---

Job scheduling often require that both a Computing Service (a.k.a. Computing Element) to run the job and a Storage Service to provide for an output storage extent or for an input file. Currently, the selection of a proper pair is done on the base of static relationships betweem these services set by SiteAdmins. The CE-SE bind schema aims at providing the mean for advertising such kind of relationship with eventual per-pair data. At the moment, this is limited to the mount point that a job running in a Computing Service should use to locally access the Storage Space provided by the Storage Service.

For each Computing Service, there is a two-level binding advertisement. At the first level (Group Level), a list of the SE unique ID is provided. At the Single Level (for each SE), specific CE-SE attributes can be provided. At the moment, at the single level, only the mount point for a CE is published. This in case that the SE is locally accessible by jobs running on that CE.

In the near future, when the Glue Network Element will be in place, the selection would be done on a dynamic base using a Network Cost function and proper quality of service attributes.