AI_INFN Technical Meeting

Monday 17 Jun 2024, 16:00 → 17:10 Europe/Rome

Virtual meeting room (zoom): https://l.infn.it/ai-infn-meeting

Date: 2024-06-17

Tracked developments:

Automation of RKE2 deployments in INFN Cloud

• NTR

Develop monitoring and accounting infrastructure (R. Petrini)

• [Open issue since two weeks ago] There is a problem with the migration to the database developed by Nadir: when writing “TYPES” the SQL console hangs forever. We rolled back to the previous database, and we will give it another try as soon as possible.

Environment setup (M. Barbetti, S. Giagu, S. Bordoni, L. Cappelli)

• Environment quantum, fino alla settimana scorsa gli environment in questione erano due.
  • Uno su cuda 11.8 in cui coesistevano TF, Jax e Torch, ma non adava Pennylane
  • Jax ha modificato le sue release che impedisce di far coesistere Jax e Pytorch (hanno tolto la release che usa cudnn)
  • Si passa a cuda 12, abbiamo due environment basati in cui sembrano funzionare i Jupyter
  • Pennylane è risolto: PyTorch+Jax (ma senza TensorFlow).
• Numpy 2.0.0 broke everything…
  • Questo rompe la maggior parte degli environment conda dove non è stato imposto a priori numpy<2.00
• We agree that we should move forward with two different environments, one with Torch and Quantum and one with TensorFlow and quantum.

Offloading tests with virtual kubelets (G. Bianchini, D. Ciangottini)

• CloudVeneto is in downtime this week.

Acquisto FPGA

• NTR

Status legend

Active
Priority
Problems
Postponed or Blocked by others
Completed

16:00 → 16:15 News and setup

Speaker: Lucio Anderlini (Istituto Nazionale di Fisica Nucleare)

16:15 → 16:35 Discussion: Take-home messages from the first user's forum

16:35 → 16:50 Discussion on tasks and priorities

Speaker: All

16:50 → 17:00 Any other business