Amit Samanta

mail_outline amit@cs.utah.edu location_on South Central Campus Drive photo_camera Photo

I am a PhD student at the Kahlert School of Computing of the University of Utah, working with Prof. Ryan Stutsman as a member of the Utah Scalable Computer Systems Lab.
I'm broadly interested in computer systems, including storage and file systems, distributed systems, operating systems, and networking.
I build systems that enable reliable, efficient, and defensible systems in the cloud. My work aims to:

• Understand the performance variability of existing and emerging cloud systems.
• Propose fair and efficient policies for end-to-end resource management to existing services.
• Design novel abstractions and frameworks to simplify and debug concurrent and fault-tolerant applications.
• Specialize applications to run efficiently on new cloud architectures.

Before joining Utah, I obtained my master's degree in Computer Science from IIT Kharagpur, India. I was an exchange scholar at Tsinghua University, China. I also worked as
a visiting research scholar at MPI-SWS in Germany, scientific research assistant at TU Darmstadt in Germany and research Intern at TCS Innovation Labs in India.

Selected Publications

• Fair, Efficient Multi-Resource Scheduling for Stateless Serverless Functions with Anubis. [pdf] [bibtex]
  Amit Samanta, and Ryan Stutsman
  In IEEE/ACM International Symposium on Cluster, Cloud and Internet Computing (CCGrid), Philadelphia, USA, May 2024.

• Persistent-Memory-Aware Scheduling for Serverless Workloads. [pdf] [link] [bibtex]
  Amit Samanta, Faraz Ahmed, Lianjie Cao, Ryan Stutsman and Puneet Sharma.
  In IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW), Florida, USA, May 2023.

• A Case of Multi-Resource Fairness for Serverless Workflows. [pdf] [bibtex]
  Amit Samanta, and Ryan Stutsman.
  In ACM/SPEC International Conference on Performance Engineering (ICPE), Coimbra, Portugal, April 2023.

• ASAP: Adaptive and Scalable Microservice Provisioning for Edge-IoT Networks. [pdf] [bibtex]
  Amit Samanta, Flavio Esposito, and Tri Gia Nguyen.
  In IEEE/IFIP Wireless On-Demand Network Systems and Services Conference (WONS), Italy, February 2023.

• Dyme: Dynamic Microservice Scheduling in Edge Computing Enabled IoT. [pdf] [bibtex]
  Amit Samanta, and Jianhua Tang.
  IEEE Internet of Things Journal (IoTJ), vol. 7, no. 7, pp. 6164-6174, July 2020.

• No DNN Left Behind: Improving Inference in the Cloud with Multi-Tenancy. [pdf] [bibtex]
  Amit Samanta, Suhas Shrinivasan, Antoine Kaufmann, and Jonathan Mace.
  In arXiv:1901.06887, January 2019 (Vision Paper). [pdf]
  Poster: In ACM European Conference on Computer Systems Dresden (EuroSys), Dresden, German, March 2019. [pdf]
  Poster: In USENIX Symposium on Networked Systems Design and Implementation (NSDI), Boston, USA, February 2019. [pdf]

• Incentivizing Microservices for Online Resource Sharing in Edge Clouds. [pdf] [bibtex]
  Amit Samanta, Lei Jiao, Max Mühlhäuser, and Lin Wang.
  In IEEE International Conference on Distributed Computing Systems (ICDCS), Dallas, TX, USA, July 2019.

• Battle of Microservices: Towards Latency-Optimal Heuristic Scheduling for Edge Computing. [pdf] [bibtex]
  Amit Samanta, Yong Li, and Flavio Esposito.
  In IEEE Conference on Network Softwarization (NetSoft), pp. 223-227, Paris, France, June 2019.

For more details, please refer to full publication list .

Selected Talks

• "Persistent-Memory-Aware Scheduling for Serverless Workloads"
   IEEE ESSA@IPDPS (Florida, USA): May 2023
   Hewlett-Packard Labs (Remote): April 2023

• "Incentivizing Microservices for Online Resource Sharing in Edge Clouds"
   IEEE ICDCS (Dallas, Texas, USA): July 2019

• "(Poster) No DNN Left Behind: Improving Inference in the Cloud with Multi-Tenancy"
   ACM EuroSys (Dresden, Germany): March 2019
   USENIX NSDI (Boston, MA, USA): February 2019

• "EReM: Energy-Efficient Resource Management in Body Area Networks with Fault Tolerance"
   ACM India Annual Event IRISS (Nagpur, India): February 2018
   IEEE GLOBECOM (Singapore): December 2017
   Tsinghua University, FIB Lab Seminar (Beijing, China): September 2017