Book your place!
This training is organized by the PUNCH4NFDI Young Academy (PYA) and intended for scientists and researchers interested in using the infrastructures, services and tools developed by PUNCH4NFDI for PUNCH communities.
Relevant Information for Participants
Join Zoom Meeting (details will be shared few days before the event)
Meeting ID:
Passcode:Please 🔗 join this Slack Workspace that will be used during the training for announcements, questions and troubleshooting. Use the #preparation channel if you incur any troubles during installation and setup.
The PUNCH Young Academy PYA training on
JupyterHub service*,
BAT.jl tool*,
Compute4PUNCH and Storage4PUNCH resources usage*,
and REANA platform**, will take place on 26 and 27 November at Rheinische Friedrich-Wilhelms-Universität Bonn.
Important:
- (*) The tutorial will be offered in-person at Lecture Hall ROT/Uni Bonn, a remote participation to the tutorial would be based on best effort, only via a chat channel in Zoom or Slack.
For eaching the building, please check "How to reach".- (**) The tutorial will be exclusively offered in virtual form, registrants located in Bonn will have the possibility to join the tutorial from the Lecture Hall ROT via Zoom.
Over two days, the trainees will learn about a number of tools and services that are developed and/or widely used by the PUNCH communities. The selection of these tools has be performed to meet the needs and expectations of the scientists in PUNCH4NFDI consortium and beyond. The tools will be explored in a thrilling alternation of explanation, discussion and practical sessions.
JupyterHub service:
Jupyter notebooks are the state-of-the-art in interactive data analysis environments. University of Münster provides a JupyterHub service to PUNCH members with an assortment of preinstalled scientific software packages available through the notebook interface.
In this session the trainee will learn how to access the service, which will be used for the training on the BAT.jl tool. We will quickly go over the available software and demonstrate some useful techniques and workflows.
BAT.jl tool:
BAT.jl is a high performance Bayesian Ananlysis Toolkit implemented in Julia, a general-purpose programming language. Applications are parameter inference given a model, the comparison of different models in the light of a given data set and the test of the validity of a model. More information about the tool capabilities and applications are available at BAT.jl documentation.
Compute4PUNCH and Storage4PUNCH infrastructures:
The federated compute and storage infrastructures made available to the PUNCH4NFDI consortium, Compute4PUNCH and Storage4PUNCH, will be presented. These intrastructures comprise a variety of heterogeneous compute and storage systems and are managed by an HTCondor overlay batch system and COBalD/TARDIS metaschedulers. The analysis workflow based on the use of containers, the job submission process and the way to access the storage resources will be presented, followed by a hands on session dedicated to the submission and handling of jobs on the Compute4PUNCH infrastructure. Documentation about Compute4PUNCH and Storage4PUNCH is available in the sections TA2/WP1 and TA2/WP2 of the PUNCH4NFDI internal pages respectively. The installation of the oidc-agent on your personal laptop is required to participate to the session. Documentation about how to set up the oidc-agent is available here.
REANA Platform:
REANA is a central tool in PUNCH development plan, as it can support several container technologies, workflow engines, shared storage systems, and compute cloud infrastructures used by the community.
REANA is a REusable ANAlysis platform allowing scientists to run containerized data analysis pipelines on remote compute clouds. This makes it easy to structure the analysis in reproducible way, enabling future reuse.
REANA training is divided into 2 parts: a short introduction, where the trainees will learn "What is REANA?"; then, we will dive in with Handouts: "Moving Your Own Code to REANA Workflow". At the end, the trainees will be able to master the foundation commands of REANA and run different analyses on the platform, both thanks to the exercises provided and through testing their own code.
Requirements for the training:
-
Common pre-requisites: In order to actively participate into the 3 parts of the training, you should bring your personal laptop, or have access to a workstation with Linux or Mac environment. Web browser.
-
Compute4PUNCH and Storage4PUNCH and REANA trainings require the registration of the applicant to the PUNCH-AAI service. Applicants who are not PUNCH members should use the instruction provided here to request a PUNCH-AAI "trainee" account. The same email used for the training registration form has to be used to register to PUNCH-AAI. Basic knowledge of Bash/Unix shell commands is required.
-
JupyterHub and BAT.jl trainings require access to the JupyterHub service at the University of Muenster. This is in principle granted if you have an active account at an institution member of DFN/eduGAIN federation. Check your access to the JupyterHub service @WWU Muenster by logging to the Uni-Muenster-JupyterHub. If successful, you should see a message that you need a resource token to proceed further. To get support about your login to the portal, contact JupyterHub-Support.
-
REANA training requires installing the platform before the training start. Follow the instruction provided here to install REANA.
Please check carefully the fulfillment of the requirements and do not hesitate to get back to the organizers to get further information and support during the registration period.
More information and material will be uploaded to the training gitlab repository, accessible through the link: PYA-training-material-GitLab on PUNCH GitLab.
Venue: The event takes place at the Faculty of Physics at Bonn University. It can easily be reached by buses 601, 602, 603 from Bonn train main station (4 stops, 5min), get off at Bonn Am Botanischen Garten.
