> On Oct 21, 2016, at 10:35, Tom Whyntie <[log in to unmask]> wrote:
>
> Hello all,
>
> Following the successful addition of the MoEDAL experiment to the grid (thanks again everyone!), one of my MoEDAL colleagues Daniel Felea (Institute of Space Science, Bucharest, CCed) was wondering about the best way of setting up a new cluster at the ISS that would be: 1) capable of running LHCb/MoEDAL simulations (i.e. with CVMFS, AFS, interfacing with DIRAC, etc.) and 2) potentially compatible with joining the WLCG at a later date.
The Vac-in-a-Box way of running Vac is designed for exactly this kind of scenario, since it is entirely self contained (doesn't even require DNS from the site.)
If you've got MoEDAL jobs running on DIRAC then the existing GridPP DIRAC VMs will run jobs like any other DIRAC site. The LHCb DIRAC VMs are available too of course for LHCb work, and are the basis for the future "official" DIRAC VMs.
Suitable ATLAS VMs are already in production, and CMS will be too again once their updated way of operating them is ready. We're currently discussing with ALICE how to do ALICE VMs. Future WLCG members (Belle II?) which are DIRAC based will run straightforwardly too.
Cheers
Andrew
> Given recent discussions on "grid lite" sites, are there any good sources of information or guidelines about what hardware/software would be good to go for at this point? Any talks at CHEP that might be worth a read, or recent experiences people have had?
>
> Any advice or thoughts much appreciated, as ever (and Happy Friday!).
>
> Cheers, Tom
>
|