/reporting infrastructure to big data processors and diagnostics log management services. Prior to that, I graduated with my Masters from University of Michigan, Ann Arbor, where my research was focused on context aware computing for structural health monitoring applications like highway bridge inspections. I went to BITS, Pilani, Goa campus, for my undergraduate studies in Electrical and Electronics engineering.

Research Interests: Distributed systems, mobile/ubiquitous computing, edge computing and IoT applications, adaptive systems



  1. Tarek Elgamal, Atul Sandur, Phuong Nguyen, Klara Nahrstedt, and Gul Agha. DROPLET: distributed operator placement for iot applications spanning edge and cloud resources. In 11th IEEE International Conference on Cloud Computing, CLOUD 2018, San Francisco, CA, USA, July 2-7, 2018, 1–8. IEEE Computer Society, 2018.
  2. Tarek Elgamal, Atul Sandur, Gul Agha, and Klara Nahrstedt. Distributed placement of machine learning operators for iot applications spanning edge and cloud resources. In SysML Conference 2018, Stanford, CA, USA, February 15-16, 2018, 1–3. 2018.
  3. Atul Sandur, Tarek Elgamal, Gul Agha, and Klara Nahrstedt. Complexity analysis for droplet: distributed operator placement for iot applications spanning edge and cloud resources. In Illinois Digital Environment for Access to Learning and Scholarship, 1–6. 2018.
  4. Gul Agha, Minas Charalambides, Kirill Mechitov, Karl Palmskog, Atul Sandur, and Reza Shiftehfar. Inferring and enforcing use patterns for mobile cloud assurance. In R. H. Campbell and K. A. Kwiat C. A. Kamhoua, editors, Assured Cloud Computing, pages 237–376. Wiley/IEEE Press, 2018.