OpenSHMEM and Related Technologies 2021 Workshop Program
http://www.openshmem.org/workshops/openshmem2021/
*All Times are
in PDT (Pacific US Time)
08:00-08:10 |
Welcome from Steve Poole |
08:10-09:00 |
Keynote Tittle: Bale 3.0 Applications and Conveyors [Slides] Speaker: Jason Devinney, Johns Hopkins University |
09:00-09:30 |
A Study in SHMEM: Parallel Graph Algorithm Acceleration with Distributed Shared Memory. [Slides] Michael Ing and Alan George Speaker: Michael Ing, University of Pittsburgh |
09:30-10:00 |
OpenFAM: A library for programming
disaggregated memory. [Slides] Sharad Singhal,
Clarete Crasta, Mashood Abdulla, Faizan Barmawer, Dave Emberson, Ramya
Ahobala, Gautam Bhat, Rishikesh Kumar and N Soumya Speaker: Sharad
Singhal, HPE |
10:00-10:30 |
Break |
10:30-11:00 |
OpenSHMEM
over MPI as a Performance Contender: Thorough Analysis and Optimizations.
[Slides] Min Si, Huansong Fu, Jeff
Hammond and Pavan Balaji Speaker: Min Si, Argonne National Laboratory (now Facebook) |
11:00-11:15 |
Adjourn & Announcements |
08:00-08:10 |
Welcome from Michael Raymond |
08:10-09:00 |
Keynote Address: NVSHMEM: GPU-Integrated Communication
for NVIDIA GPU Clusters [Slides] Speakers: Jim Dinan, NVIDIA |
09:00-09:30 |
SKaMPI-OpenSHMEM: Measuring OpenSHMEM Communication Routines. [Slides] Camille Coti and Allen Malony Speaker: Camille Coti, Universit ́e Sorbonne
Paris Nord / University of Oregon |
09:30-10:00 |
A Tools Information Interface for
OpenSHMEM. [Slides] Md Rahman, David
Ozog and Kieran Holland. Speaker: Md
Rahman, Intel Corporation |
10:00-10:30 |
Break |
10:30-11:00 |
CircusTent: A Tool for Measuring PGAS Performance on Emerging Architectures. [Slides] Brody Williams, John D. Leidel, Xi Wang, David Donofrio and Yong Chen Speaker: Brody Williams, Texas Tech University |
11:00-11:30 |
SHMEM-ML:
Leveraging OpenSHMEM and Apache Arrow for Scalable, Composable Machine
Learning. [Slides]
Max Grossman, Steve Poole, Howard Pritchard and Vivek Sarkar. Speaker: Max Grossman, Georgia Institute of Technology |
11:30-11:45 |
Adjourn & Announcements |
08:00-08:10 |
Welcome from Oscar Hernandez/Matt Baker |
08:10-09:00 |
Keynote Address: Multiresolution Support for Aggregated Communication in Chapel [Slides] Speaker: Brad Chamberlain, HPE (featuring contributions by Elliot Ronaghan and Engin Kayraklioglu) |
09:00-09:30 |
OpenSHMEM
Active Message Extension for Shared-Memory Tasking System Interoperability.
[Slides] Wenbin Lu, Tony
Curtis and Barbara Chapman. Speaker: Wenbin
Lu, Stony Brook University |
09:30-10:00 |
ifunc:
UCX Programming Interface for Remote Function Injection and Invocation.
[Slides] Luis E. Peña, Wenbin
Lu, Pavel Shamis and Steve Poole. Speakers: Luis
E. Pena, Arm Research and Wenbin Lu, Stony Brook University |
10:00-10:30 |
Break |
10:30-11:00 |
Can Deferring Small Messages Enhance the Performance of OpenSHMEM Applications? [Slides] David Ozog, Md Rahman and Kieran Holland Speaker: David Ozog, Intel Corporation |
11:00-11:30 |
Remote programmability model for SmartNICs in HPC workloads. [Slides] Morad Horany and Alex Margolin. Speaker: Morad Horany, Toga Networks |
11:30-12:00 |
Dynamic Heap Allocation in NVSHMEM.[Slides] Speaker: Akhil Langer, NVIDIA |
12:00-12:15 |
Closing Remarks & Wrap up |