Hi Stephen,
I do not know the official recommendations, but getting cpu (and managing data) through DIRAC is fairly straightforward. It is easy to submit jobs and I can show you how if it is not already shown somewhere.
About CVMFS, doesn't the RAL Tier-1 already run a stratum-0 for this purpose?
Cheers,
Raja.
On 01/11/19 13:29, Stephen Jones wrote:
> Hi all,
>
> In the storage mail list, I was asking about current protocols for supporting a smaller experiment wrt DPM/DOME etc.
>
> With respect to CPU, I'm thinking of putting them, for now, in the gridpp "umbrella" VO, and giving them a DIRAC client.
>
> And perhaps using Ganga.
>
> I'm not well up with the current best practice. Anybody got any tips on the best way to give these guys some CPU?
>
> Also, on a related note, I'm thinking of using the CVMFS system to distribute their software, so I need to do some research on how to access this as a user in an experiment that, as yet, has not formal VO.
>
> (I'm wondering if gridpp has a place on cvmfs where these people can put their files... just an idea.)
>
> Anyway, if anyone has any tips, I'm trying to form a view on this so I can recommend a direction. Let me know if you have an view, one way or the other.
>
> BTW: I'm also doing up Tom Whyntie's user guide at the moment, and this feedback will help form a basis for all/any new small experiments that come along ... so it's quite important.
>
> Cheers,
>
> Ste
>
>
>
>
########################################################################
To unsubscribe from the TB-SUPPORT list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=TB-SUPPORT&A=1
|