NOWLAB :: Publications

Journals (5)
1	DK Panda, H. Subramoni, C. Chu, and M. Bayatpour, The MVAPICH project: Transforming Research into High-Performance MPI Library for HPC Community , Journal of Computational Science (JOCS), Special Issue on Translational Computer Science, Oct 2020.
2	J. Hashmi, C. Chu, S. Chakraborty, M. Bayatpour, H. Subramoni, and DK Panda, FALCON-X: Zero-copy MPI Derived Datatype Processing on Modern CPU and GPU Architectures, Journal of Parallel and Distributed Computing (JPDC), Volume 144, October 2020, Pages 1-13, doi.org/10.1016/j.jpdc.2020.05.008,
3	Ammar Awan, A. Jain, C. Chu, H. Subramoni, and DK Panda, Communication Profiling and Characterization of Deep Learning Workloads on Clusters with High-Performance Interconnects, IEEE Micro, vol. 40, no. 1, pp. 35-43, 1 Jan.-Feb. 2020.,
4	Ammar Awan, K. Vadambacheri Manian, C. Chu, H. Subramoni, and DK Panda, Optimized Large-Message Broadcast for Deep Learning Workloads: MPI, MPI+NCCL, or NCCL2?, Volume 85, July 2019, Pages 141-152, https://doi.org/10.1016/j.parco.2019.03.005,
5	C. Chu, X. Lu, Ammar Awan, H. Subramoni, Bracy Elton, and DK Panda, Exploiting Hardware Multicast and GPUDirect RDMA for Efficient Broadcast, IEEE Transactions on Parallel and Distributed Systems (TPDS), vol. 30, no. 3, pp. 575-588, 1 March 2019,

Conferences & Workshops (22)
1	Designing a ROCm-aware MPI Library for AMD GPUs: Early Experiences K. Khorassani, J. Hashmi, C. Chu, C. Chen, H. Subramoni, and DK Panda, ISC HIGH PERFORMANCE 2021, Jun 2021 [Bib - Plain]
2	Designing High-Performance MPI Libraries with On-the-fly Compression for Modern GPU Clusters Q. Zhou, C. Chu, N. Senthil Kumar, P. Kousha, M. Ghazimirsaeed, H. Subramoni, and DK Panda, 35th IEEE International Parallel & Distributed Processing Symposium, May 2021 [Best Paper Finalist] [Bib - Plain]
3	Adaptive and Hierarchical Large Message All-to-all Communication Algorithms for Large-scale Dense GPU Systems K. Khorassani, C. Chu, Q. Anthony, H. Subramoni, and DK Panda, The 21st IEEE/ACM International Symposium on Cluster, Cloud and Internet Computing, May 2021 [Bib - Plain]
4	Dynamic Kernel Fusion for Bulk Non-contiguous Data Transfer on GPU Clusters C. Chu, K. Khorassani, Q. Zhou, H. Subramoni, and DK Panda, 22nd IEEE International Conference on Cluster Computing (IEEE Cluster 2020), Sep 2020 [Bib - Plain]
5	NV-Group: Link-Efficient Reductions for Distributed Deep Learning on Modern Dense GPU Systems C. Chu, P. Kousha, Ammar Awan, K. Khorassani, H. Subramoni, and DK Panda, The 34th ACM International Conference on Supercomputing (ICS-2020), Jun 2020 [Bib - Plain]
6	Designing a Profiling and Visualization Tool for Scalable and In-Depth Analysis of High-Performance GPU Clusters P. Kousha, B. Ramesh, K. Suresh, C. Chu, A. Jain, N. Sarkauskas, H. Subramoni, and DK Panda, 26th IEEE International Conference on High Performance Computing, Data, Analytics and Data Science, Dec 2019 [Bib - Plain]
7	High-Performance Adaptive MPI Derived Datatype Communication for Modern Multi-GPU Systems C. Chu, J. Hashmi, K. Khorassani, H. Subramoni, and DK Panda, 26th IEEE International Conference on High Performance Computing, Data, Analytics and Data Science (HiPC '19), Dec 2019 [Bib - Plain]
8	OMB-UM: Design, Implementation, and Evaluation of CUDA Unified Memory Aware MPI Benchmarks K. Vadambacheri Manian, C. Chu, Ammar Awan, K. Khorassani, H. Subramoni, and DK Panda, 10th International Workshop in Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems, Nov 2019 [Bib - Plain]
9	Communication Profiling and Characterization of Deep-Learning Workloads on Clusters With High-Performance Interconnects Ammar Awan, A. Jain, C. Chu, H. Subramoni, and DK Panda, 26th Symposium on High-Performance Interconnects (HotI '19), Aug 2019 [Bib - Plain]
10	Performance Evaluation of MPI Libraries on GPU-enabled OpenPOWER Architectures: Early Experiences K. Khorassani, C. Chu, H. Subramoni, and DK Panda, International Workshop on OpenPOWER for HPC, held in conjunction with ISC'19, Jun 2019 [Bib - Plain]
11	C-GDR: High-Performance Container-aware GPUDirect MPI Communication Schemes on RDMA Networks J. Zhang, X. Lu, C. Chu, and DK Panda, 33rd IEEE International Parallel & Distributed Processing Symposium (IPDPS '19), May 2019 [Bib - Plain]
12	Scalable Distributed DNN Training using TensorFlow and CUDA-Aware MPI: Characterization, Designs, and Performance Evaluation Ammar Awan, J. Bedorf, C. Chu, H. Subramoni, and DK Panda, The 19th Annual IEEE/ACM International Symposium in Cluster, Cloud, and Grid Computing (CCGRID 2019), May 2019 [Bib - Plain]
13	Characterizing CUDA Unified Memory (UM)-AwareMPI Designs on Modern GPU Architectures K. Vadambacheri Manian, Ammar Awan, A. Ruhela, C. Chu, and DK Panda, 12th Workshop on General Purpose Processing Using GPU (GPGPU 2019) @ ASPLOS 2019, Apr 2019 [Bib - Plain]
14	OC-DNN: Exploiting Advanced Unified Memory Capabilities in CUDA 9 and Volta GPUs for Out-of-Core DNN Training Ammar Awan, C. Chu, H. Subramoni, X. Lu, and DK Panda, 25th IEEE International Conference on High Performance Computing, Data, and Analytics, Dec 2018 [Bib - Plain]
15	Optimized Broadcast for Deep Learning Workloads on Dense-GPU InfiniBand Clusters: MPI or NCCL? Ammar Awan, C. Chu, H. Subramoni, and DK Panda, The EuroMPI 2018 Conference, Sep 2018 [Bib - Plain]
16	MPI-GDS: High Performance MPI Designs with GPUDirect-aSync for CPU-GPU Control Flow Decoupling A. Venkatesh, C. Chu, K. Hamidouche, S. Potluri, Davide Rossetti, and DK Panda, ICPP 2017 : International Conference on Parallel Processing, Aug 2017 [Bib - Plain]
17	Efficient and Scalable Multi-Source Streaming Broadcast on GPU Clusters for Deep Learning C. Chu, X. Lu, Ammar Awan, H. Subramoni, J. Hashmi, Bracy Elton, and DK Panda, ICPP 2017 : International Conference on Parallel Processing, Aug 2017 [Bib - Plain]
18	Efficient Reliability Support for Hardware Multicast-based Broadcast in GPU-enabled Streaming Applications C. Chu, K. Hamidouche, H. Subramoni, A. Venkatesh, B. Elton, and DK Panda, First Workshop on Optimization of Communication in HPC runtime systems (COMHPC, SC Workshop), Nov 2016 [Bib - Plain]
19	Designing High Performance Heterogeneous Broadcast for Streaming Applications on GPU Clusters C. Chu, K. Hamidouche, H. Subramoni, A. Venkatesh, B. Elton, and DK Panda, 28th International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD'16), Oct 2016 [Bib - Plain]
20	Exploiting Maximal Overlap for Non-Contiguous Data Movement Processing on Modern GPU-enabled System C. Chu, K. Hamidouche, A. Venkatesh, D. Banerjee, H. Subramoni, and DK Panda, The 30th IEEE International Parallel & Distributed Processing Symposium (IPDPS '16), May 2016 [Bib - Plain]
21	CUDA Kernel based Collective Reduction Operations on Large-scale GPU Clusters C. Chu, K. Hamidouche, A. Venkatesh, Ammar Awan, and DK Panda, 16th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGrid'16), May 2016 [Bib - Plain]
22	A Case for Non-Blocking Collectives in OpenSHMEM: Design, Implementation, and Performance Evaluation using MVAPICH2-X Ammar Awan, K. Hamidouche, C. Chu, and DK Panda, OpenSHMEM 2015 for PGAS Programming in the Exascale Era, Aug 2015 [Bib - Plain]

Ph.D. Disserations (1)
1	C. Chu, Accelerator-enabled Communication Middleware for Large-scale Heterogeneous HPC Systems with Modern Interconnects, Jul 2020

NOWLAB: Network Based Computing Lab

http://web.cse.ohio-state.edu/~chuc/

This page lists the publications by Ching-Hsiang Chu

Journals (5)

Conferences & Workshops (22)

Designing a ROCm-aware MPI Library for AMD GPUs: Early Experiences

Designing High-Performance MPI Libraries with On-the-fly Compression for Modern GPU Clusters

Adaptive and Hierarchical Large Message All-to-all Communication Algorithms for Large-scale Dense GPU Systems

Dynamic Kernel Fusion for Bulk Non-contiguous Data Transfer on GPU Clusters

NV-Group: Link-Efficient Reductions for Distributed Deep Learning on Modern Dense GPU Systems

Designing a Profiling and Visualization Tool for Scalable and In-Depth Analysis of High-Performance GPU Clusters

High-Performance Adaptive MPI Derived Datatype Communication for Modern Multi-GPU Systems

OMB-UM: Design, Implementation, and Evaluation of CUDA Unified Memory Aware MPI Benchmarks

Communication Profiling and Characterization of Deep-Learning Workloads on Clusters With High-Performance Interconnects

Performance Evaluation of MPI Libraries on GPU-enabled OpenPOWER Architectures: Early Experiences

C-GDR: High-Performance Container-aware GPUDirect MPI Communication Schemes on RDMA Networks

Scalable Distributed DNN Training using TensorFlow and CUDA-Aware MPI: Characterization, Designs, and Performance Evaluation

Characterizing CUDA Unified Memory (UM)-AwareMPI Designs on Modern GPU Architectures

OC-DNN: Exploiting Advanced Unified Memory Capabilities in CUDA 9 and Volta GPUs for Out-of-Core DNN Training

Optimized Broadcast for Deep Learning Workloads on Dense-GPU InfiniBand Clusters: MPI or NCCL?

MPI-GDS: High Performance MPI Designs with GPUDirect-aSync for CPU-GPU Control Flow Decoupling

Efficient and Scalable Multi-Source Streaming Broadcast on GPU Clusters for Deep Learning

Efficient Reliability Support for Hardware Multicast-based Broadcast in GPU-enabled Streaming Applications

Designing High Performance Heterogeneous Broadcast for Streaming Applications on GPU Clusters

Exploiting Maximal Overlap for Non-Contiguous Data Movement Processing on Modern GPU-enabled System

CUDA Kernel based Collective Reduction Operations on Large-scale GPU Clusters

A Case for Non-Blocking Collectives in OpenSHMEM: Design, Implementation, and Performance Evaluation using MVAPICH2-X

Ph.D. Disserations (1)