Past Talks

IEEE Hot Interconnects Symposium 2024 - Online
(Aug 21 - 23, 2024)

Time Location Event Speaker(s)

Wednesday, August 21

All times are in UTC -7 (America/Los_Angeles)
9:00AM - 10:00AM Online

Characterizing Communication in Distributed Parameter-Efficient-Fine-Tuning for Large Language Models

[Technical Paper]

N. Alnaasan
H. Huang
A. Shafi
H. Subramoni
DK Panda

Thursday, August 22

All times are in UTC -7 (America/Los_Angeles)
1:00PM - 2:00PM Online

Demystifying the Communication Characteristics for Distributed Transformer Models

[Technical Paper]

Q. Anthony
B. Michalowicz
J. Hatef
L. Xu
M. Abduljabbar
A. Shafi
H. Subramoni
DK Panda
1:00PM - 2:00PM Online

OHIO: Improving RDMA Network Scalability in MPI_Alltoall through Optimized Hierarchical and Intra/Inter-Node Communication Overlap Design

[Technical Paper]

T. Tran
G. Kuncham
B. Ramesh
S. Xu
H. Subramoni
M. Abduljabbar
DK Panda

Friday, August 23

All times are in UTC -7 (America/Los_Angeles)
8:00AM - 10:45AM Online

High-Performance and Smart Networking Technologies for HPC and AI

[Tutorial]

DK Panda
H. Subramoni
2:00PM - 4:45PM Online

Principles and Practice of Scalable and Distributed Deep Neural Networks Training and Inference

[Tutorial]

DK Panda
H. Subramoni
A. Shafi
N. Alnaasan

Practice and Experience in Advanced Research Computing - Providence, RI
(Jul 21 - 25, 2024)

Time Location Event Speaker(s)

Monday, July 22

All Times are EST
9:00AM - 12:30PM Room 550B

TIES 2024: Artificial Intelligence and Big Data Analytics at Extreme Scale

[Workshop]

A. Shafi
DK Panda
M. Abduljabbar
9:00AM - 12:30PM Room 553A

Scalable Big Data Processing with Dask and Spark on High Performance Computing Systems

[Tutorial]

A. Shafi
DK Panda
1:30PM - 5:00PM Room 551A

ConvoHPC: Second International Workshop on Conversational AI Interfaces for HPC

[Workshop]

H. Subramoni
M. Abduljabbar
1:30PM - 5:00PM Room 551B

Principles and Practice of Scalable and Distributed Deep Neural Networks Training and Inference

[Tutorial]

A. Shafi
N. Alnaasan
H. Subramoni
DK Panda

Tuesday, July 23

All Times are EST
11:25AM - 11:50AM Ballroom B

OMB-FPGA: A Microbenchmark Suite for FPGA-aware MPIs using OpenCL and SYCL

[Technical Paper]

N. Contini
M. Abduljabbar
H. Subramoni
DK Panda
3:25PM - 3:50PM Room 553 A&B

OMB-CXL: A Micro-Benchmark Suite for Evaluating MPI Communication Utilizing Compute Express Link Memory Devices

[Technical Paper]

T. Tran
M. Abduljabbar
H. Ahn
S. Kim
Y. Park
W. Han
S. Ahn
H. Subramoni
DK Panda
3:50PM - 4:15PM Room 551 A&B

Infer-HiRes: Accelerating Inference for High-Resolution Images with Quantization and Distributed Deep Learning

[Technical Paper]

R. Gulhane
Q. Anthony
A. Shafi
H. Subramoni
DK Panda
3:50PM - 4:15PM Ballroom B

A Novel LLM-enabled Framework for Accelerating the Creation of Knowledge Graphs for HPC

[Technical Paper]

P. Kousha
V. Sathu
M. Han
J. Jani
N. Alnaasan
H. Subramoni
DK Panda

Wednesday, July 24

All Times are EST
11:25AM - 11:50AM Room 553 A&B

Design and Implementation of an IPC-based Collective MPI Library for Intel GPUs

[Technical Paper]

C. Chen
G. Kuncham
P. Kousha
H. Subramoni
DK Panda
4:45PM - 5:45PM Ballroom B

Launching the ICICLE User Group (iUG)

[BoF]

R. Machiraju
DK Panda

38th IEEE International Parallel & Distributed Processing Symposium - San Francisco, California
(May 27 - 31, 2024)

Time Location Event Speaker(s)

Monday, May 27

8:45AM - 5:00PM None

Heterogeneity in Computing

[Workshop]

DK Panda
H. Subramoni
3:30PM - 5:00PM None

Impact of LLMs and Generative AI on Future Heterogeneous Systems?

[Panel]

DK Panda

Thursday, May 30

8:30AM - 10:30AM None

HINT: Designing Cache-Efficient MPI_Alltoall using Hybrid Memory Copy Ordering and Non-Temporal Instructions

[Talk]

B. Ramesh
N. Contini
N. Alnaasan
K. Suresh
M. Abduljabbar
A. Shafi
H. Subramoni
DK Panda
1:30PM - 2:50PM None

Exploiting Inter-Layer Expert Affinity for Accelerating Mixture-of-Experts Model Inference

[Talk]

J. Yao
Q. Anthony
A. Shafi
H. Subramoni
DK Panda

Friday, May 31

11:30AM - 12:00PM None

PML-MPI: A Pre-Trained ML Framework for Efficient Collective Algorithm Selection in MPI

[Talk]

M. Han
G. Kuncham
B. Michalowicz
R. Vaidya
M. Abduljabbar
A. Shafi
H. Subramoni
DK Panda

ISC HIGH PERFORMANCE 2024 - Hamburg Germany
(May 12 - 16, 2024)

Time Location Event Speaker(s)

Sunday, May 12

All Times in Central European Summer Time
9:00AM - 1:00PM Hall Y2 - 2nd floor

High-Performance and Smart Networking Technologies for HPC and AI

[Tutorial]

H. Subramoni
DK Panda

Monday, May 13

All Times in Central European Summer Time
3:00PM - 4:00PM Foyer D-G - 2nd floor

High-Performance Semi-Supervised Learning with HARVEST: A Distributed Computer Vision Framework for Expert Labeling

[Poster Presentation] [Best Poster Finalist]

N. Alnaasan
3:00PM - 4:00PM Foyer D-G - 2nd floor

Profiling, Storing and Monitoring HPC Communication Data at Scale by OSU INAM

[Poster Presentation]

H. Subramoni
3:00PM - 4:00PM Foyer D-G - 2nd floor

A Temporal Fusion Framework for Efficient Autoregressive Model Parallel Inference

[Poster Presentation]

DK Panda
3:00PM - 4:00PM Foyer D-G - 2nd floor

High Performance & Scalable MPI Library Over Broadcom RoCE

[Poster Presentation]

S. Xu

Tuesday, May 14

All Times in Central European Summer Time
11:10AM - 11:35AM Hall F - 2nd floor

Accelerating MPI AllReduce Communication with Efficient GPU-Based Compression Schemes on Modern GPU Clusters

[Paper Presentation]

H. Subramoni
2:30PM - 3:30PM Hall G1 - 2nd floor

Research Poster Pitch & Awarding

[Poster Presentation]

N. Alnaasan
S. Xu
H. Subramoni
DK Panda

Thursday, May 16

All Times in Central European Summer Time
2:00PM - 6:00PM Hall Y7 - 2nd floor

Ninth International Workshop on Communication Architectures for HPC, Big Data, Deep Learning and Clouds at Extreme Scale

[Workshop]

H. Subramoni
A. Shafi
DK Panda

The 20th annual OpenFabrics Alliance (OFA) Virtual Workshop - Online
(Apr 22 - 23, 2024)

Time Location Event Speaker(s)

Monday, April 22

All Times in PT
10:15AM - 10:45AM Online

Accelerating MPI AllReduce Communication with Efficient GPU-Based Compression Schemes on Modern GPU Clusters

[Talk]

H. Subramoni
Q. Zhou
11:00AM - 11:30AM Online

High Performance & Scalable MPI library over Broadcom RoCE

[Talk]

M. Abduljabbar
H. Shah, Broadcom Inc.
S. Xu
1:00PM - 1:30PM Online

Scaling Large Language Model Training using Hybrid GPU-based Compression in MVAPICH

[Talk]

A. Shafi
L. Xu

Tuesday, April 23

All Times in PT
10:15AM - 10:45AM Online

Optimized All-to-all Connection Establishment for High-Performance MPI Libraries over InfiniBand

[Talk]

M. Abduljabbar
DK Panda