Room 300 Room 301 Room 302 Room 304 Room 311
HP-Cast 2020 Singapore NIVIDIA
Fundamentals of
Deep Learning for
Computer Vision
IBM
Spectrum Scale Forum
AM: Altair
PBS Works User Group
PM: RIST
Manycore Architecture Tutorial
PM: Pawsey
Containers in HPC
0900 – 0930

0900 – 1030

Join us at HP-CAST 2020 (co-located with SupercomputingAsia 2020) as we share with you the latest HPC innovations, including key highlights on:

Directions for the Exascale Era

Bill Mannel, Vice President & General Manager, HPC AI, HPE

Exascale Systems – The Next Stage in High Performance Computing

Nicholas Dube, HPE Fellow, Office of CTO, HPC and AI, HPE

A New Generation of HPC Storage

Ulrich Plechschmidt, Worldwide Product Marketing HPC Storage at Cray, HPC & AI, HPE

Memory-Driven Computing

Mike Woodacre, HPE Fellow, Chief Technologist, Compute Solutions

Data Management Framework

Customer sharing

0900 – 0915
Introduction
> Meet the instructor.
> Create an account at courses.nvidia.com/join

0915 – 1030
Unlocking New Capabilities
> Learn the biological inspiration behind deep neural networks (DNNs).
> Explore training DNNs with big data.
> Train neural networks to perform image classification by harnessing the three main ingredients of deep learning: deep neural networks, big data, and the GPU.

0930 – 0945

New users: Welcome

0945 – 1030

New users: Spectrum Scale use cases

Speaker – Chris Maestas (IBM)

0900 – 1000

PBS Works User Group Inauguration and road-map

Dr Bill Nitzberg – CTO, PBS Works

1000 – 1020

HPC reliability improvement and container integration with PBS Works

Mr SOH Hwee Jin, Melvin – Senior Assistant Director – High Performance Computing Centre, NTU

Mr Edwin Tan Seng Tat – Senior Assistant Manager – High Performance Computing Centre, NTU

More details coming up!
0930 – 1000
1000 – 1030
1030 – 1100 Tea Break
1100 – 1130
1100 – 1230

HP-Cast (cont.)

1100 – 1145
Unlocking New Capabilities
(cont.)

1145 – 1230
Unlocking New Capabilities and Measuring and Improving Performance
> Deploy trained neural networks from their training environment into real applications.
> Optimize DNN performance.
> Incorporate object detection into your DNNs.

1100 – 1130

New users: How to design a Spectrum Scale cluster

Speaker – Xiang Zhan (IBM)

1130 – 1215

New users: Orientation for New users

Speaker – Chris Maestas (IBM)

1215 – 1230

New users: Q&A


1100 – 1120

GPU monitoring and management with NVIDIA data center GPU manager (DCGM)

Dr Gabriel NoajeSenior Solutions Architect, NVIDIA APAC

1120 – 1150

PBS Works for HPC and AI

Subhasis Bhattacharya – Senior Director, Software Development

1150 – 1220

User Group Open Forum Discussion

An open-forum discussion to discuss and share challenges, ideas and best practices for PBS Works.

1220 – 1230

Closing Notes and Awards

Awards for a challenge on “An ideal system architecture for HPC and AI converged requirements”

More details coming up!
1130 – 1200
1200 – 1230
1230 – 1330 Lunch
1330 – 1400 1330 – 1530

HP-Cast (cont.)

 

1330 – 1445
Unlocking New Capabilities and Measuring and Improving Performance
(cont.)

1445 – 1530
Final Project
> Validate learnings by applying the deep learning application development workflow (load dataset, train, and deploy model) to a new problem.
> Learn how to set up your GPU-enabled environment to begin work on your own projects.
> Explore additional project ideas and resources to get started with NVIDIA AMI in the cloud, nvidia-docker, and the NVIDIA DIGITS container.

1330 – 1340

Welcome

Speaker – John Ramieri (IBM Software Defined Storage Executive)

1340 – 1410

Spectrum Scale V5 & ESS update

Speaker – Chris Maestas (IBM)

1410 – 1430

Spectrum Scale Erasure coding edition benchmark

Speaker – Antony Steel (Systemethix)

1430 – 1450

National Data Fabric on Spectrum Scale AFM

Speaker – Jake Carrol (University of Queensland)

1450 – 1520

Tiering data from Spectrum Scale to NAS

Speaker – Moonwalk

1330 – 1500

Manycore Architecture Lecture

Mr Gilles Gouaillardet, HPC Consultant, Department of HPC Support, RIST

1) Introduction

Why do we need manycore platforms?

2) Single-thread optimization

a) Vectorization

Topics: SIMD, Use of compiler report, Hierarchical memory, Data layout (AoS vs. SoA), Indirect access

b) Performance analysis (Roofline analysis)

Topics: Arithmetic intensity, Roofline analysis via Intel Advisor

c) Mixed precision

3) Parallelization

a) Performance scaling Topics: Strong scaling, Weak scaling

b) Parallel programming model: MPI+X

c) OpenMP Topics: Basic usage, Data locality (on cc-NUMA arch.), MPI+OpenMP binding

4) Fugaku architecture

1330 – 1530

Containers in HPC Tutorial

Dr Marco De La Pierre, Supercomputing Applications Specialist, Pawsey Supercomputing Centre

Mr Mark Gray, Head of Scientific Platforms, Pawsey Supercomputing Centre

 

• Overview of containers

• Basics of Singularity

• Mounting external filesystems

• Writable containers

• MPI enabled workflows

• Using GPUs

1400 – 1430
1430 – 1500
1500 – 1530
1530 – 1600 Tea Break
1600 – 1630

1600 – 1715
Final Project
(cont.)

1715 – 1730
Final Review
> Review key learnings and wrap up questions.
> Complete the assessment to earn a certificate.
> Take the workshop survey.

1550 – 1610

Storage for AI

Speaker – Frank Kraemer (IBM)

1610 – 1640

AI infrastructure solutions built for data centers

Speaker – Gabriel Noaje (Nvidia)

1640 – 1710

Network Offload For Storage Computing

Speaker – Ashrut Ambastha (Mellanox)

1710 – 1740

Manage massive unstructured data in Spectrum Scale

Speaker – Nilesh Bhosale (IBM)

1740 – 1800

Spectrum Scale troubleshooting

Speaker – Bing Zhu (IBM Support team)

1800

Get Together

1600 – 1800

• Building container images

• GUI applications (RStudio)

• Python workflows

• Other tools

• Best practices and tips

1630 – 1700
1700 – 1730
1730 – 1800

DO YOU HAVE A QUESTION ABOUT SCA20?

You can contact us at [email protected]

Contact Us