Clarify the currently running Pangeo platforms configuration on EOSC

[guillaumeeb]

I know we have been discussing it by email, but it is still not perfectly clear for me. @j34ni, could you clarify the configuration of Pangeo platforms currently running on EOSC?

For example, I thought with #3 (comment) that dask-gateway was not working on pangeo-foss4g, but from pangeo-data/foss4g-2022#47 (comment) it looks it does?

So for each platform:

• Which helm charts are deployed? Working?
• For daskhub, is dask-gateway working? Dask dashboard? EGI check-in?
• what are the resources behind those platforms?
• what is the IM Dashboard name, the DNS name?
• etc.

Bonus: are there Open Stack resources not used by these two platforms?

Thanks!

[sebastian-luna-valero]

Hi,

Sorry this has been a moving target in the past, and it may also be the case in the future. We try to allocate resources/deployments to meet the demands of upcoming events/projects, also depending on availability.

We have currently settled with https://pangeo-eosc.vm.fedcloud.eu/ where

• the (modified) helm chart is here
• dask-gateway is working with dask dashboard and EGI Check-in [1]
• this deployment is done in a baremetal k8s cluster. Current quotas are: 200 vCPUs, 2TB RAM memory, 20 TB storage. Again, this may change in the future depending on project funding, that's why I am a bit reluctant to add this to the docs and the forget about updating it.
• IM Dashboard is available at https://im.egi.eu/. We still have access to this but right now we are focusing on the baremetal k8s cluster.

I hope that clarifies a bit. Should we close this issue or how would you like to proceed?

Best regards,
Sebastian

[1] EGI Check-in (aai-dev.egi.eu) will be down from today Fri 14th Jul until Mon 16th Jul for an scheduled maintenance and therefore https://pangeo-eosc.vm.fedcloud.eu/ won't be accessible during that period.