# AiiDAlab Meeting Notes 2024-06-06 ###### tags: `Meeting` :::info - **Location:** https://psich.zoom.us/j/69116696576 *(Please request password on slack)* - **Date:** 06-06, 2024 14h (CET) - **Agenda** - [25 min] Go through - [Urgents of Roadmap](https://github.com/orgs/aiidalab/projects/15/views/2) (click me) - [priority panel of Roadmap](https://github.com/orgs/aiidalab/projects/15) (click me) - [5 min] Prepare session: go to ["hackmd notes"](https://hackmd.io/team/aiidalab?nav=overview) (click me) for general issues to discuss. - [~30 min] Hackmd session: The catch up round for everyone and bring general topics about AiiDAlab. Buffer for discussion: Continue of hackmd session if not finished and focus on one or two general discussion. - main purpose is to make sure all participants speak and give update. - **Participants:** Before the meeting, the host should check and update the list below. - Jason [PSI] (jy) - Sasha [Empa] (ay) - Xing [PSI] (xw) - Edan [PSI] (eb) - Giovanni [PSI] (gp) - Marnik [PSI] (mb) - Carlo [Empa] (cp) - Daniel [ISPG] (dh) - Andres [Empa] (ao) - Miki [PSI] (mib) - **Host:** jy ::: [TOC] ### Catch-up round *Max. 3 minutes each* Edan: * Sorry to cut in line 😬 - need to work on slides for tomorrow * Nothing on AiiDAlab - too many deliverables 😮‍💨 * MARVEL Junior Seminar talk tomorrow * Showing short versions of demos * Permission to plaster your faces on some slide? 🙏 * Will work on AiiDAlab starting week of 17th Jason: * Proof of concept on hyperqueue as scheduler for container: * Hardcoded 4 fake CPUs for a container instance, which means maximum 4 `pw.x` process can run at the same time. (TODO) This requires more test on performance, since probabaly MPI is not happy with such over-subscribing. * Implemented now in aiidalab-docker-stack, will do it for QeApp image for test purpose. Would be good if Daniel can test on the https://github.com/aiidalab/aiidalab-docker-stack/pull/463 image using hyperqueue and compare with slurm whether all your use cases are covered. * TODO: benchmarking on the open of the AiiDAlab page when calculations are running. * I make aiida-hyperqueue able to set number of CPUs and RAM by `num_cpus` and `memory_mb`, this will _NOT_ compatible with the resource setting which using the maner of slurm. (TODO), see if can find commond setup option for the resources that fit for all type of scheduler. Xing: * [AiiDAlab PSI deployment (staging)](https://dev-aiidalab.psi.ch/hub/login?next=%2Fhub%2F) * `https` works now * Working on AzureAD * Docker image for Demo server. The image size reduce to 4.36 GB (only ~ 400MB larger than orginal image). Use `aiidalab-launch` to test: * ~12 s for the first time * ~5 s for the second time Miki * solved issue installation of vibroscopy plugin on ARM: https://github.com/aiidalab/aiidalab-qe/issues/745