Home

BWUniCluster Instructions

You can find all application steps on this side: https://wiki.bwhpc.de/e/BwUniCluster_2.0_User_Access#Registration

1. You need to register with a short abstract on this side:
   https://entitlements.uni-mannheim.de/
   with the following text:
   "My field of research is Data integration using Deep Learning in the chair of the Data and Web Science Group at the University of Mannheim. I'm currently doing a group project and our supervisors are Christian Bizer and Ralph Peeters. The project will explore the benefits of using Deep Learning techniques (e.g. relational transformers) for data integration tasks such as identity resolution, schema matching (attribute and relation annotation), and slot filling (value prediction). I will experiment with different relational pre-training as well as fine-tuning techniques. The experiments will be conducted using Python and Huggingface Transformers."

2. Then you can go on with STEP B of the registration.

3. In STEP C you have to fill out a questionnaire as shown in the picture but in "bwHPC-Fachbereich" you select "Ingenieurwissenschaften" and in the field "numerische Methoden" you write "Transformer Modelle":

4. Now you have to wait until you get the positive feedback from the bwHPC.
5. Then you have to install MobaXTerm as shown here: https://wiki.bwhpc.de/e/BwUniCluster_2.0_User_Access#Login
6. For the usage you have to be in the VPN of the university. Here the instructions for the usage of the VPN: https://www.uni-mannheim.de/it/anleitungen/vpn/#c124270
7. Now you can start MobaXTerm and select "Session" then "SSH" and
   fill out the "Remote name" with "uc2.scc.kit.edu"
   "User-ID" is "ma_" and your uni id.
   Then you can connect with the Token and the service passwort you created in STEP B.