DiRAC-3 Launch Day

Programme

Welcome: Jeremy Yates, Chair of DiRAC Technical Directorate

Overview of the new HPC Systems and hardware across all services: Mark Wilkinson, Director

Overview of User Guide: Anushka Sharma, Senior Technical Programme Coordinator

Overview of DiRAC Training: Richard Regan, Training Manager

TauREx on DiRAC-3: Dr. Ahmed Faris Al-Refaie, University College London

A Lattice Field Theory Ecosystem for DiRAC-3: L Del Debbio, University of Edinburgh

Large cosmological runs on the new DiRAC MI “Cosma-8” facility: Matthieu Schaller, Lorentz Institute, on behalf of the Virgo Consortium

The Technical Working Group experience of deploying DiRAC3: Alastair Basden, Technical Manager

The Research Software Engineer Experience: James Richings, Edinburgh and Athena Elafrou, Cambridge

Closing remarks: Simon Hands, DiRAC Community Development Director


Welcome: Jeremy Yates, Chair of DiRAC Technical Directorate


Overview of the new HPC Systems and hardware across all services: Mark Wilkinson, Director

Overview of User Guide: Anushka Sharma, Senior Technical Programme Coordinator

Overview of DiRAC Training: Richard Regan, Training Manager

TauREx on DiRAC-3: Dr. Ahmed Faris Al-Refaie, University College London

A Lattice Field Theory Ecosystem for DiRAC-3: L Del Debbio, University of Edinburgh

Large cosmological runs on the new DiRAC MI “Cosma-8” facility: Matthieu Schaller, Lorentz Institute, on behalf of the Virgo Consortium

The Technical Working Group experience of deploying DiRAC3: Alastair Basden, Technical Manager

The Research Software Engineer Experience: James Richings, Edinburgh and Athena Elafrou, Cambridge

Closing remarks: Simon Hands, DiRAC Community Development Director

Course Testimonials

Machine Learning Techniques for Science – November 2021

“I found the course to be the ideal mix between theory and practical examples. Very informative on how to use ML for cutting-edge science research questions.”

“This course was the perfect introduction to machine learning from a scientific perspective; instructors were enthusiastic and very helpful, and the lectures and materials were approachable for someone with no previous experience, while remaining detailed enough to be of practical use. I felt I gained a good overview of a range of important techniques and algorithms and will certainly apply the knowledge gained during my chemistry PhD.”

AMD Induction Training (AIT)

13th October 2021 (Registration is closed)

Introduction

This is a hands on workshop where participants will use code snippets to illustrate the best approach to accessing the power of  our new AMD ROME CPUs.

The workshop will cover:

  • The ROME Microarchitecture & Memory channels
  • Compilation, & optimization
  • Numa regions & Pinning
  • Maths and Scientific Libraries
  • uProf a new profiler

Format

The virtual workshop will be run on multiple DiRAC sites so you will be able to experience exactly what you need to do on the system you use.

  • Support will be there from AMD and the local technical support teams.
  • A shared slack channel will be available to ask questions, highlight any issues and share good practice between sites.

Target Audience

The target audience is researchers:

  • Who are interested in building and running their code on our new DiRAC-3 AMD systems.
  • Who want to run their code efficiently to get the best performance.
  • Who want to take advantage of the new AMD features and tools.

Requirements

Only basic experience of a language is required C/C++, Fortran, or Python.


Registration

Closing date is the 6th of October. Now closed

N Ways to GPU Programming

30th Sep 2021, full day course, TBA

Learning Objectives

With the release of NVIDIA CUDA in 2007, different approaches to GPU programming have evolved. Each approach has its own advantages and disadvantages. By the end of this bootcamp session, participants will have a broader perspective on GPU programming approaches to help them select a programming model that better fits their application’s needs and constraints. The bootcamp will teach how to accelerate a real-world scientific application using the following methods:

  • Standard: C++ stdpar, Fortran Do-Concurrent
  • Directives: OpenACC, OpenMP
  • Programming Language Extension: CUDA C, CUDA Fortran.

Bootcamp Outline

During this lab, we will be working on porting mini applications in Molecular Simulation (MD) domain to GPUs. You can choose to work with either version of this application.

Bootcamp Duration

The lab material will be presented in an 8-hour session. A Link to the material is available for download at the end of the lab.

Content Level

Beginner, Intermediate

Target Audience and Prerequisites

The target audience for this lab are researchers/graduate students and developers who are interested in learning about various ways of GPU programming to accelerate their scientific applications.

Basic experience with C/C++ or Fortran programming is needed. No GPU programming knowledge is required.

Registration

Registration is closed.

EPFL OpenACC Bootcamp

Overview

Learn how to write a portable parallel program that can run on multicore CPUs and accelerators like GPUs and how to apply incremental parallelization strategies using the OpenACC programming model to accelerate an example application that simulates heat distribution across a 2-dimensional metal plate and applying this knowledge to accelerate a mini-application.

Event Format 

The GPU Bootcamp will be hosted online in the Central European Summer Time zone. All communication will be done through Zoom, Slack and email. 

Prerequisites

Basic experience with C/C++ or Fortran. No GPU programming knowledge is required.

This event has limited capacity, so please make sure that prerequisites are met before applying. You will be receiving an acceptance email with details on how to participate by September 2nd, 2021.  

Compute Resources

Attendees will be given access to a GPU cluster for the duration of the GPU Bootcamp.


Agenda

Day 1: September 9, 2021 (9:00 AM to 5:00 PM CEST)

  • Welcome: 9:00 AM 
  • Connecting to a cluster:  9:00 AM – 09:15 AM 
  • Introduction to GPU programming: 9:15 AM – 9:30 AM CAT(Lecture)
    • What is a GPU and Why Should You care?
    • What is GPU Programming?
    • Available Libraries, Programming Models, Platforms.
  • Introduction to OpenACC: 9:15 AM – 10:00 AM (Lecture + Lab)
    • What is OpenACC and Why Should You Care?
    • Profile-driven Development.
    • First Steps with OpenACC.
    • Lab 1.
  • OpenACC Data Management: 10:00 AM – 10:45 AM (Lecture + Lab) 
    • CPU and GPU Memories.
    • CUDA Unified (Managed) Memory.
    • OpenACC Data Management.
    • Lab 2.
  • Break 11:00-11:15
  • Gangs, Workers, and Vectors Demystified: 11:30 AM – 12:15 PM (Lecture + Lab)
    • GPU Profiling.
    • Loop Optimizations.
    • Lab 3.
  • Mini-application challenge (12:15-12:30)
    • Overview of the mini-application 
    • Review steps to acceleration 
  • Application challenge  (13:00-17:30)

Day 2: September 10, 2021  (9:00 AM to 12:30 PM CEST)

Welcome (Moderator): 9:00

Mini-application Solution Walk-through (9:15-9:30)

Introduction to NVIDIA ® Nsight™ Tools (9:30-10:00)

  • Overview of Nsight Tools
  • How to profile a serial application with NVIDIA Tools Extension (NVTX) 
  • Overview of optimization cycle with Nsight Systems

Profiling mini-application (10:00-12:30)

  • Profile a sequential weather modeling application (integrated with NVTX APIs) with NVIDIA Nsight Systems to capture and trace CPU events and time ranges
  • Understand how to use NVIDIA Nsight Systems profiler’s report to detect hotspots and apply OpenACC compute constructs to the serial application to parallelise it on the GPU
  • Learn how to use Nsight Systems to identify issues such as underutilized GPU device and unnecessary data movements in the application and to apply optimization strategies steps by steps to expose more parallelism and utilize computer’s CPU and GPU

Wrap-up (12:25-12:30)

Director’s Time

Call for DiRAC Director’s Discretionary proposals

Application deadline 23rd November 2020.

The DiRAC Director will make a number of large discretionary awards of time on two DiRAC services during the period 1st December 2020 – 31st March 2021. In total, approximately 32m core hours are available in this call.

In addition, Research Software Engineering (RSE) projects of up to 3 months’ duration will be considered.

Scope of call:

DiRAC will be able to make a number of large discretionary awards of time on two of our services during the period August 2020-March 2021. Proposals should be ambitious in scope – these awards will be made to proposals of outstanding scientific interest and timeliness and which can demonstrate significant benefit from concurrent access to large numbers of nodes.

All codes used must have undergone a technical review as part of the RAC process. Codes which have not done cannot be considered in this call.

All awarded time must be used by March 31st 2021 and no requests to extend beyond this date will be accepted.

Application process:

Proposals should be submitted using the Discretionary Call application form or the RSE Support application form. The forms are available below:

You must include the application number where the required technical case associated with the software to be used in this application can be obtained. You do not need to resubmit technical information that you have already provided however, you may be asked to provide evidence of the scalability of your codes if you will be using the code at a larger scale than was previously requested.

Assessment process:

Proposals will be reviewed by the Director, the RAC chairs and members of the RAC panels.

Assessment criteria:

Proposals will be assessed according the following criteria:

    1. Scientific impact – 60%
    2. Track record of user/group – 20%
    3. Timeliness/urgency – 20%

Closing date:

Applications should be submitted by e-mail to DiscretionaryCall@leicester.ac.uk by 5pm on Monday, 23rd November 2020.


Seedcorn Time

Access to DiRAC is co-ordinated by The STFC’s DiRAC Resource Allocation Committee, which puts out an annual Call for Proposals to request time for a Full Project Allocation and also operates a Seedcorn Time programme.

Seedcorn Time Programme

For researchers who would like to try the DiRAC resources, get a feel for HPC, test codes, benchmark or see what the DiRAC resources can do for you before making a full application for resources, an application can be made for seedcorn time.

Existing users may also apply for seedcorn allocations to enable code development/testing on a service which is not currently part of their project allocation.

You can apply for seedcorn time at any time.

Seedcorn Application

The maximum allocation of seedcorn time will be 100,000 x86 core hours or 1,000 GPU hours or 1,000 KNL node hours. Your hours can be used at any time within three months from the date it is allocated. If you require a larger seedcorn allocation in order to carry out the necessary testing of your code, please contact the DiRAC Director to discuss your requirements.

An overview of the DiRAC hardware resources can be found here.

If you would like to apply please fill in this simple DiRAC Seedcorn application form and send it to dirac-support@epcc.ed.ac.uk

You will need to provide your personal details, the codes that you intend to run and what libraries and tools are required for you to run the code, if known. We also would like to know what you intend to achieve with your allocation and of course which of the four DiRAC systems you would like to use.

You should be able to complete the form in around 15 minutes.

The decision process for Seedcorn applications is streamlined and should take no more that 1 week.