Hari Subramoni

Dr. Hari Subramoni


Research Focus & Metrics

Academic Career Length

17Years (2010-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
13 April 2011 in Computer Science - Research and Development
Verified
Journal Article
Closed access
PDF file
KKKrishna Kandalla
Krishna Kandalla
HSHari Subramoni
Hari Subramoni
KTKaren Tomko
Karen Tomko
DPDmitry Pekurovsky
Dmitry Pekurovsky
SSSayantan Sur
Sayantan Sur
DPDhabaleswar K. Panda
Dhabaleswar K. Panda

29 May 2022
Verified
Book Chapter
Closed access
PDF file
QZQinghua Zhou
Qinghua Zhou
PKPouya Kousha
Pouya Kousha
QAQuentin Anthony
Quentin Anthony
KKKawthar Shafie Khorassani
Kawthar Shafie Khorassani
ASAamir Shafi
Aamir Shafi
DPDhabaleswar K. Panda
Dhabaleswar K. Panda

17 June 2021
Verified
Book Chapter
Closed access
PDF file
KKKawthar Shafie Khorassani
Kawthar Shafie Khorassani
JHJahanzeb Hashmi
Jahanzeb Hashmi
CCChing-Hsiang Chu
Ching-Hsiang Chu
CCChen-Chun Chen
Chen-Chun Chen
HSHari Subramoni
Hari Subramoni
DPDhabaleswar K. Panda
Dhabaleswar K. Panda

26 August 2018 in IEEE Transactions on Parallel and Distributed Systems
Verified
Journal Article
Closed access
CCChing-Hsiang Chu
Ching-Hsiang Chu
XLXiaoyi Lu
Xiaoyi Lu
AAAmmar A. Awan
Ammar A. Awan
HSHari Subramoni
Hari Subramoni
BEBracy Elton
Bracy Elton
DPDhabaleswar K. Panda
Dhabaleswar K. Panda
Broadcast is a widely used operation in many streaming and deep learning applications to disseminate large amounts of data on emerging heterogeneous High-Performance Computing (HPC) systems. However, traditional broadcast schemes do not fully utilize hardware features for Graphics Processing Unit (GPU)-based applications. In this paper, a model-oriented analysis is presented to identify performance bottlenecks of existing broadcast schemes on GPU clusters. Next, streaming-based broadcast schemes are proposed to exploit InfiniBand hardware multicast (IB-MCAST) and NVIDIA GPUDirect technology for efficient message transmission. The proposed designs are evaluated in the context of using Message Passing Interface (MPI) based benchmarks and applications. The experimental results indicate improved scalability and up to 82 percent reduction of latency compared to the state-of-the-art solutions in the benchmark-level evaluation. Furthermore, compared to the state-of-the-art, the proposed design yields stable higher throughput for a synthetic streaming workload, and 1.3x faster training time for a deep learning framework.

12 August 2024
Verified
Conference Paper
Open access
QAQuentin Anthony
Quentin Anthony
JHJacob Hatef
Jacob Hatef
DNDeepak Narayanan
Deepak Narayanan
SBStella Biderman
Stella Biderman
SBStas Bekman
Stas Bekman
DPDhabaleswar Panda
Dhabaleswar Panda
While GPUs are responsible for training the vast majority of state-of-the-art deep learning models, the implications of their architecture are often overlooked when designing new deep learning (DL) models. As a consequence, modifying a DL model to be more amenable to the target hardware can significantly improve the runtime performance of DL training and inference. In this paper, we provide a set of guidelines for users to maximize the runtime performance of their transformer models. These guidelines have been created by carefully considering the impact of various model hyperparameters controlling model shape on the efficiency of the underlying computation kernels executed on the GPU. We find the throughput of models with “efficient” model shapes is up to 39% higher while preserving accuracy compared to models with a similar number of parameters but with unoptimized shapes.

12 May 2017
Verified
Book Chapter
Closed access
PDF file
HSHari Subramoni
Hari Subramoni
SCSourav Chakraborty
Sourav Chakraborty
DPDhabaleswar K. Panda
Dhabaleswar K. Panda