Re: [Condor-devel] RFC: partitionable slots as "per-machine concurrency limits"

[Erik Erlandson <eje@xxxxxxxxxx>]

> 
> A better question is: "what are the possible benefits versus drawbacks
> of extracting mem/disk accounting out of slots and into some other kind
> of per-node accounting mechanism"

After kicking this around with Matt I don't think moving the accounting
complexity around would buy any useful improvement in code
maintainability.  It might even undermine the core classad job/slot
matching.