# AiiDA Team Meeting 2021-06-01 ###### tags: `team meetings` ###### time: 10am CET [TOC] ### Present * Simon * Marnik * Leopold * Sebastiaan * Francisco * Chris * Giovanni ### Catch-up round *Max. 3 minutes each* Giovanni: * Have next meetings in person by default? Identify people in charge of CoViz2 booking and setting up Zoom for any remote participant (Chris + Jason as backup?) * Milestones (see below in specific agenda items) * Next events: 28 June (new internal design of AiiDA core, for core developers) + 30 June: AiiDA webinar for users. Is there enough non-overlap with what we already did * Final discussion: the developer meeting is canceled, one can ask directly Chris. For the AiiDA webinar, a more general discussion of what can be done with AiiDA (e.g. AiiDA QE workflows) will happen together with the tutorial. Chris will make a very short video on how to "start" with AiiDA 2.0, put on YouTube on the AiiDA channel, and link it from the documentation (together with the recording of the webinar for plugin developers of a few weeks ago - NOTE! recordings get deleted after 30 days, remember to download it!) Sebastiaan: * **Open PRs**: * Deprecate `Profile.repository_path` [[#5516]](https://github.com/aiidateam/aiida-core/pull/5516) * `DynamicEntryPointCommandGroup`: make actual command dynamic [[#5548]](https://github.com/aiidateam/aiida-core/pull/5548) * ORM: raise when trying to pickle instance of `Entity` [[#5549]](https://github.com/aiidateam/aiida-core/pull/5549) * Docs: add `-U` flag to PostgreSQL database backup command [[#5550]](https://github.com/aiidateam/aiida-core/pull/5550) * Docs: correct snippet for workchain context nested keys [[#5551]](https://github.com/aiidateam/aiida-core/pull/5551) * **Merged PRs** * CLI: normalize use of colors [[#5547]](https://github.com/aiidateam/aiida-core/pull/5547) * `BaseRestartWorkChain`: allow to override priority in `handler_overrides` [[#5546]](https://github.com/aiidateam/aiida-core/pull/5546) * CLI: allow setting options for config without profiles [[#5544]](https://github.com/aiidateam/aiida-core/pull/5544) * `Process`: refactor out input handling specific to `CalcJob` [[#5539]](https://github.com/aiidateam/aiida-core/pull/5539) * Engine: fix bug when caching from process with nested outputs [[#5538]](https://github.com/aiidateam/aiida-core/pull/5538) * Engine: fix bug that allowed non-storable inputs to be passed to process [[#5532]](https://github.com/aiidateam/aiida-core/pull/5532) Leopold: * AiiDA was no longer raising when detecting multiple entry points of the same name in the same group [fixed](https://github.com/aiidateam/aiida-core/issues/5530) * [fsspec](https://github.com/fsspec/filesystem_spec) provides a uniform Pyton API to a [wide range of local and cloud file systems](https://filesystem-spec.readthedocs.io/en/latest/api.html?highlight=azure#built-in-implementations) ([background](https://filesystem-spec.readthedocs.io/en/latest/intro.html), [features](https://filesystem-spec.readthedocs.io/en/latest/features.html)). In particular, [s3fs, gcfs, abfs, ... already depend on fsspec](https://filesystem-spec.readthedocs.io/en/latest/api.html?highlight=azure#other-known-implementations). I'm not 100% sure but it seems to me that having a storage adapter in AiiDA that restricts itself to `fsspec` operations in principle provides interoperability with all of these implementations (+ the built-in ones in fsspec) in one shot (in pratice one may want to do fs-dependent optimizations) Marnik: * Did a short AiiDA tutorial and AiiDAlab demo/trial at the University of Antwerp. * On holiday last week, so not much to report. * Two weeks holiday the next weeks, then week of conference in Italy. * [GP] Can we fix a date for a first brainstorming on the tutorial in Oct? Francisco: * [PR-5552](https://github.com/aiidateam/aiida-core/pull/5552): Add autocomplete functionality to the `verdi group list` command (add tests). Simon: - _Mainly worked on AiiDAlab and MarketPlace issues._ - Created AiiDAlab deployment for AiiDA tutorial @ University of Antwerpen (https://uatutorial.aiidalab.io) - Deployment was reportedly fast, smooth, and robust. - Deployed in eu-west with Standard_D2as_v5 for the core node pool and Standard_D32as_v5 for the user node pool. - Encountered some issues with configuring the DNS zone for aiidalab.io; the domain was protected which required intervention from Giovanni. - There were only 12 participants so there were unfortunately no learnings from autoscaling as the chosen node size (32 CPU, 128 GB) was oversized. - Costs: ~ 50 USD / day - Cluster will be destroyed today. - In progress of finalizing the aiidalab-on-azure template. ### Specific Agenda Items - Add here 1 or 2 main issues that you want to work on related to AiiDA (in the next ~3 months, focus on high priority ones) and report in the next meeting, so we prepare a roadmap and know who will work on them. Examples: critical bugs (recursion errors), sharing capabilities, containerized codes, REST API changes/improvements, ... - Check topics here: https://github.com/aiidateam/aiida-core/wiki/AiiDA-release-roadmap, don't pick more than 2 (3 max), be realistic - Add your name and what you want to work on: - Marnik: [Task Farming](https://github.com/aiidateam/aiida-core/wiki/AiiDA-release-roadmap#task-farming-disguise-multiple-jobs-as-one) -> Updating/improving `aiida-hyperqueue` - Simon: [Workflow management through the REST API](https://github.com/aiidateam/aiida-core/wiki/AiiDA-release-roadmap#workflow-management-through-the-rest-api) - Finalize choice of date of tutorial, write on aiida.net/events (people start asking), define "organizing committee" (Marnik + Chris), have timeline