FORGOT YOUR DETAILS?

2022 OFA Virtual Workshop

The 18th annual OpenFabrics Alliance (OFA) Workshop took place irtually April 26-28, 2022. It is the only event of its kind where fabric developers and users can discuss emerging fabric technologies, collaborate on future industry requirements, and address problems that exist today.

This year, we hosted daily trainings and tutorials alongside the technical sessions track. Additional details can be found below.

Download the abstracts from here.

Below is the OFA Virtual Workshop 2022 agenda.

Trainings/Tutorials

Tuesday, April 26

Fabric Software Develop Platform (FSDP) “Office Hours”
Doug Ledford, Red Hat; Tatyana Nikolova, Intel Corp.; Brian Chae, Red Hat; Afom Michael, Red Hat
8:30am-1:30pm

Wednesday, April 27

Introduction to Networking Technologies for High-Performance Computing
Dhabaleswar Panda, The Ohio State University; Hari Subramoni, The Ohio State University
8:00-11:30am

Thursday, April 28

High Performance Machine Learning, Deep Learning, and Data Science
Dhabaleswar Panda, The Ohio State University; Hari Subramoni, The Ohio State University; Arpan Jain, The Ohio State University; Aamir Shafi, The Ohio State University
8:00-11:30am

Technical Sessions

Tuesday, April 26

Accelerating MPI and Deep Learning Applications with the DPU Technology
VIDEO
Dhabaleswar Panda, The Ohio State University
8:00-8:30am

Performance Evaluation of MPI on the Slingshot Interconnect
VIDEO
Dhabaleswar Panda, The Ohio State University
8:30-9:00am

Designing High-Performance Alltoall Solutions for Dense GPU Systems
VIDEO
Hari Subramoni, The Ohio State University
9:00-9:30am

Break
9:30-10:00am

Performance Characterization of MPI Libraries on Arm-based AWS HPC Cloud Instances with Elastic Fabric Adapters
VIDEO
Aamir Shafi, The Ohio State University
10:00-10:30am

Offloading Scatter-Gather via Custom Accelerators on COPA FPGA Network Platform
VIDEO
Shweta Jain, Intel Corp.
10:30-11:00am

Omni-Path Express (OPX) Libfabric Provider: Overview & Status
VIDEO
Tim Thompson, Cornelis Networks
11:00-11:30am

Break
11:30am-12:30pm

Scalable Storage Management Demonstrations – SNIA Swordfish™: In Action Demonstration
VIDEO
Richelle Ahlvers, Vice Chair and Executive Committee, SNIA Board of Directors/Storage Technology Enablement Architect, Intel Corp.
12:30-1:00pm

Computational Storage Demonstrations
1:00-2:00pm


High Performance Object Storage (HPOS) with SmartSSD Demonstration
VIDEO
Mayank Saxena, SNIA Member, SNIA Computational Storage Special Interest Group/Senior Director,  Engineering, Samsung 

---Far Edge Retail chain and Far Edge/Near Edge Data Security Use Case Demonstrations
VIDEO
J B Baker, SNIA Member, SNIA Computational Storage Special Interest Group/Senior Director, of Product Management, ScaleFlux

---AI/ML Workloads Run in Drive and Distributed/Parallel Processing Demonstrations
VIDEO
Eli Tiomkin, SNIA Member, SNIA Computational Storage Technical Work Group/VP Business Development, NGD Systems

Wednesday, April 27

SOFA-Storage: Creating a vendor agnostic framework to enable seamless storage offload using SmartNICs
VIDEO
Raphael Polig. IBM
8:00-8:30am

Exploiting RDMA Mistakes in NVMe-oF Storage Applications
VIDEO
Konstantin Taranov, ETH Zurich
8:30-9:00am

Graphcore IPU over Fabric (IPUoF)
VIDEO
Wei Lin Guay, Graphcore
9:00-9:30am

Break
9:30-10:00am

In-Network Collective Communication Accelerations
VIDEO
Sean Hefty, Intel Corp.
10:00-10:30am

Communication and Computation API Composability - XPU support in OFI
VIDEO
Sean Hefty, Intel Corp.
10:30-11:00am

PCIe® 6.0 Specification: A High-Performance I/O Interconnect for Advanced Networking Applications
VIDEO
Debendra Das Sharma, PCI-SIG
11:00am-11:30am

Break
11:30am-12:30pm

Towards Universal Management of Fabric Interfaces
VIDEO
Richelle Ahlvers, Intel Corp.
12:30-1:00pm

Diving Into the New Wave of Storage Management
VIDEO
Richelle Ahlvers, Intel Corp.
1:00pm-1:30pm

OpenFabrics Management Framework (OFMF) Demonstration
VIDEO
Russ Herrell, HPE; Jim Hull, IntelliProp
1:30-2:00pm

Thursday, April 28

Bridging RDMACM Traffic between InfiniBand and ROCE
VIDEO
Christoph Lameter, Deutche Boerse
8:00-8:30am

sPIN: High-performance streaming Processing in the Network
VIDEO
Salvatore Di Girolamo, ETH Zurich
8:30-9:00am

HatRPC: Hint-Accelerated Apache Thrift RPC over RDMA
VIDEO
Xiaoyi Lu, University of California Merced
9:00-9:30am

Break
9:30-10:00am

Update on PSM3 Capabilities and Architecture
VIDEO
James Erwin, Intel Corp.
10:00-10:30am

Cloud-Native Supercomputing: Bare-Metal, Secured Supercomputing Architecture
VIDEO
Gilad Shainer, NVIDIA
10:30-11:00am

Status of OpenFabrics Interfaces (OFI) Support in MPICH
VIDEO
Yanfei Guo, Argonne National Laboratory
11:00am-11:30pm

Benefits of Compute Express Link™ (CXL™) for High-Performance Computing
VIDEO
CXL Consortium
11:30am-12:00pm

TOP