UL HPC School Program
All sessions will take place at the Kirchberg Campus, mainly in the Paul Feidert Room.
Important You are expected to bring your personal laptop for all sessions since there will be no workstation available on site.
All tutorials proposed as practical sessions are available on GitHub.
The detailed program is available here.
Agenda - Day 1 (May 6th, 2014)
| May 6th | By | Title | Location |
|---|---|---|---|
| 9h00-9H10 | S. Varrette | Opening | P. Feidert |
| 9h10-10h15 | S. Varrette | Keynote: HPC platforms @ UL: Overview and Usage | P. Feidert |
| 10h15-10h30 | Coffee break | P. Feidert | |
| 10h30-11h15 | S. Varrette | PS1: Getting Started (ssh, scp, rsync, oar, modules, builds) | P. Feidert |
| 11h15-12h30 | H. Cartiaux | PS2: HPC workflow with sequential jobs (C,python,java) | P. Feidert |
| 12h30-13h30 | LUNCH | ||
| 13h30-14h30 | S. Varrette | PS3: HPC workflow with MPI jobs Ex: OSU MB/HPL * | P. Feidert |
| 14h30-16h30 | S. Grimonet | PS4: Direct, Reverse and parallel Memory debugging with TotalView | P. Feidert |
| 16h30-17h30 | Roguewave Open discussion and Day 1 Closing | P. Feidert |
PS = Practical Session using your laptop
*this session have been rescheduled on May 7th
Agenda - Day 2 (May 7th, 2014)
| May 7th | By | Title | Location |
|---|---|---|---|
| 9h30-10h00 | S.Varrette | Follow-up Keynote: Last HPC challenges | P. Feidert |
| 10h00-10h45 | Keynote: Experience reports from top users | P. Feidert | |
| J. Berryman | HPC usage in the UL Soft Matter Theory Group | P. Feidert | |
| V. Satagopam | HPC@Uni.lu: a Bio-Core user experience | P. Feidert | |
| X. Besseron | XDEM Research on UL HPC platform | P. Feidert | |
| J.Muszynski | P2P Java experiments on the UL HPC Platform | P. Feidert | |
| F. Pinel | Uni.lu HPC usage | P. Feidert | |
| 10h45-11h00 | Coffee break | P. Feidert | |
| 11h00-11h55 | V.Plugaru | PS5: Using Matlab on the UL HPC Platform | P. Feidert |
| 11h55-12h30 | J. Emeras | PS6: Using R on the UL HPC Platform | P. Feidert |
| 12h30-13h30 | LUNCH | ||
| 13h30-14h30 | V.Plugaru | PS7: Bio-informatic softwares on the UL HPC Platform | P. Feidert |
| 14h30-16h00 | P. Wohlschlegel | PS8: Unified profiling and debugging with Allinea | P. Feidert |
| 16h00-16h30 | Allinea Open discussion | P. Feidert | |
| 16h30-17h00 | S.Varrette | (Rescheduled) PS3: HPC workflow with MPI jobs Ex: OSU MB/HP | P. Feidert |
| 17h00-17h10 | S.Varrette | Closing | P. Feidert |
Detailed Program for the practical sessions
Practical Session 1
Getting Started (ssh, scp, rsync, oar, modules, builds, screen), by: S. Varrette
This tutorial will guide you through your first steps on the UL HPC platform. we will cover the following topics:
- Platform access via SSH
- Overview of the working environment
- Usage of the web monitoring interfaces (monika, ganglia etc.)
- Reserving resources with OAR and Job management
- Using modules
- Advanced job management and Persistent Terminal Sessions using GNU Screen.
- illustration on Linux kernel compilation
Practical Session 2
HPC workflow with sequential jobs (C,python,java), by H. Cartiaux
For many users, the typical usage of the HPC facility is to execute 1 (sequential) program with many parameters. On your local machine, you can just start your program 100 times sequentially. However, you will obtain better results if you are able to parallelize the executions, either across many thread and/or many nodes of the HPC platform
During this session, we will cover 3 use cases:
- Use the serial launcher (1 node, in sequential and parallel mode) over a C program.
- Use the generic launcher, distribute your executions on several nodes over a python program.
- A more advanced usage over a complex Java program JCell, a Java library design to work with cellular genetic algorithms (cGAs)
Practical Session 3
HPC workflow with MPI jobs. Ex: OSU MB/HP, by S. Varrette
Online instructions for OSU Microbenchmarks and for HPL
The objective of this session is to compile and run MPI programs over various reference MPI implementations (Intel MPI, OpenMPI or MVAPICH2, all available on the cluster). In practice, this session will focus on the following MPI applications:
- the OSU Micro-benchmarks which permit to measure the performance of an MPI implementation
- the High Performance Linpack (HPL) which is used to rank supercomputers across the world for the Top500 list
Practical Session 4
Direct, Reverse and parallel Memory debugging with TotalView, by S. Grimonet
The objective of this session is to get a brief overview of the TotalView, a GUI-based source code defect analysis tool that gives you unprecedented control over processes and thread execution and visibility into program state and variables. The following topics will be covered:
- Startup and Overview
- IHM Navigation and process control
- Action points (TP)
- Examination and Data Analysis (TP)
- Debugging Parallel Applications (TP)
- Memory reports using MemoryScape (TP)
- Remote Debugging
- CUDA Debugging
- Xeon Phi Debugging
- Memory Debugging with MemoryScape
- Events and memory errors (TP)
- Delayed Scripted debugging (non-interactive) (TP)
- Reverse debugging using ReplayEngine (TP)
- Asynchronous control of Parallel Applications (TP)
- Type Transformation (TP)
Practical Session 5
Using Matlab on the UL HPC Platform, by V. Plugaru
The objective of this session is to exemplify the execution of MATLAB - a high-level language and interactive environment for numerical computation, visualization and programming, on top of the UL HPC platform. The following topics will be covered:
- how to run MATLAB in interactive mode, with either the full graphical interface or the text-mode interface
- how to run MATLAB in passive (batch) mode, enabling unattended execution on the clusters
- how to use MATLAB script (.m) files
- how to plot data, saving the plots to file
- how to take advantage of some of the parallelization capabilities of MATLAB to speed up your tasks
Practical Session 6
Using R on the UL HPC Platform, by J. Emeras
Through this session you will learn how to use R from your local machine or from one of the resources of the UL HPC platform clusters. We will also use the ggplot library to generate nice graphics and export them as pdf files. Then, we will see how to organize and group data. Finally we will illustrate how R can benefit from multicore and cluster parallelization.
Practical Session 7
Bio-informatic softwares on the UL HPC Platform, by V. Plugaru
The objective of this session is to exemplify the execution of several Bioinformatics packages on top of the UL HPC platform. The targeted applications are:
In particular, the following topics will be covered:
- how to load and run pre-configured versions of these applications on the clusters
- how to download and use updated versions of Bowtie2/TopHat
- discuss the parallelization capabilities of these applications
Practical Session 8
Unified profiling and debugging with Allinea, by P. Wohlschlegel
Developing an HPC application designed for MPI or hybrid environments can be a very challenging task - especially when it comes to fixing bugs, optimizing workload or even resolving both type of issues simultaneously. Those challenges are made easier with Allinea Tools. Using Allinea environment, it is now possible for developers to adopt instantly efficient and scalable development tools and to focus immediately on their core activity : science.
Allinea tools are available for you in University of Luxembourg, and this workshop has been arranged to help you get started with Allinea MAP and Allinea DDT. During this workshop, we will review the capabilities of Allinea tools and see, through hands-on exercises, how this environment can help reduce your time to results to the minimum.