Aishwarya Sarkar

Aishwarya Sarkar

Position
  • PhD Student
Aishwarya is a Ph.D. student in the Software Analytics and Pervasive Parallelism Lab, working under the supervision of Dr. Ali Jannesari. Her research focuses on developing advanced techniques to optimize communication and computational efficiency in distributed neural network architectures. As an intern at Pacific Northwest National Laboratory (PNNL), she is currently improving the scalability and performance of Graph Neural Networks (GNNs) and Large Language Models (LLMs) in large-scale distributed environments.

Area of Expertise

  • Deep Learning
  • Parallel Computing
  • High Performance Computing

Education

  • B.S., Computer Science and Engineering, Institute of Engineering & Management, Maulana Abul Kalam Azad University of Technology
  • M.S., Computer Science, Iowa State University

Publications

  1. Aishwarya Sarkar, Sayan Ghosh, Nathan R. Tallent, Ali Jannesari: “MassiveGNN: Efficient Training via Prefetching for Massively Connected Distributed Graphs”, 2024 IEEE International Conference on Cluster Computing (CLUSTER), Kobe, Japan, 2024
  2. Chen, L., Ahmed, N.K., Dutta, A., Bhattacharjee, A., Yu, S., Mahmud, Q.I., Abebe, W., Phan, H., Sarkar, A., Butler, B. and Hasabnis, N., 2024. The Landscape and Challenges of HPC Research and LLMs. CoRR https://arxiv.org/pdf/2402.02018v3
  3. Aishwarya Sarkar, Jien Zhang, Chaoqun Lu, Ali Jannesari: "Domain-Aware Scalable Distributed Training for Geo-Spatiotemporal Data." 2023 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW). IEEE, 2023. https://ieeexplore.ieee.org/abstract/document/10196650
  4. Aishwarya Sarkar, Chaoqun Lu, Ali Jannesari: Accelerating Domain-aware Deep Learning Models with Distributed Training. Workshop on Multi-scale, Multi-physics and Coupled Problems on highly parallel systems (MMCP), co-located with HPC Asia, pages 1–5, February 2023 https://arxiv.org/pdf/2301.11787
  5. Aishwarya Sarkar, Jien Zhang, Chaoqun Lu, Ali Jannesari: Transfer Learning Approaches for Knowledge Discovery in Grid-based Geo-Spatiotemporal Data. NeurIPS 2021 AI for Science Workshop, 2021 https://openreview.net/pdf?id=mC6-nsYtacP