Conversation
|
I had two general comments: a) is there perhaps someplace upstream (hsf?) that could host the non-fcc specific pieces of this? b) perhaps docs on dirac, rucio, grid certs (perhaps better tokens instead) should wait until they are relevant for most (or at least many) users? [eg, rucio is much more advanced in fcc than this doc suggests] |
|
@davidlange6 Thanks for the review! a) Right now, no place comes to my mind. Some time ago we considered the posibility of writting HSF modules related to grid computing, but we gave up since it not trivial to make generic enough material for examples and exercises (PanDA vs DIRAC vs glideinWMS, flat vs hierarchical namespaces, VO policies, etc). And we assumed each experiment has its own tailored version. b) Yes you are right, I was lazy not digging deeper on the status of Rucio adoption. I can update that part. And about waiting that it becomes relevant for most users, the logic is that we already have quite some material about grid computing and we want to put some context for those not familiar with grid. I see no harm on adding it right now. |
| For example, CERN runs [HTCondor](https://batchdocs.web.cern.ch/) on its farm, reachable from `lxplus`. Many universities run HTCondor or Slurm. In batch system | ||
| you already have an account, the shared filesystem (AFS, EOS) is mounted on the worker nodes, and you can `ssh` in to look at things. | ||
|
|
||
| The grid glues many batch systems together behind a single interface. In exchange for scale, there is no shared filesystem across sites, |
There was a problem hiding this comment.
There is CVMFS on all the sites as the one shared filesystem.
There was a problem hiding this comment.
Good point. I meant filesystem for users to write, but certainly better to be more accurate about it.
This branch has not been deployed
3 participants
Adding a chapter on basic concepts of distributed computing in HEP, intended for beginners.
Resolves #193 .