Summer School on Effective HPC for Climate and Weather

Making effective use of HPC environments becomes increasingly challenging for PhD students and young researchers. As their primary intent is to generate insight, they often struggle with the technical nature of the tools and environments that enable their computer-aided research: computation, integration, and analysis of relevant data.

The scope of the summer school is the training of young researchers and software engineers in methods, tools, and theoretical knowledge to make effective use of HPC environments and generate insights.

Date 23-28 August 2020
Venue University of Reading, Whiteknights Campus, Reading, UK, RG6 6UR
Contact Julian Kunkel (University of Reading, UK)

While the school aims to prepare the attendees for large scale simulation runs and data processing, it does also cover a representative selection of modern concepts such as machine learning, domain-specific languages, containerisation, and analysis of climate/weather data using Python 1).

We will also provide an outlook of challenges and strategies for HPC for climate and weather. Additionally, we aim to foster networking among scientists bringing together users of specific models and tools and enabling them to exchange their knowledge.

A certificate of attendance will be provided to all attendees. Additionally, we offer the opportunity to obtain the currently available certificate of the HPC Certification Program.

The summer school will also support the mission of the European Network for Earth System modelling (ENES).

The ESiWACE project funds this summer school.

The ESiWACE Summer School is structured along with topical sessions in the morning/afternoon followed by an Academic Group Projects Session.

A topical session typically consists of an academic lecture and it may contain hands-on/lab practicals, group work and discussion. Experts in the respective field will organise each of these sessions.

During the first Academic Group Projects Session (Monday), groups of 4-5 people will be assigned considering the topics suggested in the applications. Over the week, the Academic Group Projects Sessions will provide you with the opportunity to develop the projects in teams with joint interests. The project teams will then present their results to their peers on the last day of the summer school (Friday).

Academic Group Projects

The Academic Group Projects will be organised as follows:

  • With your application, you (optionally) submit a tentative idea of a project you want to conduct. The project must be related to summer school topics.
  • We will collect the project ideas, gather them appropriately and prepare suggestions for the group's formation. This compilation will be shared with attendees and speakers one month before summer school.
  • On the first day, during the Academic Group Projects Session, the group members will define one project (or multiple small ones) to pursue.
  • It is your choice to engage in a predefined project or group. You may also adjust the project idea(s) and develop a project by yourself. However, we strongly recommend you team up to benefit from others' expertise.
  • During the Academic Group Projects Session (90 minutes at the end of the day), you will work on your project.
  • On Friday, you will present your outcome as a group (can use few slides, a live demonstration is preferred).

Topics

The topics covered in the summer school are as follows:

  • Big Data Analytics
  • Code Development
  • Extreme-Scale Computation
  • Machine Learning
  • Middleware and File Formats
  • Modern Storage
  • Post-Processing and Visualisation
  • Software Engineering
  • Supporting Tools
  • Workflows

Agenda

The following shows the general structure of the tentative agenda.

Sunday (Optional Attendence)

  • 17:00 Registration opens and coffee
  • 18:00 Get together

Monday - Computing

  • 08:30 Registration opens and coffee
  • 09:00 Extreme-Scale Computation – Chair: Rupert Ford (STFC, UK), Carlos Osuna (MeteoSwiss, Switzerland)
    Learning objectives: Domain-Specific Languages, Workflows, Cylc
    • 09:00 Morning Academic Session
    • 10:30 Refreshment Break
    • 11:00 Lab Session
  • 12:30 Lunch Break
  • 13:30 Code Development – Chair: Sophie Valcke (Cerfacs, France), Christopher Maynard (University of Reading, UK)
    Learning objectives: Code Coupling, OASIS3-MCT, Software Engineering, Agile Programming
    • 13:30 Afternoon Academic Session
    • 15:00 Refreshment Break
    • 15:30 Lab Session
    • 16:30 Breakout Session
  • 17:00 Academic Group Projects – Hands-on
  • 18:00 Cocktail – Kick-off of Academic Group Projects

Tuesday - Storage

  • 09:00 Modern Storage – Chair: Sai Narasimhamurthy (Seagate, UK), Jean-Thomas Acquaviva (DDN, France)
    Learning objectives:
    • Posix versus MPI-IO versus hdf5 versus other high-level io libraries (e.g. Sionlib)
    • Darshan for I/O patterns identification
    • Benchmarking tools and tricks
    • Architectural considerations of modern object stores suitable for extreme-scale computing
    • Students will learn about the fundamental design aspects of such object-based storage systems
  • 09:00 Morning Academic Session
    • 10:30 Refreshment Break
    • 11:00 Lab Session
  • 12:30 Lunch Break
  • 13:30 Middleware and File Formats – Chair: Julian Kunkel (University of Reading, UK), Luciana Pedro (University of Reading, UK)
    Learning objectives:
    • Describe the general layers involved in I/O on a supercomputer
    • Design a data model for NetCDF/CF
    • Analyse the implications of parallel I/O on application efficiency
    • Identify typical I/O performance issues and their causes
    • (we will include architectures if not part of the session before)
  • 13:30 Afternoon Academic Session
    • 15:00 Refreshment Break
    • 15:30 Lab Session
    • 16:30 Breakout Session
  • 17:00 Academic Group Projects – Hands-on
  • 18:30 Break
  • 20:00 Networking Dinner

Wednesday - Data Analytics

The morning session takes place at ECMWF.

  • 09:30 Machine Learning – Chair: Peter Dueben (ECMWF, UK)
    Learning objectives:
    • Provide an introduction to machine learning
    • Provide an overview of how machine learning can be used to improve weather and climate predictions
    • Present examples for the use of machine learning at ECMWF
    • Discuss the challenges for machine learning in weather and climate science.
  • 09:30 Morning Academic Session
    • 10:30 Refreshment Break
    • 11:00 Visit ECMWF
  • 12:30 Lunch Break
  • 13:30 High-Performance Data Analytics and Visualisation – Chair: Sandro Fiore (CMCC, Italy), Niklas Röber (DKRZ, Germany)
    Learning objectives:
    • Challenges of big data over HPC
    • Introduction to some HPDA tools
    • Real use cases on HPDA
  • 13:30 Afternoon Academic Session
    • 15:00 Refreshment Break
    • 15:30 Lab Session: A hands-on to extract and visualise traces.
    • 16:30 Breakout Session
  • 17:00 Academic Group Projects – Hands-on

Thursday - Supporting Tools

  • 09:00 Performance Analysis – Chair: Kim Seradell, Mario Acosta (BSC, Spain)
    Learning objectives:
    • Learn about performance analysis fundamentals (objectives, methods, metrics, hardware counters, etc.)
    • Provide specific knowledge about performance analysis applied to earth system modelling
    • Describe BSC tools suite (Extrae, Paraver, Dimemas)
    • Present uses cases from Earth System Models (IFS, NEMO, etc.) to show how to identify and solve performance issues
  • 09:00 Morning Academic Session
    • 10:30 Refreshment Break
    • 11:00 Lab Session
  • 12:30 Lunch Break
  • 13:30 Containers – Chair: Alberto Madonna (ETH Zürich, Switzerland), Simon Wilson (NCAS, UK)
    Learning objectives:
    • General concepts about containerisation
    • Advantages of using containers for scientific applications
    • Introduction to Docker: build and run containers on personal computers
    • Introduction to containers on HPC systems with the Sarus container engine
  • 13:30 Afternoon Academic Session
    • 15:00 Refreshment Break
    • 15:30 Lab Session
  • 17:00 Academic Group Projects – Hands-on

Friday - Group Presentation

  • 09:00 Group Presentation Preparation
    • Additional time to prepare the presentations and test the types of required equipment.
  • 10:00 Group Presentation
    • 10:00 Groups 1 to 4
    • 11:00 Refreshment Break
    • 11:30 Groups 5 to 8
  • 12:30 Closing Remarks and Farewall
  • 13:00 Lunch Break (Optional)
  • 14:00 End of the Summer School Programme

We will provide a document with detailed information about the local organisation and spare-time activities for the attendees timely before the school.

Are you planning to attend? The ESiWACE2 project partially funds the summer school. We require attendees to register and optionally apply for subsidies.

Use the following form to register. It also provides means to indicate the subsidy required for your attendance if submitted before the deadline.

If you have any question regarding registration, send an email to Luciana Pedro l.r.pedro@reading.ac.uk

Important Dates

  • 24 January 2020 – Applications for attending the summer school open (with optional subsidy)
  • 28 February 2020 – Applications for the summer school close for applications requesting subsidy
  • 13 March 2020 – Applicants receive notification of the decision on subsidy and acceptance
  • 12 May 2020 – General applications for attending the summer school close
  • 26 May 2020 – All applicants receive information of the decision on acceptance
  • 23-28 August 2020 – Summer School on Effective HPC for Climate and Weather
  • Mid-September – Selected presentations will be made available, and certificates will be sent to all attendees

The ESiWACE2 project will subsidise the attendees and provide waived conference fees.

If you do not receive a subsidy, the fee structure for the event is as follows:

Full Package Registration £800
Minimal Registration £350

The Full Package Registration covers:

  • Attendance to all sessions and social events
  • Accommodation on university premises (Check Section Accommodation)
  • All-inclusive meals (breakfast and three-course self-service dinner with tea and coffee)
  • Transport between London Heathrow Airport and the venue (by bus)

The Minimal Registration covers:

  • Attendance to all sessions and social events
  • All-inclusive meals

While we recommend that you choose the Full Package Registration, you have the option to select the Minimal Registration and organise accommodation and transport by yourself.

For information about subsidy, check Section Funding of Attendees.

Travel Grants

A committee will conduct the selection procedure for subsidising attendees, which will be funded according to their financial situation, the support provided by their home institution, and other possible fundings. We appreciate partial financing.

During the selection procedure, the applicants that request subsidy will be classified into one of the following categories:

  • Travel Subsidy: The Full Registration fee for the summer school will be covered. In addition, attendees will receive the reimbursement of their travel costs to Reading after the event. The receipts must be provided for the payment to be made in full. Note that we have only a few travel subsidy grants, so please attempt to support your funding.
  • Attendance Subsidy: The Full Registration fee for the summer school will be included.
  • Registration Subsidy: The Minimal Registration fee for the summer school will be included.
  • No Subsidy: Attendees will have to pay for the total cost of the summer school.

In all the cases in which a subsidy is granted, a fee prepayment (also described as a deposit) with a value of £100 will be charged to secure the participant's place on the event 2). Please, note that none of the subsidy categories covers personal expenses.

In addition to their (application process), applicants who want to be considered for subsidies will have to submit:

  • A one-page document describing:
    • the justification for the subsidisation 3),
    • the requested level of subsidy, and
    • a tentative cost estimate (for travel subsidy).

Applicants will receive an answer from the committee four months before the event.

The selection procedure for applicants will be conducted by a committee that oversees the registration to make sure we balance the attendance across countries and gender and support those in need. The committee will prioritise diversity and internationality.

Applicants will have to submit:

  1. Up to one-page motivational letter including:
    • a tentative idea of a project that can be conducted as part of the Academic Group Projects, including no more than five keywords, and
    • how you will act as a multiplier of the gathered information.
  2. Up to two-page CV showing that the applicant satisfies the description of the target audience including a reference to one paper/thesis/dissertation/project in a related area in which the applicant is the author (or one of the authors).

Considering that the main idea behind the summer school is spreading concepts on effective HPC for climate and weather, typically only one applicant will be selected per university/company. Applicants that do not require funding from the summer school will have priority to ensure that more people will benefit from the proposal.

The selected applicants should bring their personal computers. Please, indicate the availability of a laptop to the organisers with your application.

Target Audience

The target audience for the summer school is Earth system scientists, including PhD students and young researchers, and software engineers in the domain. While each of our topics will be introduced, the attendees should have a basic understanding of:

  • Python
  • Linux
  • The general computational aspects of a climate/weather model

For attendees without prior experience, the following links provide some references to cover significant aspects of the contents mentioned above.

Organisers

Programme Committee

  • Julian Kunkel (University of Reading)
  • Sophie Valcke (Cerfacs)
  • Luciana Pedro (University of Reading)
  • Bryan Lawrence (University of Reading)

The University of Reading is situated in the Thames Valley leading to London. The main campus, Whiteknights, is the intended venue for the summer school. Download the Campus Map.

The campus has won seven consecutive Green Flag awards, recognising it as one of Britain's top green spaces. It is set in 130 hectares of parkland and is two miles from Reading station and the beautiful shopping district in Reading city centre (downtown).

Accommodation

We provide accommodation in the En-suite self-catered rooms (Premium en-suite rooms) as part of the Full Package Registration.

For attendees with Minimal Registration, a viable option is the Cedars Hotel at the Campus. Alternatively, the city centre offers more than 30 hotels nearby where several renowned companies provide cosy accommodation.

Transport

There is a regular direct bus connection to welcome international guests (RailAir) between the Reading Main Station and the London Heathrow Airport.

The public transport system with buses 4) is fully developed in Reading, particularly around the Whiteknights campus and Reading city centre. Buses between the central railway station and the campus are typically scheduled on a regular 7-minute interval and, due to the fair fares, enable quick connection for participants. All buses offer free WiFi and many USB charging points on board.

The Centre of Excellence in Simulation of Weather and Climate in Europe (ESiWACE) addresses world challenges pushing the limits of science. It benefits the broader community by providing services and training opportunities.

As part of the ESiWACE2 project, we are organising this summer school to bridge the gap between scientists and computational science and increase the effectiveness of young scientists. The main goal of this event is the training of representative scientists from different institutions with state-of-the-art concepts tailored to the domain, but that also stretches beyond climate and weather, allowing them to act as a multiplier and increase productivity overall.

ESiWACE is funded by the European Union's Horizon 2020 research and innovation programme under grant agreement No. 823988.

1)
A detailed list of topics can be found in the programme.
2)
This value will be fully refunded when the participant arrives.
3)
Why do you not have any other funding source? Why can't your institution pay for it?