[glue-wg] GLUE2.1 draft: request for final discussion and approval

[Sill, Alan]

Dear GLUE working group members,

For long-term sustainability, we are looking to host most of the OGF document preparation process in settings we can host in the cloud and in the process modernize some of the tools OGF uses for document preparation and production.

I'd like to see if the folks in this working group would be willing to experiment with producing documents in GitHub using GitHub-Flavored Markdown (GFM) and using pull requests to track and review proposed changes. For this purpose I have created a GitHub-hosed copy of the current document draft and added it to the OGF-GLUE repository in the OpenGridForum organizational pages. Anyone can request to join, and we can designate certain people as authors with write privileges and others as administrators for the repository.

For now this is an experiment, but I would be happy to walk the group through the procedures to propose and make document changes using git rather than exchanging documents by email or editing them directly in Google Docs. This can produce a more coherent way of discussing potential changes and adopting or rejecting them by the usual pull request process familiar to anyone who has used this method to prepare and maintain common source code.

The OGF organizational pages are at https://github.com/OpenGridForum and the repository for the GLUE experiment is at https://github.com/OpenGridForum/OGF-GLUE . We'll be happy to add anyone in the team and to appoint people to administrator or author (write- and pull-enabled) status as required and agreed by the group.

Let me know if you have any questions.

Alan

On Jun 4, 2018, at 4:57 AM, Stephen Burke - UKRI STFC <stephen.burke at stfc.ac.uk<mailto:stephen.burke at stfc.ac.uk>> wrote:

Paolo Andreetto <paolo.andreetto at pd.infn.it<mailto:paolo.andreetto at pd.infn.it>> said:
Even better, I think we can remove the suffix "Info" from both the class names.

That's probably OK, as long as all the names are still different.

It's not easy to design a model that can be suitable for any kind of future accelerator devices.
We can have multiple (different) cards on the same WN, and we have some examples about that.

OK, if you think it's possible then you should allow for that - you shouldn't make the model more complicated than it needs to be but it's reasonable to allow for likely future developments. Also you have quite a few accelerator objects, the arguments aren't necessarily the same for all of them.

I read about new architecture with multi-chip GPU on the same card.

I think the objects usually only need to be one instance per GPU type, if you have multiple GPUs of the same type you could just publish a count in a single object.

There're GPU appliances (Nvidia Quadro VCA) that can be shared among different WNs, and I'm not sure that cloning the AcceleratorEnvironment could get the point.

You'd need to have some idea of what the sharing model would be - in general you'd need a new kind of resource object for that, but it may be that you could deal with it within the existing structure.

Stephen

_______________________________________________
glue-wg mailing list
glue-wg at ogf.org<mailto:glue-wg at ogf.org>
https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.ogf.org%2Fmailman%2Flistinfo%2Fglue-wg&data=02%7C01%7Calan.sill%40ttu.edu%7C1c7e47606981456feb4c08d5ca022cb7%7C178a51bf8b2049ffb65556245d5c173c%7C0%7C0%7C636637033054435909&sdata=jAOIUTesKlfHpR1L5kl30HQfHOx8S5Mtl0itw0l7axM%3D&reserved=0

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.ogf.org/pipermail/glue-wg/attachments/20180605/74b00dc7/attachment-0001.html>